JP3750396B2 - Image processing device - Google Patents

Description

表１において、状態欄の図形は、図３のステップＳ１０１で作成した２値のエッジリストのうち、注目画素に対応する部分の状態を模式的に示したものであり、白色矩形は塗り潰さない画素、黒色画素は塗り潰す画素を表す。なお、この算出方法については後述する。
【００３３】
次に、変数ｎ２が「３」に等しいどうかを調べ（ステップＳ３０５）、等しければステップＳ３０６以降の処理を行い、等しくなければステップＳ３０９以降の処理を行う。変数ｎ２が「３」に等しい場合、変数ｎ３が「０」または「１２」に等しいかどうかを調べ（ステップＳ３０６）、どちらかに等しい場合には変数ｎ２を表３に示すＬＵＴで濃度値に変換して変数ｍに格納する（ステップＳ３０７）。一方、どちらにも等しくない場合には変数ｎ２を表２に示すＬＵＴで濃度値に変換して変数ｍに格納する（ステップＳ３０８）。
【表２】

【００３４】
【表３】

【００３５】
一方、変数ｎ２が「３」に等しくない場合、変数ｎ２が「１２」に等しいかどうかを調べ（ステップＳ３０９）、等しければステップＳ３１０以降の処理を行い、等しくなければステップＳ３１２以降の処理を行う。変数ｎ２が「１２」に等しい場合、変数ｎ１が「０」または「３」に等しいかどうかを調べ（ステップＳ３１０）、等しければ表３に示すＬＵＴで濃度値に変換して変数ｍに格納し（ステップＳ３１１）、等しくなければ表２に示すＬＵＴで濃度値に変換して変数皿に格納する（ステップＳ３１２）。
【００３６】
上記ステップＳ３０７、ステップＳ３０８、ステップＳ３１１、ステップＳ３１２のいずれかでの処理が終了すると、濃度値が格納されている変数ｍと変数ｐｖが「０」より大きいかどうかを調べる（ステップＳ３１０）。そして、変数ｍと変数ｐｖがともに「０」より大きければ、変数ｍと変数ｐｖが等しいかどうかを調べ（ステップＳ３１２）、等しければ変数ｒｌをインクリメントし（ステップＳ３１３）、等しくなければ変数ｓｔ、変数ｐｖおよび変数ｒｌをリストに登録し（ステップＳ３１４）、変数ｓｔにインデックスｉ、変数ｐｖに変数ｍ、変数ｒｌに「１」を設定する（ステップＳ３１５）。
【００３７】
一方、変数ｍが「０」であり、変数ｐｖが「０」より大きければ、変数ｓｔ、変数ｐｖおよび変数ｒｌをリストに登録し（ステップＳ３１７）、変数ｐＶに「０」を設定する（ステップＳ３１８）。また、変数ｍが「０」より大きく、変数ｐｖが「０」であれば、変数ｓｔにインデックスｉ、変数ｐｖに変数ｍ、変数ｒｌに「１」を設定する（ステップＳ３１６）。つまり、ステップＳ３１０からステップＳ３１９における処理は、ランレングスコーディング処理である。
【００３８】
次に、ステップＳ３１３、ステップＳ３１５、ステップＳ３１６またはステップＳ３１８での処理が終了するか、ステップＳ３１０で変数ｍと変数ｐｖがともに「０」であった場合には、インデックスｉをインクリメントし（ステップＳ３１９）、インデックスｉがＰに等しいかどうかを調べる（ステップＳ３２０）。そして、等しければステップＳ３２１以降の処理を行い、等しくなければステップＳ３０４以降の処理を行う。
【００３９】
インデックスｉがＰに等しい場合、変数ｐｖが「０」より大きいかどうかを調べ（ステップＳ３２１）、大きければ変数ｓｔ、変数ｐｖおよび変数ｒｌをリストに登録する（ステップＳ３２２）。そして、ステップＳ３２２での処理が終了するか、ステップＳ３２１での判定結果が「ＮＯ」であった場合、ＥＯＬをリストに登録してインデックスｊをインクリメントし（ステップＳ３２３）、インデックスｊがＬから「１」を減じた値に等しいかどうかを調べる（ステップＳ３２４）。そして、等しければ後述する最終ライン処理を行って（ステップＳ３２５）、全ての処理を終了し、等しくなければステップＳ３０３以降の処理を行う。
【００４０】
ここで、図５のステップＳ３０１における先頭ライン処理について説明する。図８および図９は、先頭ライン処理の詳細を示すフローチャートである。なお、図６において、図５ないし図７と同じ名前のインデックス、変数、記号および関数は、図５ないし図７と同様である。
【００４１】
まず、ＳＴ３，ＲＬ３〜ＳＴ６，ＲＬ６を読み込むとともに、処理に必要な変数を初期化する（ステップＳ４００）。次に、画素の状態を表す変数ｎ２、ｎ３を算出する（ステップＳ４０１）。次に、変数ｎ２が「３」に等しいどうかを調べ（ステップＳ４０２）、等しければステップＳ４０３以降の処理を行い、等しくなければステップＳ４０５以降の処理を行う。変数ｎ２が「３」に等しい場合には、変数ｎ３が「０」または「１２」に等しいかどうかを調べ（ステップＳ４０３）、どちらかに等しい場合には、変数ｎ２を表３に示すＬＵＴで濃度値に変換して変数ｍに格納する（ステップＳ４０４）。一方、どちらにも等しくない場合には、変数ｎ２を表２に示すＬＵＴで濃度値に変換して変数ｍに格納する（ステップＳ４０５）。
【００４２】
上記ステップＳ４０４、ステップＳ４０５のいずれかでの処理が終了すると、濃度値が格納されている変数ｍと変数ｐｖが「０」より大きいかどうかを調べ（ステップＳ４０６）、変数ｍと変数ｐｖとがともに「０」より大きければ、変数ｍと変数ｐｖとが等しいかどうかを調べる（ステップＳ４０７）。そして、等しければ、変数ｒｌをインクリメントし（ステップＳ４０８）、等しくなければ、変数ｓｔ、変数ｐｖおよび変数ｒｌをリストに登録し（ステップＳ４０９）、変数ｓｔにインデックスｉ、変数ｐｖに変数ｍ、変数ｒｌに「１」を設定する（ステップＳ４１０）。一方、変数ｍが「０」であり、変数ｐｖが「０」より大きければ、変数ｓｔ、変数ｐｖおよび変数ｒｌをリストに登録し（ステップＳ４１２）、変数ｐｖに「０」を設定する（ステップＳ４１３）。また、変数ｍが「０」より大きく、変数ｐｖが「０」であれば、変数ｓｔにインデックスｉ、変数ｐｖに変数ｍ、変数ｒｌに「１」を設定する（ステップＳ４１１）。つまり、ステップＳ４０６からステップＳ４１３における処理は、先頭ラインのランレングスコーディング処理である。
【００４３】
次に、ステップＳ４０８、ステップＳ４１０、ステップＳ４１１またはステップＳ４１３での処理が終了するか、ステップＳ４０６で変数ｎと変数ｐｖがともに「０」であった場合には、インデックスｉをインクリメントし（ステップＳ４１４）、インデックスｉがＰに等しいかどうかを調べ（ステップＳ４１５）、等しければステップＳ４１６以降の処理を行い、等しくなければステップＳ４０１以降の処理を行う。インデックスｉがＰに等しい場合、変数ｐｖが「０」より大きいかどうかを調べ（ステップＳ４１６）、大きければ変数ｓｔ、変数ｐｖおよび変数ｒｌをリストに登録する（ステップＳ４１７）。そして、ステップＳ４１７での処理が終了するか、子テップＳ４１６での判定結果が「ＮＯ」であった場合、ＥＯＬをリストに登録して（ステップＳ４１８）、処理を終了する。
【００４４】
次に、図７のステップＳ３２５における最終ライン処理について説明する。図１０および図１１は、最終ライン処理の詳細を説明するためのフローチャートである。なお、図１０および図１１において、図５ないし図７と同じ名前のインデックス、変数、記号および関数は、図５ないし図７と同様である。
【００４５】
まず、ＳＴ１，ＲＬ１〜ＳＴ４，ＲＬ４を読み込むとともに、処理に必要な変数を初期化する（ステップＳ５００）。次に、画素の状態を表す変数ｎ１〜ｎ２を算出する（ステップＳ５０１）。次に、変数ｎ１が「１２」に等しいどうかを調べ（ステップＳ５０２）、等しければステップＳ５０３以降の処理を行い、等しくなければステップＳ５０５以降の処理を行う。変数ｎ２が「１２」に等しい場合には、変数ｎ１が「０」または「３」に等しいかどうかを調べ（ステップＳ５０３）、どちらかに等しい場合には、変数ｎ２を表３に示すＬＵＴで濃度値に変換して変数ｍに格納する（ステップＳ５０４）。一方、どちらにも等しくない場合には、変数ｎ２を表２に示すＬＵＴで濃度値に変換して変数ｍに格納する（ステップＳ５０５）。
【００４６】
上記ステップＳ５０４、ステップＳ５０５のいずれかでの処理が終了すると、濃度値が格納されている変数ｍと変数ｐｖが「０」より大きいかどうかを調べ（ステップＳ５０６）、変数ｍと変数ｐｖがともに「０」より大きければ変数ｍと変数ｐｖとが等しいかどうかを調べ（ステップＳ５０７）、等しければ変数ｒｌをインクリメントする（ステップＳ５０８）。一方、等しくなければ変数ｓｔ、変数ｐｖおよび変数ｒｌをリストに登録し（ステップＳ５０９）、変数ｓｔにインデックスｉ、変数ｐｖに変数ｍ、変数ｒｌに「１」を設定する（ステップＳ５１０）。一方、変数ｍが「０」であり、変数ｐｖが「０」より大きければ、変数ｓｔ、変数ｐｖおよび変数ｒｌをリストに登録し（ステップＳ５１３）、変数ｐｖに‘０’を設定する（ステップＳ５１４）。また、変数ｍが「０」より大きく、変数ｐｖが「０」であれば、変数ｓｔにインデックスｉ、変数ｐｖに変数ｍ、変数ｒｌに「１」を設定する（ステップＳ５１２）。つまり、ステップＳ５０６からステップＳ５１４における処理は、最終ラインのランレングスコーディング処理である。
【００４７】
次に、ステップＳ５０８、ステップＳ５１０、ステップＳ５１２またはステップＳ５１４での処理が終了するか、ステップＳ５０６で変数ｍと変数ｐｖとがともに「０」であった場合には、インデックスｉをインクリメントし（ステップＳ５１４）、インデックスｉがＰに等しいかどうかを調べる（ステップＳ５１５）。そして、双方が等しければステップＳ５１７以降の処理を、等しくなければステップＳ５０１以降の処理を行う。インデックスｉがＰに等しい場合、変数ｐＶが「０」より大さいかどうかを調べ（ステップＳ５１７）、大きければ変数ｓｔ、変数ｐｖおよび変数ｒｌをリストに登録する（ステップＳ５１８）。ステップＳ５１８での処理が終了するか、ステップＳ５１７での判定結果が「ＮＯ」であった場合には、ＥＯＬをリストに登録して（ステップＳ５１９）、処理を終了する。
【００４８】
次に、図５のステップＳ３０４、図８のステップＳ４０１および図１０のステップＳ５０１における変数ｎ１、ｎ３の算出方法について説明する。なお、この算出方法は、変数ｎ１〜ｎ３について同じであるため、変数ｎ１の場合についてのみ説明する。図１２および図１３は、変数ｎ１を算出するアルゴリズムを説明するためのフローチャートである。なお、図１２および図１８において、図ないし図７と同じ名前のインデックス、変数および記号は、図５ないし図７と同様であり、記号「｜」はビットＯＲ演算を示すものとする。
【００４９】
まず、変数ｎ１を初期化し（ステップＳ６００）、注目画素の主走査方向での位置を示すインデックスｉが変数ＳＴ１以上、かつ変数ＳＴ１と変数ＲＬ１との和より小さいか否かを調べる（ステップＳ６０１）。つまり、注目画素が塗潰領域にあるかどうかを調べる。ステップＳ６０１での判定結果が「ＹＥＳ」の場合、すなわち注目画素が塗潰領域にある場合には、変数ｎ１と定数「１０００ｂ」とのビットＯＲ演算結果を変数ｎ１に格納する（ステップＳ６０２）。一方、「ＮＯ」の場合、すなわち注目画素が塗潰領域にない場合には、インデックスｉが変数ＳＴ１と変数ＲＬ１の和以上であるか否かを調べる（ステップＳ６０３）。ステップＳ６０３での判定結果が「ＹＥＳ」の場合には、リストからＳＴ１およびＲＬ１を読み込む（ステップＳ６０４）。
【００５０】
次に、ステップＳ６０２またはステップＳ６０４での処理が終了するか、ステップＳ６０３での判定結果が「ＮＯ」である場合には、インデックスｉに「１」を加えた値が変数ＳＴ１以上、かつ変数ＳＴ１と変数ＲＬ１との和より小さいか否かを調べる（ステップＳ６０５）。つまり、注目画素の右の画素（以下、単に右画素という）が塗潰領域にあるか否かを調べる。ステップＳ６０５での判定結果が「ＹＥＳ」の場合、すなわち右画素が塗潰領域にある場合には、変数ｎ１と定数「１００ｂ」とのビットＯＲ演算結果を変数ｎ１に格納する（ステップＳ６０６）。一方、「ＮＯ」の場合、すなわち右画素が塗潰領域にない場合には、インデックスｉに「１」を加えた値が変数ＳＴ１と変数ＲＬ１との和以上であるか否かを調べる（ステップＳ６０７）。そして、ステップＳ６０７での判定結果が「ＹＥＳ」の場合には、リストからＳＴ１およびＲＬ１を読み込む（ステップＳ６０８）。
【００５１】
次に、ステップＳ６０６またはステップＳ６０８での処理が終了するか、ステップＳ６０７での判定結果が「ＮＯ」である場合には、インデックスｉが変数ＳＴ２以上で、かつ変数ＳＴ２と変数ＲＬ２との和より小さいか否かを調べる（ステップＳ６０９）。つまり、注目画素の下の画素（以下、単に下画素という）が塗潰領域にあるかどうかを調べる。ステップＳ６０９での判定結果が「ＹＥＳ」の場合、すなわち下画素が塗潰領域にある場合には、変数ｎ１と定数「１０ｂ」とのビットＯＲ演算結果を変数ｎ１に格納する（ステップＳ６１０）。一方、「ＮＯ」の場合、すなわち下画素が塗潰領域にない場合には、インデックスｉが変数ＳＴ２と変数ＲＬ２との和以上であるか否かを調べる（ステップＳ６１１）。そして、ステップＳ６１１での判定結果が「ＹＥＳ」の場合には、リストからＳＴ２およびＲＬ２を読み込む（ステップＳ６１２）。
【００５２】
次に、ステップＳ６１０またはステップＳ６１２での処理が終了するか、ステップＳ６１１での判定結果が「ＮＯ」である場合には、インデックスｉに「１」を加えた値が変数ＳＴ２以上で、かつ変数ＳＴ２と変数ＲＬ２との和より小さいか否かを調べる（ステップＳ６１３）。つまり、注目画素の右下の画素（以下、単に右下画素という）が塗潰領域にあるかどうかを調べる。ステップＳ６１３での判定結果が「ＹＥＳ」の場合、すなわち右下画素が塗潰領域にある場合には、変数ｎ１と定数「１ｂ」とのビットＯＲ演算結果を変数ｎ１に格納し（ステップＳ６１４）、当該処理を終了する。一方、「ＮＯ」の場合、すなわち次画素が塗潰領域にない場合には、インデックスｉに「１」を加えた値が変数ＳＴ２と変数ＲＬ２との和以上であるか否かを調べる（ステップＳ６１５）。ステップＳ６１５での判定結果が「ＹＥＳ」の場合には、リストからＳＴ２およびＲＬ２を読み込み（ステップＳ６１６）、当該処理を終了する。また、ステップＳ６１５での判定結果が「ＮＯ」である場合には、何もせずに当該処理を終了する。なお、上記処理において、画素の相対的な位置関係は、ラインの先頭画素を最左端、先頭ラインを最上端ラインとして記述している。
【００５３】
ここで、明朝体の文字「一」を例にとり、処理の具体例を示す。図１４は、処理過程でのラスタライズ画像ならびに比較対照のための概念図である。また、図１５は、図１４（ａ），（ｅ）の部分拡大図である。なお、図１４および図１５において、矩形は画素を表し、白色矩形以外の矩形の意味についてはそれぞれ右横の注釈の通りである。また、図１４においては、１つの文字を主走査方向１６画素、副走査方向１６画素の領域に展開するものとし、該領域の左上端を座標原点とし、上述した画素の相対的な位置関係と同様の位置関係で示している。また、図１５においては、破線矢印は、プリンタ解像度のｎ倍の解像度の２値の画像データを表すエッジリストを作成する際に仮想的に設定されるスキャンラインであり、主走査および副走査の両方向のスキャンラインの交点が図４における仮想座標点ＴＰ（ｉ，ｊ）である。なお、図１５では、上述と同様にｎ＝２の場合について示しているが、ｎの値が他の値であっても同様に仮想座標点を設定できるのは明らかである。
【００５４】
図１４（ａ）は、文字「一」を理想的に展開した場合の塗潰領域を示す概念図であるが、固有の解像度を有するプリンタに出力するためには、塗潰領域を含む画素の値を何らかの論理に従って決定する必要がある。例えば、図１４（ａ）に示す塗潰領域を少しでも含む画素を印字、含まない画素を非印字と決定すると、図１４（ｂ）に示すような印字結果が得られる。両図を見てもわかるように、この画素値決定論理では、高い画質の印字結果を得ることはできない。一方、前述した従来技術に記載されているように、ＰＤＬから作成したプリンタ解像度のｎ倍の解像度の２値の画像データを表すエッジリストから単純にプリンタ解像度の多値の画像データを表すエッジリストに変換してラスタライズすると、図１４（ｃ）に示すラスター画像が得られ、このラスター画像を印字すると図１４（ｄ）に示す印字結果が得られる。この場合、特に文字「一」の横棒部分が想定していた線幅よりも細い線幅で印字されてしまい、場合によっては途切れたり印字されないことがある。
【００５５】
一方、上述した本発明の実施形態によってラスタライズすると、図１４（ｅ）に示すラスター画像が得られ、このラスター画像を印字すると図１４（ｆ）に示す印字結果が得られる。図１５（ａ）および図１５（ｂ）は、それぞれ図１４（ａ）および図１４（ｅ）の部分拡大図である。以下、本発明によるラスタライズ処理例を図１４（ａ）、図１４（ｅ）、図１４（ｆ）および図１５を参照しながら説明する。まず、本実施形態において、入力されたＰＤＬのうち、多値展開を指定された文字についてＰＤＬからアウトラインデータ、さらにエッジリストへと変換が行われる。図４ないし図７に示したアルゴリズムに従って、各仮想座標点が塗潰領域にあるか否かが判定され、その結果エッジリストが作成されていく。
【００５６】
例えば、図１５（ａ）の７ライン目について見てみると、上側の主走査方向のスキャンライン（以下、単に主走査スキャンラインという）では、１３番目の画素の右側の仮想座標点が塗潰領域内にある。これは、図４のステップＳ２０２において、ｉ＝２７，ｊ＝１４のときに判定結果が「ＹＥＳ」となることを意味している。この仮想座標点までに、ステップＳ２０２における判定結果が「ＹＥＳ」となることはなく、また、この仮想座標点以降の同じ主走査スキャンライン上の仮想座標点についてもステップＳ２０２における判定結果が「ＮＯ」となるため、この主走査スキャンラインが終了した時点でのリストには、ｓｔ＝２７，ｒｌ＝１のみが登録されている。
【００５７】
一方、下側の主走査スキャンラインでは、１３番目の画素の左右両側および１４番目の画素の左側の仮想座標点が連続して塗潰領域内にある。従って、上側と同様に処理を行うと、この主走査スキャンラインが終了した時点でのリストには、ｓｔ＝２６，ｒｌ＝３がリストに追加されている。同様に８ライン目についても処理を行うと、上側の主走査スキャンラインでは、ｓｔ＝２，ｒｌ＝２６がリストに追加され、下側の主走査スキャンラインでは、何もリストに追加されない。上述したライン以外については、いずれの仮想座標点も塗演領域にはないため、最終的には表４に示すようなリストが得られる。
【表４】

【００５８】
表４に示すリストは、一形態例であってリストを見やすい形で表現したものにすぎず、リストの記憶媒体でのデータ形式は、表４に隈定されるわけではなく、どの主走査スキャンラインにどのようなｓｔ，ｒｌが登録されているかが分かればどのような形態であっても構わない。なお、表４においては、ＥＯＬ符号のみが登録されている主走査スキャンラインおよびＥＯＬ符号については省略している。
【００５９】
次に、表４に示すリストに基づいて多値のエッジリストが作成される。この作成の具体例を図１５（ｂ）の７ライン目および８ライン目を例にとり説明する。まず、７ライン目では、１番目の画素から１２番目の画素までは、図１２または図１３におけるｎ１、ｎ２の算出でステップＳ６０１、ステップＳ６０５、ステップＳ６０９およびステップＳ６１３での判定結果がいずれも「ＮＯ」となり、ｎ３の算出でステップＳ６０１およびステップＳ６０５での判定結果がいずれも「ＹＥＳ」、ステップＳ６０９およびステップＳ６１３での判定結果がいずれも「ＮＯ」となるので、ｎ１＝０、ｎ２＝０、ｎ３＝１２となり、図５のステップＳ３０５およびステップＳ３０９での判定結果がいずれも「ＮＯ」となる。従って、変数ｎ２が表２に示すＬＵＴで変換され、画素値「０」が得られる。ゆえに、多値エッジリストには何も登録されない。同ラインの１３番目の画素の時点でｎ１＝０、ｎ２＝７、ｎ３＝１２、１４番目の画素の時点でｎ１＝０、ｎ２＝２、ｎ３＝１２となるので、それぞれ表２に示すＬＵＴから画素値「１９２」および画素値「６４」が得られる。
【００６０】
先頭画素および１５番目以降の画素については、いずれもｎ１＝０、ｎ２＝０、ｎ３＝０となるので、７ライン目の処理が終了した時点で、多値エッジリストには、ｓｔ＝１３、ｐｖ＝１９２、ｒｌ＝１およびｓｔ＝１４、ｐｖ＝６４、ｒｌ＝１が登録されている。次に、８ライン目では、１番目の画素から１２番目の画素までは、ｎ１＝０、ｎ２＝１２、ｎ３＝０となり、図５のステップＳ３０５での判定結果は「ＮＯ」、ステップＳ３０９およびステップＳ３１０での判定結果は「ＹＥＳ」となる。従って、変数ｎ２が表３に示すＬＵＴで変換され、画素値「１９２」が得られる。また、同ラインの１３番目の画素の時点で、ｎ１＝７、ｎ２＝１２、ｎ３＝０となり、１４番目の画素の時点で、ｎｌ＝２、ｎ２＝１２、ｎ３＝０となり、ともに画素値「１２８」が得られる。上述したライン以外については、いずれの画素についても、画素値は「０」となるため、最終的には、例えば表５に示すようなリストが得られる。
【表５】

【００６１】
表５に示すリストは、一形態例であってリストを見やすい形で表現したものにすぎず、リストの記憶媒体でのデータ形式は表５に限定されるわけではなく、どのラインにどのようなｓｔ，ｐｖ，ｒｌが登録されているかが分かればどのような形態であっても構わない。なお、表５においては、ＥＯＬ符号のみが登録されている主走査スキャンラインおよびＥＯＬ符号については省略している。以上により、多値エッジリストが得られ、多値文字処理部１２から合成部１８に出力される。
【００６２】
次に、２値文字処理部１３での処理を説明する。まず、描画する文字に対応したアウトラインデータをロードし、２値エッジリストを作成する。ＴＹＰＥコマンドで指定された文字列中の全ての文字の描画が終了すると、合成部１８に２値エッジリストを出力して処理を終了し、描画してない文字がある場合には、アウトラインデータのロード以降を繰り返す。
【００６３】
なお、上述した処理のうちアウトラインデータのロードは、多値文字処理部１２と同じであり、文字輪郭からの２値エッジリストの作成は、多値文字処理部１２における高解像度エッジリストの作成と解像度の設定、つまり仮想座標点の設定が異なるのみで他は同じであるので、詳細な説明は省略する。文字「一」から作成したエッジリストを表６に示す。
【表６】

【００６４】
次に、多値ベクトル処理部１４での処理を説明する。図１６は、多値ベクトル処理部１４での処理を示すフローチャートである。まず、指定されたベクトルを描画し（ステップＳ７００）、高解像度エッジリストを作成し（ステップＳ７０１）、該エッジリストを多値エッジリストに変換する（ステップＳ７０２）。全てのベクトルの描画が終了すると、合成部１８に多値エッジリストを出力して処理を終了し、描画していないベクトルがある場合には、ステップＳ７００以降を繰り返す。ステップＳ７０１以降は、多値文字処理部１２と同じであるので、ここでは、ステップＳ７００についてのみ説明する。まず、指定されたベクトルは、指定された座標点情報に従って仮想座標平面上に描画される。次に、仮想座標平面上に描画されたベクトルを中心として、指定された線幅の半分の距離にある領域を塗潰領域とする。そして、多値文字処理部１２と同様に仮想座標点を設定し、高解像度エッジリストの作成、および該エッジリストの多値エッジリストヘの変換を行っていく。
【００６５】
ここで、図１７は、多値ベクトル処理部１４での処理の具体例を説明するための概念図である。まず、指定されたベクトルが図１７（ａ）に示すハッチング領域の中心線のように描画される。次に、該中心線から指定された線幅の半分の距離にある領域が図１７（ａ）に示すハッチング領域のように塗潰領域に指定される。このとき、多値文字処理部１２においては、アウトラインデータをロードし、塗潰領域を設定したときと同じ状態であるので、上記処理以降では、多値文字処理部１２と同様の処理を行い、一旦、高解像度エッジリストを取得した後、多値エッジリストを取得する。表７は、高解像度エツジリストを示し、表８は、多値エッジリストを示す。また、図１７（ｃ）は、最終的に得られた多値エッジリストを印字した結果を示す図である。
【表７】

【００６６】
【表８】

【００６７】
次に、２値ベクトル処理部１５での処理を説明する。まず、指定された図形を描画し、２値エッジリストを作成する。全ての図形の描画が終了すると、合成部１８に２値エッジリストを出力して処理を終了し、描画していない図形がある場合には、描画以降を繰り返す。なお、上述した処理のうち、図形の描画は、多値ベクトル処理部１４と同じであり、ベクトルからの２値エッジリストの作成は、多値文字処理部１２における高解像度エッジリストの作成と解像度の設定、つまり仮想座標点の設定が異なるのみで、他は同じであるので、詳細な説明は省略する。図１７（ａ）に示す図形から作成したエッジリストを表９に示す。
【表９】

【００６８】
次に、グラフィクス処理部１６での処理を説明する。ここでは、指定された領域を指定された色および指定された塗り潰し方法で塗り潰す処理が行われるが、塗り潰し方法が指定された領域を一様に塗り潰す方法である場合には詳述するまでもないので、一様に塗り潰さない方法について説明する。このような塗り潰し方法では、仮想座標平面上の座標位置の関数として各座標位置の画素値が決定される。この関数としては、例えば矩形領域内で主走査方向に最高濃度から最低濃度まで画素値が単調減少する関数がある。
【００６９】
ここで、図１８は、一様に塗り潰さない方法を説明するための概念図である。図１８（ａ）は、主走査方向１４画素、副走査方向１ラインの仮想座標平面上の領域に、該領域の主走査方向の左端を最高濃度、右端を最低濃度とし、その間は主走査方向の左から右に向かって単調減少する関数により描画した例、図１８（ｂ）は、プリンタ解像度と同じ解像度で仮想座標点を設定して各画素値を決定した例、図１８（ｃ）は、図１８（ｂ）に示す領域をプリントアウトした例である。なお、図１８（ｂ）の上側の数値は、その直下の画素の値を示している。なお、図１８に示す例は、一具体例であり、座標位置から画素値を決定する関数を様々に変えることにより、どのような領域の塗り潰し方法にも対応できるのは明らかである。ここで、座標位置によらず画素値が同じ値に決定される関数を用いれば、指定された領域を一様に塗り潰すことができるので、一様に塗り潰す方法も上述した塗り潰し方法に含まれる。画素値の決定が終了すると、決定した画素偉から多値エッジリストを作成し、合成部１８に出力する。表１０に図１８（ｂ）に示す領域の各画素値から作成したエッジリストを示す。
【表１０】

【００７０】
次に、イメージ処理部１７での処理を説明する。ここでは、ラスターイメージデータに従って指定された領域に、指定された解像度および指定された色空間で描画するのみであるが、指定された解像度がプリンタ解像度と一致しない場合がある。このような場合には、プリンタ解像度と同じ解像度で設定した仮想座標点と、ラスターイメージデータの各画素の位置とが一致しないため、ある座標点の近傍にあるラスターイメージデータの各画素の値から近似画素値を算出することによりプリントアウトを可能にする。
【００７１】
つまり、解像度変換処理によって、ラスターイメージデータをプリンタ解像度に合せることにより、プリントアウトを可能にする。このときの近似画素値の算出方法は、周知の方法でよい。また、指定された色空間がプリントアウト可能な色空間と一致しない、あるいはラスターイメージデータの各画素値が表す色がプリンタで再現できない場合には、ラスターイメージデータの画素値を変更する。このときの画素値の変更方法も周知の方法でよい。上述したように、ラスターイメージデータをプリントアウト可能な解像度および色空間の画像に変換した後、変換した画像から多値エッジリストを作成し、合成部１８に出力する。
【００７２】
最後に、合成部１８での処理を説明する。合成部１８は、多値文字処理部１２、２値文字処理部１３、多値ベクトル処理部１４、２値ベクトル処理部１５、グラフィクス処理部１６およびラスターイメージ処理部１７から出力されたエッジリストを合成し、ラスター画像に変換して、出力端Ａを通じて外部に出力する。なお、この合成処理は、エッジリスト間の操作、例えばリストの連結・挿入・選択であり、優先順位はコマンド分離部１１に入力されたＰＤＬ内での描画順序に従う。例えば、何も指定がない場合には、ＰＤＬの先頭から最後尾に向けて優先順位が高くなる。
【００７３】
つまり、後から描画される文字またはベクトルが先に描画した文字またはベクトルに上書きされることとなる。このとき、後から描画される文字またはベクトルのエッジリストは、既存の出力用エッジリストに強制挿入され、既存の出力用エッジリストの該当部分は書きつぶされる。また、何らかの指定があった場合には、上記描画順序に従わずに指定された描画順序に従う。例えば、ＰＤＬの先頭に記述されている文字またはベクトルを最優先で描画するよう指定があった場合には、後から描画される文字またはベクトルのエッジリストを既存の出力用エッジリストに合成する際に、同じ画素位置にリストのデータがなければリストを連結し、同じものがある場合には何もしない。
【００７４】
次に、上述した合成部１８での処理の具体例を説明する。例えば、表５に示す多値エッジリストが多値文字処理部１２から出力され、その後、表１１に示す２値エッジリストが２値ベクトル処理部１５から出力されたとする。
【表１１】

【００７５】
なお、ＰＤＬの記述では、表５に示す多値エッジリストに関する記述、すなわち文字「一」を多値展開する記述が、表１１に示す２値エッジリストに関する記述より先、すなわち副走査方向の画索幅「１」の直線を描画する記述より先にあるとする。ここで、描画順序に関する指定がない場合には、後から描画される文字またはベクトルが優先されるため、表１２に示すようなエッジリストが作成される。つまり、表５に示す多値エッジリストにデータがない６ライン目、７ライン目および９ライン目の８番目の画素については、表１１に示す２値エッジリストのデータが連結され、８ライン目の８番目の画素についてはともにデータが存在しているため、表１１に示す２値エッジリストのデータが挿入されている。
【表１２】

【００７６】
一方、文字「一」を最優先で多値展開する指定があった場合には、同じ画素位置にデータがあっても、指定された文字またはベクトルが優先されるため、表１３に示すようなエッジリストが作成される。つまり、表５に示す多値エッジリストにデータがない６ライン目、７ライン目および９ライン目の８番目の画素については、表１１に示す２値エッジリストのデータが連結され、８ライン目の８番目の画素についてはともにデータが存在しているが、指定により多値エッジリストのデータが優先されるため、リストの操作は何も行われない。
【表１３】

【００７７】
以上説明したように、合成部１８では、リストの合成処理が行われ、最終的にラスター画像に変換される。ラスター画像への変換は、ランレングス形式で記述されているエッジリストをその記述内容に従ってラスター画像に変換するのみであるので、詳細な説明は省略する。また、合成部１８では、多値エッジリストまたは２値エッジリストが多値文字処理部１２、２値文字処理部１３、多値ベクトル処理部１４、２値ベクトル処理部１５のいずれから出力されたかを検出し、その結果を識別データとして上記ラスター画像の画素毎に出力端Ｂから外部に出力する。なお、識別データ作成の際には、上述したリストの合成時の描画順序に関する処理結果が反映される。
【００７８】
ところで、上述したリストの合成処理の際には、同じ画素位置に、リストにデータがあるかどうかを調べるため、異なる処理部から出力されたエッジリストが表す画像の境界を容易に知ることができる。したがって、このような境界に対する処理、例えばグラフィクス領域中に黒色文字が埋め込まれているような場合に、プリント時のレジストレーションのずれを目立たなくするような境界処理も容易に行える。この境界処理は、以下のようにして実現できる。
【００７９】
まず、多値文字処理部１２、２値文字処理部１３、多値ベクトル処理部１４、２値ベクトル処理部１５のいずれかから出力されたエッジリストから画像の境界を検出する。この検出は、当該エッジリストをラスター画像に変換した場合に「０」でない画素値の画素群の主走査方向の始点と終点とを検出すること、つまりエッジリスト内で同じラインに位置する各リストデータの主走査方向の始点と終点とを調べ、さらに１つのリストデータの終点と他のリストデータの始点とが主走査方向に隣接する場合には、その２つのリストデータの隣接する始点と終点とを検出対象からはずすことにより行うことができる。
【００８０】
次に、合成部１８において、グラフィクス処理部１６またはイメージ処理部１７から出力されたエッジリストと、多値文字処理部１２、２値文字処理部１３、多値ベクトル処理部１４、２値ベクトル処理部１５のいずれかから出力されたエッジリストとを合成する際には、例えばグラフィクス処理部１６またはイメージ処理部１７から出力されたエッジリスト中のリストデータで、検出した画像の境界と始点あるいは終点とが一致するとき、一致した点が始点であった場合には、その値をデクリメントし、終点であった場合には、インクリメントする。このように境界処理を行えば、プリント時のレジストレーションのずれを目立たなくすることができる。
【００８１】
上述した処理に対して、エッジリスト合成時にリストデータを変更するのではなく、境界であることをどの処理部から出力されたかを示す識別データとは異なる内容の識別データとして出力端Ｂから外部に出力するようにしてもよい。
【００８２】
また、別の境界処理として、多値文字と背景との重ね合わせ処理がある。以下では、当該重ね合わせ処理の例を図１４を参照して説明する。ここで、図１９（ａ）は、色成分をＹＭＣＫの４成分として背景上に理想的に展開した例を示す概念図であり、各色成分の各画素は２５６階調である。図１９（ｂ）、（ｃ）は、図１９（ａ）に示した画像のうち、文字を多値展開した画像のそれぞれＫ成分・Ｃ成分を示す概念図、図１９（ｄ）は、図１９（ｂ）、（ｃ）に示す画像をプリントアウトした例を示す概念図である。なお、図１９において、矩形は画素を表し、白色でない矩形の画素値は階調を有する画素であり、図中の凡例に従う。
【００８３】
通常、文字を白色ではない背景に重ね合わせる場合、図１９（ｃ）に示すように、背景の画素のうち文字の有色画素と同じ位置にある画素は、文字の画素に置き換えられる。これに対して、多値展開した文字は、パルス幅変調等の解像度を拡張する印字方法により印字されるため、特に画素値が中間調である画素については、図１９（ｄ）に示すように、色成分間で白隙を生じることがある。したがって、多値展開文字の中間調画素を背景に重ね合わせる場合には、背景画素を多値展開文字の画素に単純に置き換えるのではなく、背景画素の色成分のうち、多値展開文字とは異なる色成分の画素の値を保存し、同一色成分の画素のみ背景画素を文字画素に置き換えるようにすればよい。図１９（ｅ）は、図１９（ａ）に示す画像を文字と背景の重ね合わせを上述した方法で行った場合のプリントアウト例を示す概念図である。上述した境界処理の実現方法は、プリント時のレジストレーションのずれを目立たなくする境界処理と同様であるため、詳細な説明は省略する。なお、上述において、ラスターイメージは、ラスターイメージ処理部１７でエッジリストに変換された後、他の処理部から出力されたエッジリストと合成処理されるとしているが、ラスターイメージ処理部１７では、ラスターデータの取り出しだけを行い、ラスターイメージ処理部１７から出力されるエッジリストを合成し、ラスタライズした後、当該ラスターデータとの合成処理を行うようにしてもよい。
【００８４】
Ｃ．変形例
以上、本発明の一実施形態について説明したが、本発明の実施形態は、これに限定されるわけではなく、例えば上記実施形態の合成部１８においては、複数の処理部から出力されたエッジリストをラスタライズする前に合成処理を行っているが、各エッジリストをラスタライズした後に合成処理を行っても、本発明の趣旨は何ら逸脱しない。以下では、本発明の変形例として後者の実施形態について説明する。
【００８５】
この変形例では、上記実施形態とは、合成部１８における合成処理が異なるのみで、コマンド分離部１１および各処理部における処理については同じであるため説明を省略する。合成部１８は、まず、各処理部から出力されたエッジリストをラスタライズする。ラスタライズした結果得られるラスター画像は、合成処理を行うために合成部１８内のメモリ（図示せず）に書き込まれる。メモリへは、メモリ上のラスター画像がそのまま外部に出力できるよう書き込まれる。具体的には、書き込もうとする画素のメモリ上での位置関係が画像全体での位置関係と一致するように書き込まれる。
【００８６】
しかしながら、書き込みの際には、他の処理部から出力されたエッジリストをラスタライズしたラスト画像が書き込まれている場合がある。したがって、この場合には、上記実施形態と同様に、ＰＤＬ内での描画順序に応じて書き込もうとする画素を、書き込もうとする位置の既存の画素に上書きするのか、上書きせずに既存の画素を保存するのかを決定する。この画素の上書きの可否決定は、上記実施形態の合成部１８におけるエッジリスト間の合成処理に相当するが、エッジリストとラスター画像との違いはあるが合成論理は変わらない。したがって、出力端Ａから外部に出力される画像については何ら相違はない。
【００８７】
また、上記実施形態の合成部１８における文字と背景との境界処理についてもエッジリスト間の合成処理と同様にラスター画像間で処理を行うことができる。上記実施形態では、文字と背景との境界を検出した後、境界部分のエッジリストの内容を操作することにより、プリント時のレジストレーションのずれを防止する処理あるいは多値展開文字と背景とを重ね合わせる際の多値文字周辺の白隙を防止する処理を行っている。これに対して、本変形例では、境界を検出した後、メモリに書き込む際に上記処理を行う。具体的には、プリント時のレジストレーションのずれを防止するために、境界部分の文字の画素をメモリに書き込む前に、その画素の主走査方向の１つ前あるいは１つ後の画素値のうち、文字とは異なる色成分の画素値を書き込む文字の画素の同一色成分の画素値とする。また、多値展開文字と背景とを重ね合わせる際に多値文字周辺の白隙を防止するために、多値展開文字の中間調画素をメモリに書き込む前に、その画素の主走査方向の１つ前あるいは１つ後の画素値のうち、文字とは異なる色成分の画素値を、書き込む文字の画素の同一色成分の画素値とする。
【００８８】
いずれも、上記実施形態でエッジリストの内容を繰作する処理を、ラスター画像間の処理に置き換えただけであり、両者とも最終的に出力端Ａから出力される画像の内容に相違はない。なお、書き込む画素の文字とは異なる色成分の画素値を、当該画素の主走査方向の１つ前あるいは１つ後の画素の同一色成分の値としているが、１つ前あるいは１つ後の画素のどちらを選択するかは上記実施形態における、プリント時のレジストレーションのずれを防止する処理での選択と同一論理とする。
【００８９】
ところで、上述した実施形態および変形例は、いずれもＰＤＬを出力するアプリケーションソフトウェアとプリントエンジンとの間に実装すればよいので、いくつかの異なる形態により実現することができる。例えば、印字を行うプリントエンジンと、プリントエンジンの制御、外部との通信の制御等を行うプリントコントローラとが１つの筐体に納められているプリンタでは、プリントコントローラで動作するソフトウェアとして実現できる。また、プリントエンジンとプリントコントローラとが別個の筐体となっている場合にも、プリントコントローラで動作するソフトウェアとして実現できる。あるいは、ＰＤＬを出力するアプリケーションソフトウェアが搭載されているＰＣあるいはＷＳで動作するソフトウェアとしても実現できる。さらに、処理速度を速めるために上記実施形態における処理の一部あるいは全部をハードウェアで実現してもよく、いずれの実施形態で実現しても本発明の趣旨を何ら逸脱するものではない。
【００９０】
以上、本発明による画像処理装置の実施例について述べたが、本発明による画像処理装置の実現形態は、上述した内容に限定されるものではなく、他の実現形態であっても本発明の趣旨を逸脱しない範囲ならばどのような実現形態であっても構わない。
【００９１】
【発明の効果】
以上、説明したように、この発明によれば、ページ記述言語で表され、入力手段から入力された画像情報の中から抽出手段により文字または線画部を抽出し、抽出した文字または線画部を第１の変換手段により２値の階調を有する第１の中間コードに変換し、第１の中間コードを第２の変換手段により上記第１の中間コードより低い解像度で多値の階調を有する第２の中間コードに変換し、また、タグ生成手段により、第１の中間コードまたは第２の中間コードに基づいて、入力された画像情報の所定の領域ごとに当該領域を占める画像の種類を表し、当該第２の中間コードで表された文字または線画部であることを表す情報を少なくとも含むタグ情報を生成するようにしたので、ページ記述言語で記述された画像情報をラスター画像に変換する際に必要な記憶容量を少なくすることができ、より高画質のプリントアウトを得ることができるという利点が得られる。
【図面の簡単な説明】
【図１】 本発明による画像処理装置の一構成例を示すブロック図である。
【図２】 本発明による画像処理装置に入力されるＰＤＬの一例を示す概念図である。
【図３】 エッジリスト生成過程を示すフローチャートである。
【図４】 ２値エッジリスト作成処理の詳細を示すフローチャートである。
【図５】 ２値エッジリスト作成処理の詳細を示すフローチャートである。
【図６】 ２値エッジリスト作成処理の詳細を示すフローチャートである。
【図７】 ２値エッジリスト作成処理の詳細を示すフローチャートである。
【図８】 先頭ライン処理の詳細を示すフローチャートである。
【図９】 先頭ライン処理の詳細を示すフローチャートである。
【図１０】 最終ライン処理の詳細を説明するためのフローチャートである。
【図１１】 最終ライン処理の詳細を説明するためのフローチャートである。
【図１２】 変数ｎ１を算出するアルゴリズムを説明するためのフローチャートである。
【図１３】 変数ｎ１を算出するアルゴリズムを説明するためのフローチャートである。
【図１４】 処理過程でのラスタティズ画像ならびに比較対照のための概念図である。
【図１５】 図１４（ａ），（ｂ）の部分拡大図である。
【図１６】 多値ベクトル処理部１４での処理を示すフローチャートである。
【図１７】 多値ベクトル処理部１４での処理の具体例を説明するための概念図である。
【図１８】 グラフィクス処理部１６での指定された領域を一様に塗り潰さない方法を説明するための概念図である。
【図１９】 多値文字と背景との重ね合わせ処理を説明するための概念図である。
【符号の説明】
１１ コマンド分離部（入力手段、抽出手段）
１２ 多値文字処理部（第１の変換手段）
１３ ２値文字処理部（第１の変換手段）
１４ 多値ベクトル処理部（第１の変換手段）
１５ ２値ベクトル処理部（第１の変換手段）
１６ グラフィクス処理部
１７ ラスターイメージ処理部
１８ 合成部（第２の変換手段、タグ生成手段、合成手段、境界検知手段）[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image processing apparatus, and more particularly to an image processing apparatus that converts character / line drawing data representing a character / line drawing into raster image data and outputs the raster image data.
[0002]
[Prior art]
In recent years, there are increasing opportunities to print out documents created on personal computers (hereinafter referred to as PCs) and workstations (hereinafter referred to as WSs). In particular, many documents printed out in the office have text / graphics or photographs read by a scanner laid out on the same page, and recently there is an opportunity to print out color documents colored in various colors. is increasing. In addition, a system configuration in which a plurality of PCs and WSs share a printer via a network is often adopted, and the demand for higher speed and higher image quality is increasing.
[0003]
By the way, in order to shorten the data transfer time from the PC or WS to the printer and to print out the document with the same contents as much as possible even with printers having different characteristics, a document is described in a print description language (hereinafter referred to as PDL) which is a collection of drawing commands. In general, the data is transferred to a printer, interpreted by the printer, converted into a raster image, and printed out. However, if PDL is converted into a raster image in accordance with the resolution of an existing printer and printed out, jaggies are generated particularly at the edges of characters and line drawings.
[0004]
Therefore, in order to reduce the above-described jaggy, an area where jaggy can occur is extracted from the converted raster image by pattern matching as in the technique disclosed in US Pat. No. 4,847,641, for example. A method of performing correction is proposed. However, according to this method, even areas that should not be corrected, such as the mines and folds of Mincho characters, are corrected and cause deformation of the font. On the other hand, when straight line groups with uniform thickness, equal spacing, and parallel are converted to raster images due to resolution limitations, the thickness may vary, or may be converted to parallel straight line groups with no equal spacing. In such a case, it cannot be corrected by the above prior art.
[0005]
Therefore, for example, as in the technique disclosed in Japanese Patent Laid-Open No. 6-40074, the PDL is once converted into a binary raster image having a resolution higher than the resolution of the printer, and the raster image is the same as the resolution of the printer by resolution conversion. There has been proposed a method of converting to a multi-value raster image having a resolution and changing the dot size in accordance with the pixel value. However, in the latter technique, since it is necessary to convert the PDL into a binary raster image having a resolution higher than that of the printer, a large storage area is required as compared with the former technique. Therefore, like the technique disclosed in Japanese Patent Laid-Open No. 6-168334, the PDL is once converted into binary code data representing a binary raster image having a resolution higher than the resolution of the printer, and the binary code data is converted into the printer's resolution. There has been proposed a method of converting multi-value code data into a multi-value raster image by converting into multi-value code data representing a multi-value raster image having the same resolution as the resolution.
[0006]
[Problems to be solved by the invention]
By the way, in a document printed out by a printer, there are cases where graphics and photographs are mixed in addition to characters and line drawings. In order to obtain a printout with high image quality, resolution is required for characters and line drawings, and gradation is required for patterns. Therefore, processing for reducing jaggies is applied to characters and line drawings as in the above-described technique, and area gradation processing is applied to the pattern.
[0007]
In order to further improve the image quality, there is a technique in which the classification is further refined, for example, the contents of the area gradation processing are different between graphics and photographs, and the processing applied for each classification is changed. However, it is difficult to completely separate characters, line drawings, graphics, photographs, and the like from raster images. Therefore, in the prior art, there are cases where area gradation processing is applied to characters or line drawings, or processing for reducing jaggy is applied to a pattern, and it is difficult to improve image quality.
[0008]
The present invention has been made in view of the above-described circumstances, and can reduce the storage capacity required for converting image information described in a page description language into a raster image, thereby obtaining a higher quality printout. An object of the present invention is to provide an image processing apparatus.
[0009]
[Means for Solving the Problems]
In order to solve the above-described problem, in the first aspect of the present invention, an input means for inputting image information described in a page description language, and a character or line drawing from the image information input by the input means. An extraction means for extracting a portion and a character or line drawing portion extracted by the extraction means is an intermediate code for raster development, and is converted into a first intermediate code having binary gradations A conversion unit and a second conversion for converting the first intermediate code converted by the first conversion unit into a second intermediate code having a lower resolution than the first intermediate code and having a multi-valued gradation And a predetermined intermediate value of the image information input by the input means based on the first intermediate code converted by the first conversion means or the second intermediate code converted by the second conversion means For each area Tag generating means for generating tag information representing the type of image occupying the area, and the tag information generated by the tag generating means is a character or line drawing part represented by the second intermediate code. Including at least information representing the above.
[0010]
According to the present invention, a character or line drawing part is extracted from the image information expressed in the page description language by the extraction means from the image information input from the input means, and the extracted character or line drawing part is binarized by the first conversion means. Is converted into a first intermediate code having a plurality of gradations, and the first intermediate code is converted into a second intermediate code having a multilevel gradation at a resolution lower than that of the first intermediate code by the second conversion means. To do. Further, the tag generation means represents the type of image occupying the area for each predetermined area of the input image information based on the first intermediate code or the second intermediate code, and the second intermediate code Tag information including at least information indicating that the character or line drawing portion is represented is generated. Accordingly, it is possible to reduce the storage capacity required for converting image information described in the page description language into a raster image, and to obtain a higher quality printout.
[0011]
DETAILED DESCRIPTION OF THE INVENTION
Next, embodiments of the present invention will be described with reference to the drawings. In the following, in the image data, the number of gradations per pixel of each color component is “256”, the number with “b” added at the end is binary notation, and the number with “h” added. Is in hexadecimal notation, and numbers with nothing added are in decimal notation. In the following description, only one color component process will be described, but for multiple color component processes, one color component process is applied by the number of color components, or all color components are simultaneously processed in parallel. It may be processed.
[0012]
First, code data in a format in which image data is run-length coded will be described. The code data is obtained by run-length coding a raster image in units of lines, and describes how many pixels of a certain value continue from a certain position in the main scanning direction. For example, when a pixel having a pixel value “255 (black pixel)” is 20 pixels continuous from the tenth pixel from the top of the line, a set of three integer values “255”, “10”, and “20” is set. Described as data. At this time, since it is not necessary to describe the pixel having the pixel value “0”, the data amount can be compressed as compared with the raster image. Further, when a binary image is targeted, it is not necessary to describe pixel values, so that the data amount can be further compressed. On the other hand, since various image processing with a raster image is replaced with an operation between code data, processing can be performed at higher speed. Hereinafter, the above-described code data will be referred to as an edge list, and a specific example of the edge list and operations between the edge lists will be described later.
[0013]
A. Configuration of the embodiment
FIG. 1 is a block diagram illustrating a configuration example of an image processing apparatus according to an embodiment of the present invention. In the figure, 11 is a command separation unit, 12 is a multi-value character processing unit, 13 is a binary character processing unit, 14 is a multi-value vector processing unit, 15 is a binary vector processing unit, 16 is a graphics processing unit, and 17 is a raster. An image processing unit 18 is a combining unit. The command separation unit 11 converts the PDL input to the input terminal A into a multi-value character drawing command, a binary character drawing command, a multi-value vector drawing command, a binary vector drawing command, a graphics drawing command, and a raster image drawing command. The commands are classified into types of commands and supplied to the multi-value character processing unit 12, the binary character processing unit 13, the multi-value vector processing unit 14, the binary vector processing unit 15, the graphics processing unit 16, and the raster image drawing command 17, respectively. .
[0014]
The multi-value character drawing command and the binary character drawing command are composed of a character code such as ASCII code and information indicating print characteristics such as a typeface name, the number of points, a face, a weight, and a color. Information indicating whether to perform multi-value or binary is also included. Multi-value vector drawing commands and binary vector drawing commands are information representing lines such as straight lines and curves, such as coordinate information such as start points, end points, and via points, and line widths, line types, line colors, etc. It includes image information, and further includes information indicating whether vector drawing is performed in multi-value or binary. The graphics drawing command is information representing a color for filling a region, and includes coordinate information of the region to be filled and information representing a color to be painted and a painting method. The raster image drawing command represents a raster image in PDL, and includes information such as the number of pixels, resolution, and color component configuration in the main scanning and sub-scanning directions.
[0015]
The multi-value character processing unit 12 converts the input multi-value character drawing command into a binary edge list (hereinafter referred to as a high-resolution edge list) having a resolution n times the printer resolution (n is an integer of 2 or more). After that, the edge list is converted into a multi-valued edge list (hereinafter referred to as a multi-valued edge list) having the same resolution as the printer resolution, and supplied to the synthesizing unit 18. The binary character processing unit 13 converts the input binary character drawing command into a binary edge list (hereinafter referred to as a binary edge list) having the same resolution as the printer resolution, and supplies it to the synthesis unit 18. The multi-value vector processing unit 14 converts the input multi-value vector command into a high-resolution edge list, converts the edge list into a multi-value edge list, and supplies it to the synthesis unit 18.
[0016]
The binary vector processing unit 15 converts the input vector command into a binary edge list and supplies it to the synthesis unit 18. The graphics processing unit 16 converts the input graphics drawing command into a multi-valued edge list and supplies it to the synthesis unit 18. The raster image processing unit 17 converts the input raster image drawing command into a multi-value edge list and supplies it to the synthesis unit 18. The synthesizing unit 18 receives the edge list supplied from the multi-value character processing unit 12, the binary character processing unit 13, the multi-value vector processing unit 14, the binary vector processing unit 15, the graphics processing unit 16, and the raster image processing unit 17. At the same time, it is converted into a raster image and output to the output terminal A. Simultaneously with the conversion to the raster image, classification data indicating the classification of each pixel of the raster image is generated and output to the output terminal B. Note that the classification data includes information indicating which processing unit the edge list is output from, and information indicating the boundaries of raster images respectively represented by edge lists output from different processing units.
[0017]
Here, FIG. 2 is a conceptual diagram showing an example of PDL. Hereinafter, the case where the PDL illustrated in FIG. 2 is input to the input terminal A will be described in detail. In FIG. 2, TYPE is a character drawing command, LINE is a straight line drawing command, PAINT is a graphics drawing command, and RASTER is a raster image drawing command. The part enclosed in parentheses () following the command name is the parameter of each command, and the part enclosed in parentheses {} is raster image data. The four parameters of the TYPE command are a drawing start x coordinate, a drawing start y coordinate, a font type, the number of points, a drawing character string, and a drawing mode in order from the top. The first line in FIG. 2 is a character string “abcdefg”. Means that the coordinates are (x1, y1) as a starting point and the second line is drawn in the binary mode.
[0018]
The six parameters of the LINE command are the drawing start x coordinate, the drawing start y coordinate, the drawing end x coordinate, the drawing end y coordinate, the line width, and the drawing mode in order from the top. The third line in FIG. It means that a straight line with a thickness of 1 point is drawn from (x2, y2) to (x3, y3) in the multi-value mode and the fourth line in the binary mode. Next, the fifth line in FIG. 2 fills a rectangular area whose diagonal is a straight line connecting the coordinates (x4, y4) and the coordinates (x5, y5) with the specified color RED (red) with the color space as RGB. Means that. Further, the sixth line in FIG. 2 has a raster image of p pixels × l line with coordinates (x6, y6) as the upper left corner, RGB color space, resolution rx in the main scanning direction, and resolution ry in the sub scanning direction. Means drawing according to the last image data in parentheses {}. Note that the commands and parameters of each command are examples, and the command name and parameter contents of each command are not limited to the above contents. Further, the setting of commands for each drawing operation is not limited to the above contents, and may be changed, for example, such that the TYPE command is divided into two or more commands.
[0019]
The command separation unit 11 performs multi-valued vector drawing by drawing the first line shown in FIG. 2 as the multi-valued character drawing command to the multi-valued character processing unit 12 and the second line as the binary character drawing command to the binary character processing unit 13. A third line as a command to the multi-value vector processing unit 14, a fourth line as a binary vector drawing command to the binary vector processing unit 15, a fifth line as a graphics drawing command to the graphics processing unit 16, a raster image drawing command Are output to the raster image processing unit 17, respectively.
[0020]
The multi-value character processing unit 12 interprets and executes the command on the first line input as the multi-value character drawing command. That is, the character string is drawn with the designated font in the multi-value mode. There are two types of fonts: bitmap fonts that describe characters in raster image data, and outline fonts that describe character outlines as a collection of lines such as lines and curves, with line drawing information (outline data). In this embodiment, an outline font is used because of the ease of deformation processing corresponding to the number, face, and weight and the small number of defects after the deformation processing. When the outline font is used, the outline data of the specified character is read out, the outline of the character is drawn on the virtual coordinate plane based on the outline data, and the closed area surrounded by the outline is filled with the virtual coordinates An ideal character image is obtained on a plane. Then, a multi-valued edge list is generated based on the outline of the character on the virtual coordinate plane (hereinafter simply referred to as “character outline”), and is output to the synthesis unit 18.
[0021]
The binary character processing unit 13 interprets and executes the command on the second line input as the binary character drawing command. That is, the character string is drawn with the designated font in the binary mode. Here, an ideal character image is obtained on the virtual coordinate plane in the same manner as the multi-value character processing unit 12. Then, a binary edge list is generated based on the character outline and output to the synthesis unit 18. The multi-value vector processing unit 14 interprets and executes the command on the third line input as the multi-value vector drawing command. That is, the vector is drawn in the multi-value mode. First, a straight line with two specified coordinate points on the virtual coordinate plane is drawn, and an area at half the distance of the specified line width is filled from the straight line, so that it is ideal on the virtual coordinate plane. A simple vector image. Then, a multi-value edge list is generated based on the vector contour on the virtual coordinate plane (hereinafter simply referred to as a vector contour), and is output to the synthesis unit 18.
[0022]
Next, the binary vector processing unit 15 interprets and executes the command on the fourth line input as the binary vector drawing command. That is, the vector is drawn in the binary mode. Here, an ideal vector image is obtained on the virtual coordinate plane in the same manner as the multi-value vector processing unit 14. Then, a binary edge list is generated based on the contour of the vector on the virtual coordinate plane, and is output to the synthesis unit 18. The graphics processing unit 16 interprets and executes the command on the fifth line input as the graphics drawing command. That is, the specified area is filled with the specified color. First, a filled area is set on the virtual coordinate plane, and the area is filled with a designated color. Then, a binary edge list is generated based on the painted area on the virtual plane, and is output to the synthesis unit 18. It should be noted that the color space designated at the time of painting is naturally converted to a color space suitable for the printer, and further, color approximation processing is performed according to the color reproduction characteristics of the printer. The contents of the color space conversion process and the color approximation process may be anything.
[0023]
Next, the raster image processing unit 17 interprets and executes the command on the sixth line input as the raster image drawing command. That is, drawing is performed in the specified color space and the specified resolution in the specified area according to the specified raster image data. First, a drawing area corresponding to the number of pixels in the main scanning and sub-scanning directions and the resolution is set on the virtual coordinate plane, and a raster image is drawn according to the raster image data designated in the color space designated in the area. It should be noted that the color space designated at the time of painting is naturally converted to a color space suitable for the printer, and further, color approximation processing is performed according to the color reproduction characteristics of the printer. If the designated resolution and the printer resolution do not match, resolution conversion processing is performed. The contents of the color space conversion process, the color approximation process, and the resolution conversion process may be anything.
[0024]
The synthesizing unit 18 receives the edge list supplied from the multi-value character processing unit 12, the binary character processing unit 13, the multi-value vector processing unit 14, the binary vector processing unit 15, the graphics processing unit 16, and the raster image processing unit 17. It is synthesized into one edge list, converted into a raster image, and output from the output terminal A to the outside. Further, data indicating which processing unit outputs each edge list input to the combining unit 18 is generated for each pixel, and is output to the outside from the output terminal B while being synchronized with the raster image. The synthesizing process in the synthesizing unit 18 is realized by an operation between the edge lists.
[0025]
B. Operation of the embodiment
Next, the operation of the above-described embodiment will be described. First, generation of the multi-value edge list from the outline of the character on the virtual coordinate plane in the multi-value character processing unit 12 will be described in detail. FIG. 3 is a flowchart for explaining an edge list generation process. First, outline data corresponding to a character to be drawn is loaded (step S100), a high-resolution edge list is created (step S101), and the edge list is converted into a multi-value edge list (step S102). When all the characters in the character string specified by the TYPE command have been drawn, the multi-value edge list is output to the synthesizing unit 18 to finish the process. repeat. In the following description, n = 2.
[0026]
FIG. 4 is a flowchart showing details of the processing in step S101 in FIG. In FIG. 4, i is a processing position in the main scanning direction, j is an index indicating a processing position in the sub scanning direction, st is a filling start position in the main scanning direction (hereinafter referred to as a filling start position), and rl is This is a variable for storing the number of continuous pixels to be filled (hereinafter referred to as the number of continuous pixels to be filled), and when fl is “1”, valid data remains at the fill start position st and the continuous number rl. When it is “0”, it is a flag indicating that it does not remain. P and L are the numbers of pixels in the main scanning direction and the sub-scanning direction of an area where one character is drawn. The symbol = means that the number of the right side or the calculation result is stored in the variable on the left side, and the symbol == means that both sides are equal.
[0027]
First, variables necessary for processing are initialized (steps S200 and S201). Next, it is determined whether or not the virtual coordinate point TP (i, j) indicated by the indexes i and j is in the filled area. If the virtual coordinate point TP (i, j) is in the filled area, the process from step S203 is performed. Performs the processing after step S206 (step S202). The virtual coordinate point TP (i, j) is the center point of a pixel having a resolution twice that of the printer. When the virtual coordinate point TP (i, j) is in the filled area, the filled pixel continuous number rl is incremented (step S203), whether or not the flag fl is “0”, that is, the filling start position st and the filled area. It is checked whether or not valid data remains in the number of continuous pixels rl (step S204). If there is no remaining data, the current position in the main scanning direction is set as the filling start position st in the main scanning direction (step S205).
[0028]
On the other hand, when the virtual coordinate point TP (i, j) is not in the filled area, whether or not the flag fl is “1”, that is, whether or not valid data remains in the filled start position st and the filled pixel continuous number rl. (Step S206), if it remains, the paint start position st and the paint pixel continuous number rl are registered in the list (step S207), and the paint start position st, the paint pixel continuous number rl and the flag fl are initialized. (Step S208). When step S205 or step S208 ends or when the determination result in step S204 or step S206 is “NO”, the index i is incremented (step S209), and the value obtained by doubling the number P of pixels in the main scanning direction And index i are compared (step S210). If they are equal, the process from step S211 is performed, and if they are not equal, the process from step S202 is performed.
[0029]
When the index i is equal to a value obtained by doubling the number P of pixels in the main scanning direction, whether or not the flag fl is “1”, that is, whether valid data remains in the paint start position st and the paint pixel continuous number rl. It is checked whether or not it remains (step S211), and if it remains, the paint start position st and the paint pixel continuous number rl are registered in the list (step S212). When the process in step S212 ends or the determination result in step S211 is “NO”, the end code EOL is registered in the list and the index j is incremented (step S213), and the sub-scanning direction The value obtained by doubling the number of pixels L is compared with the index j (step S214). If they are equal, the process ends. Through the processing described above, a binary edge list having a resolution twice the printer resolution is created. In the above-described processing, since the edge list to be created is binary, only the start position and the continuous number of pixels to be printed are the contents of the list. Further, as can be seen from the fact that the end code EOL is registered in the list at the end of the line, the edge list is created in units of lines.
[0030]
Next, FIGS. 5 to 7 are flowcharts showing details of the processing in step S1O2 of FIG. 5 to 7, i is a processing position in the main scanning direction, j is an index indicating a processing position in the sub scanning direction, st1 to st6 are fill start positions in the main scanning direction, and rl1 to rl6 are filled. This is a variable for storing the number of continuous pixels, st is the fill start position in the main scanning direction registered in the multi-valued edge list, rl is the number of continuous pixels to be registered in the multi-valued edge list, and pv is the multi-valued edge list This is a variable for storing the pixel value to be registered in. Further, n2 is a pixel of interest, n1 and n3 are variables representing the state of the pixels immediately above and directly below, and m is a variable for storing a pixel value. The meanings of the numbers represented by P and L, the symbol = and the symbol == are the same as those in FIG. 4, and the function LUT-A and the function LUT-B are arguments using different lookup tables (hereinafter referred to as LUT). Means a function that transforms.
[0031]
First, an initial value is set for the index i (step S300), and the first line is processed (step S301). The processing of the first line will be described later. Next, the index i is incremented (step S302), ST1, RL1 to ST6, and RL6 are read, and variables necessary for processing are initialized (step S303). STx and RLx (x = 1 to 6) are the fill start positions in the main scanning direction read from the list of the (2 · j + x−3) th line in the binary edge list created in step S102 of FIG. And the continuous number of pixels to be filled. For example, when j = 1, ST1 and RL1 are the fill start position in the main scanning direction read from the list of the 0th line (that is, the top line) and the number of continuous pixels to be filled. Next, variables n1 and n3 representing pixel states are calculated (step S304). Table 1 shows the calculated values and pixel states.
[0032]
[Table 1]

In Table 1, the figure in the state column schematically shows the state of the portion corresponding to the target pixel in the binary edge list created in step S101 in FIG. 3, and the white rectangle is not filled in. The black pixel represents a pixel to be filled. This calculation method will be described later.
[0033]
Next, it is checked whether or not the variable n2 is equal to “3” (step S305). If they are equal, the processing from step S306 is performed, and if they are not equal, the processing from step S309 is performed. When the variable n2 is equal to “3”, it is checked whether or not the variable n3 is equal to “0” or “12” (step S306). If the variable n2 is equal to either, the variable n2 is set to the concentration value by the LUT shown in Table 3. The data is converted and stored in the variable m (step S307). On the other hand, if they are not equal to each other, the variable n2 is converted into a density value by the LUT shown in Table 2 and stored in the variable m (step S308).
[Table 2]

[0034]
[Table 3]

[0035]
On the other hand, if the variable n2 is not equal to “3”, it is checked whether or not the variable n2 is equal to “12” (step S309). If they are equal, the processing from step S310 is performed, and if they are not equal, the processing from step S312 is performed. . If the variable n2 is equal to “12”, it is checked whether the variable n1 is equal to “0” or “3” (step S310). If they are equal, they are converted into density values by the LUT shown in Table 3 and stored in the variable m. (Step S311), if not equal, it is converted into a density value by the LUT shown in Table 2 and stored in a variable dish (Step S312).
[0036]
When the processing in any of the above steps S307, S308, S311 and S312 is completed, it is checked whether the variable m and the variable pv in which the density value is stored are larger than “0” (step S310). If both the variable m and the variable pv are larger than “0”, it is checked whether the variable m and the variable pv are equal (step S312). If they are equal, the variable rl is incremented (step S313). Variable pv and variable rl are registered in the list (step S314), index i is set in variable st, variable m is set in variable pv, and "1" is set in variable rl (step S315).
[0037]
On the other hand, if the variable m is “0” and the variable pv is greater than “0”, the variable st, the variable pv, and the variable rl are registered in the list (step S317), and “0” is set to the variable pV (step S317). S318). If the variable m is larger than “0” and the variable pv is “0”, the index i is set in the variable st, the variable m is set in the variable pv, and “1” is set in the variable rl (step S316). That is, the process from step S310 to step S319 is a run-length coding process.
[0038]
Next, when the processing in step S313, step S315, step S316, or step S318 ends, or if both the variable m and the variable pv are “0” in step S310, the index i is incremented (step S319). ), Whether or not the index i is equal to P is checked (step S320). If they are equal, the process from step S321 is performed, and if they are not equal, the process from step S304 is performed.
[0039]
When the index i is equal to P, it is checked whether or not the variable pv is larger than “0” (step S321). If larger, the variable st, the variable pv and the variable rl are registered in the list (step S322). When the process in step S322 ends or the determination result in step S321 is “NO”, the EOL is registered in the list and the index j is incremented (step S323). It is checked whether it is equal to the value obtained by subtracting “1” (step S324). If they are equal, a final line process described later is performed (step S325), and all the processes are terminated. If they are not equal, the process from step S303 is performed.
[0040]
Here, the head line processing in step S301 in FIG. 5 will be described. 8 and 9 are flowcharts showing details of the head line processing. In FIG. 6, indexes, variables, symbols, and functions having the same names as those in FIGS. 5 to 7 are the same as those in FIGS.
[0041]
First, ST3, RL3 to ST6, and RL6 are read, and variables necessary for processing are initialized (step S400). Next, variables n2 and n3 representing pixel states are calculated (step S401). Next, it is checked whether or not the variable n2 is equal to “3” (step S402). If they are equal, the processing from step S403 is performed, and if they are not equal, the processing from step S405 is performed. If the variable n2 is equal to “3”, it is checked whether the variable n3 is equal to “0” or “12” (step S403). If the variable n2 is equal to either, the variable n2 is determined by the LUT shown in Table 3. It is converted into a density value and stored in a variable m (step S404). On the other hand, if it is not equal to either, the variable n2 is converted into a density value by the LUT shown in Table 2 and stored in the variable m (step S405).
[0042]
When the process in either step S404 or step S405 is completed, it is checked whether the variable m and the variable pv in which the density value is stored are larger than “0” (step S406), and the variable m and the variable pv are determined. If both are larger than “0”, it is checked whether the variable m and the variable pv are equal (step S407). If they are equal, the variable rl is incremented (step S408). If they are not equal, the variable st, the variable pv, and the variable rl are registered in the list (step S409), the variable i is index i, the variable pv is variable m, the variable “1” is set to rl (step S410). On the other hand, if the variable m is “0” and the variable pv is greater than “0”, the variable st, the variable pv, and the variable rl are registered in the list (step S412), and “0” is set to the variable pv (step S412). S413). If the variable m is greater than “0” and the variable pv is “0”, the index i is set in the variable st, the variable m is set in the variable pv, and “1” is set in the variable rl (step S411). That is, the processing from step S406 to step S413 is a run-length coding process for the top line.
[0043]
Next, when the processing in step S408, step S410, step S411, or step S413 ends, or when both the variable n and the variable pv are “0” in step S406, the index i is incremented (step S414). ), It is checked whether or not the index i is equal to P (step S415). If they are equal, the processing from step S416 is performed, and if they are not equal, the processing from step S401 is performed. If the index i is equal to P, it is checked whether or not the variable pv is greater than “0” (step S416). If larger, the variable st, the variable pv and the variable rl are registered in the list (step S417). If the process in step S417 ends or if the determination result in child step S416 is “NO”, the EOL is registered in the list (step S418), and the process ends.
[0044]
Next, the final line process in step S325 in FIG. 7 will be described. 10 and 11 are flowcharts for explaining the details of the final line processing. 10 and 11, the same names, indexes, variables, symbols, and functions as those in FIGS. 5 to 7 are the same as those in FIGS.
[0045]
First, ST1, RL1 to ST4, and RL4 are read, and variables necessary for processing are initialized (step S500). Next, variables n1 to n2 representing pixel states are calculated (step S501). Next, it is checked whether or not the variable n1 is equal to “12” (step S502). If they are equal, the processing from step S503 is performed, and if they are not equal, the processing from step S505 is performed. If the variable n2 is equal to “12”, it is checked whether the variable n1 is equal to “0” or “3” (step S503). If the variable n2 is equal to either, the variable n2 is determined by the LUT shown in Table 3. It is converted into a density value and stored in a variable m (step S504). On the other hand, if it is not equal to either, the variable n2 is converted into a density value by the LUT shown in Table 2 and stored in the variable m (step S505).
[0046]
When the process in either step S504 or step S505 is completed, it is checked whether the variable m and the variable pv in which the density value is stored are larger than “0” (step S506). If greater than “0”, it is checked whether or not the variable m and the variable pv are equal (step S507), and if equal, the variable rl is incremented (step S508). On the other hand, if not equal, variable st, variable pv and variable rl are registered in the list (step S509), index i is set in variable st, variable m is set in variable pv, and "1" is set in variable rl (step S510). On the other hand, if the variable m is “0” and the variable pv is larger than “0”, the variable st, the variable pv, and the variable rl are registered in the list (step S513), and “0” is set to the variable pv (step S513). S514). If the variable m is larger than “0” and the variable pv is “0”, the index i is set in the variable st, the variable m is set in the variable pv, and “1” is set in the variable rl (step S512). That is, the processing from step S506 to step S514 is a run-length coding process for the last line.
[0047]
Next, when the processing in step S508, step S510, step S512, or step S514 ends, or when both the variable m and the variable pv are “0” in step S506, the index i is incremented (step In step S514, it is checked whether the index i is equal to P (step S515). Then, if both are equal, the processing from step S517 is performed, and if they are not equal, the processing from step S501 is performed. If the index i is equal to P, it is checked whether or not the variable pV is greater than “0” (step S517). If larger, the variable st, the variable pv, and the variable rl are registered in the list (step S518). If the process in step S518 ends or if the determination result in step S517 is “NO”, the EOL is registered in the list (step S519), and the process ends.
[0048]
Next, a method for calculating the variables n1 and n3 in step S304 in FIG. 5, step S401 in FIG. 8, and step S501 in FIG. 10 will be described. Since this calculation method is the same for the variables n1 to n3, only the case of the variable n1 will be described. 12 and 13 are flowcharts for explaining an algorithm for calculating the variable n1. 12 and 18, the same names, variables, and symbols as in FIGS. 7 to 7 are the same as those in FIGS. 5 to 7, and the symbol “|” represents a bit OR operation.
[0049]
First, the variable n1 is initialized (step S600), and it is checked whether or not the index i indicating the position of the target pixel in the main scanning direction is equal to or larger than the variable ST1 and smaller than the sum of the variable ST1 and the variable RL1 (step S601). . That is, it is checked whether or not the target pixel is in the painted area. If the determination result in step S601 is “YES”, that is, if the target pixel is in the filled region, the bit OR operation result of the variable n1 and the constant “1000b” is stored in the variable n1 (step S602). On the other hand, if “NO”, that is, if the target pixel is not in the painted area, it is checked whether or not the index i is equal to or greater than the sum of the variable ST1 and the variable RL1 (step S603). If the determination result in step S603 is “YES”, ST1 and RL1 are read from the list (step S604).
[0050]
Next, when the processing in step S602 or step S604 ends or the determination result in step S603 is “NO”, the value obtained by adding “1” to the index i is equal to or greater than the variable ST1 and the variable ST1. It is checked whether or not the sum is smaller than the sum of the variable RL1 (step S605). That is, it is checked whether or not the pixel on the right of the target pixel (hereinafter simply referred to as the right pixel) is in the painted area. If the determination result in step S605 is “YES”, that is, if the right pixel is in the filled region, the bit OR operation result of the variable n1 and the constant “100b” is stored in the variable n1 (step S606). On the other hand, in the case of “NO”, that is, when the right pixel is not in the filled region, it is checked whether or not the value obtained by adding “1” to the index i is equal to or greater than the sum of the variable ST1 and the variable RL1 (step S607). If the determination result in step S607 is “YES”, ST1 and RL1 are read from the list (step S608).
[0051]
Next, when the processing in step S606 or step S608 ends or the determination result in step S607 is “NO”, the index i is greater than or equal to the variable ST2, and the sum of the variable ST2 and the variable RL2 Whether it is smaller is checked (step S609). That is, it is checked whether a pixel below the target pixel (hereinafter simply referred to as a lower pixel) is in the painted area. If the determination result in step S609 is “YES”, that is, if the lower pixel is in the filled region, the bit OR operation result of the variable n1 and the constant “10b” is stored in the variable n1 (step S610). On the other hand, if “NO”, that is, if the lower pixel is not in the painted area, it is checked whether or not the index i is greater than or equal to the sum of the variable ST2 and the variable RL2 (step S611). If the determination result in step S611 is “YES”, ST2 and RL2 are read from the list (step S612).
[0052]
Next, when the processing in step S610 or step S612 ends or the determination result in step S611 is “NO”, the value obtained by adding “1” to the index i is equal to or greater than the variable ST2, and the variable It is checked whether it is smaller than the sum of ST2 and variable RL2 (step S613). That is, it is checked whether or not the lower right pixel of the target pixel (hereinafter simply referred to as the lower right pixel) is in the filled region. If the determination result in step S613 is “YES”, that is, if the lower right pixel is in the filled region, the bit OR operation result of the variable n1 and the constant “1b” is stored in the variable n1 (step S614). Then, the process ends. On the other hand, in the case of “NO”, that is, when the next pixel is not in the filled region, it is checked whether or not the value obtained by adding “1” to the index i is greater than or equal to the sum of the variable ST2 and the variable RL2 (step S615). If the determination result in step S615 is “YES”, ST2 and RL2 are read from the list (step S616), and the process ends. If the determination result in step S615 is “NO”, the process ends without doing anything. In the above processing, the relative positional relationship of the pixels is described with the first pixel of the line as the leftmost end and the first line as the uppermost line.
[0053]
Here, a specific example of the processing will be shown taking the Mincho character “one” as an example. FIG. 14 is a conceptual diagram for a rasterized image and a comparison in the process. FIG. 15 is a partially enlarged view of FIGS. 14 (a) and 14 (e). 14 and 15, the rectangle represents a pixel, and the meanings of the rectangles other than the white rectangle are as described on the right side. In FIG. 14, one character is developed in an area of 16 pixels in the main scanning direction and 16 pixels in the sub-scanning direction, the upper left corner of the area is the coordinate origin, and the relative positional relationship between the pixels described above is used. The same positional relationship is shown. In FIG. 15, broken line arrows are scan lines that are virtually set when creating an edge list representing binary image data having a resolution n times the printer resolution. The intersection of the scan lines in both directions is a virtual coordinate point TP (i, j) in FIG. FIG. 15 shows the case of n = 2 as described above, but it is obvious that the virtual coordinate point can be set similarly even if the value of n is another value.
[0054]
FIG. 14A is a conceptual diagram showing a filled area when the character “one” is ideally developed. In order to output to a printer having a specific resolution, the pixel including the filled area is displayed. The value must be determined according to some logic. For example, if it is determined that pixels including even a small area shown in FIG. 14A are printed and pixels not included are not printed, a printing result as shown in FIG. 14B is obtained. As can be seen from both figures, this pixel value decision logic cannot obtain a high-quality print result. On the other hand, as described in the above-mentioned prior art, an edge list that simply represents multi-valued image data of printer resolution from an edge list that represents binary image data having a resolution n times the printer resolution created from PDL. When the raster image is converted into raster and rasterized, the raster image shown in FIG. 14C is obtained. When this raster image is printed, the print result shown in FIG. 14D is obtained. In this case, in particular, the horizontal bar portion of the character “one” is printed with a line width thinner than the line width assumed, and in some cases, it may be interrupted or not printed.
[0055]
On the other hand, when rasterization is performed according to the above-described embodiment of the present invention, a raster image shown in FIG. 14E is obtained, and when this raster image is printed, a printing result shown in FIG. 14F is obtained. FIG. 15A and FIG. 15B are partially enlarged views of FIG. 14A and FIG. 14E, respectively. Hereinafter, an example of rasterization processing according to the present invention will be described with reference to FIGS. 14 (a), 14 (e), 14 (f), and 15. FIG. First, in the present embodiment, among input PDL, characters designated for multi-level expansion are converted from PDL to outline data and further to an edge list. According to the algorithm shown in FIGS. 4 to 7, it is determined whether or not each virtual coordinate point is in the painted area, and as a result, an edge list is created.
[0056]
For example, looking at the seventh line in FIG. 15A, the virtual coordinate point on the right side of the thirteenth pixel is filled in the upper main scan direction scan line (hereinafter simply referred to as the main scan scan line). In the area. This means that the determination result is “YES” when i = 27 and j = 14 in step S202 of FIG. Until this virtual coordinate point, the determination result in step S202 is not “YES”, and the determination result in step S202 is “NO” for virtual coordinate points on the same main scanning scan line after this virtual coordinate point. Therefore, only st = 27 and rl = 1 are registered in the list when the main scanning scan line is completed.
[0057]
On the other hand, in the lower main scanning scan line, the virtual coordinate points on the left and right sides of the 13th pixel and the left side of the 14th pixel are continuously in the painted area. Accordingly, when processing is performed in the same manner as in the upper side, st = 26 and rl = 3 are added to the list when the main scanning scan line is completed. Similarly, when processing is performed for the eighth line, st = 2 and rl = 26 are added to the list for the upper main scanning scan line, and nothing is added to the list for the lower main scanning scan line. For the lines other than those described above, since no virtual coordinate points are in the painting area, a list as shown in Table 4 is finally obtained.
[Table 4]

[0058]
The list shown in Table 4 is merely an example and is expressed in an easy-to-read form. The data format in the storage medium of the list is not determined in Table 4, and any main scanning scan is used. As long as it is known what st and rl are registered in the line, any form may be used. In Table 4, the main scan scan line and the EOL code in which only the EOL code is registered are omitted.
[0059]
Next, a multi-valued edge list is created based on the list shown in Table 4. A specific example of this creation will be described by taking the seventh and eighth lines in FIG. 15B as an example. First, in the seventh line, from the first pixel to the twelfth pixel, the determination results in step S601, step S605, step S609, and step S613 are all calculated by calculating n1 and n2 in FIG. “NO”, the determination results in steps S601 and S605 are both “YES” and the determination results in steps S609 and S613 are both “NO” in the calculation of n3, so that n1 = 0 and n2 = 0 , N3 = 12, and the determination results in step S305 and step S309 in FIG. 5 are both “NO”. Therefore, the variable n2 is converted by the LUT shown in Table 2, and the pixel value “0” is obtained. Therefore, nothing is registered in the multi-value edge list. At the time of the 13th pixel on the same line, n1 = 0, n2 = 7, n3 = 12, and at the time of the 14th pixel, n1 = 0, n2 = 2, and n3 = 12, respectively. Thus, the pixel value “192” and the pixel value “64” are obtained.
[0060]
For the first pixel and the fifteenth and subsequent pixels, n1 = 0, n2 = 0, and n3 = 0. Therefore, when the processing on the seventh line is completed, the multi-value edge list includes st = 13, pv = 192, rl = 1 and st = 14, pv = 64, rl = 1 are registered. Next, in the eighth line, from the first pixel to the twelfth pixel, n1 = 0, n2 = 12, and n3 = 0, and the determination result in step S305 of FIG. 5 is “NO”, step S309 and The determination result in step S310 is “YES”. Therefore, the variable n2 is converted by the LUT shown in Table 3, and the pixel value “192” is obtained. At the time of the 13th pixel on the same line, n1 = 7, n2 = 12, and n3 = 0, and at the time of the 14th pixel, nl = 2, n2 = 12, and n3 = 0, both of which are pixel values. “128” is obtained. For all the pixels other than the lines described above, the pixel value is “0” for any pixel, so that a list as shown in Table 5 is finally obtained.
[Table 5]

[0061]
The list shown in Table 5 is merely an example and is expressed in an easy-to-read form. The data format of the list storage medium is not limited to Table 5, and what type is on which line? Any form can be used as long as it is known whether st, pv, and rl are registered. In Table 5, the main scan scan line and the EOL code in which only the EOL code is registered are omitted. As described above, a multi-value edge list is obtained and output from the multi-value character processing unit 12 to the synthesis unit 18.
[0062]
Next, processing in the binary character processing unit 13 will be described. First, outline data corresponding to a character to be drawn is loaded to create a binary edge list. When drawing of all the characters in the character string specified by the TYPE command is completed, the binary edge list is output to the synthesis unit 18 and the process is terminated. If there is a character that is not drawn, the outline data Repeat after loading.
[0063]
Of the above-described processes, the loading of outline data is the same as that of the multi-value character processing unit 12, and the creation of the binary edge list from the character outline is the same as the creation of the high-resolution edge list in the multi-value character processing unit 12. Since only the setting of the resolution, that is, the setting of the virtual coordinate point is different, the others are the same, and detailed description thereof is omitted. Table 6 shows an edge list created from the character “one”.
[Table 6]

[0064]
Next, processing in the multi-value vector processing unit 14 will be described. FIG. 16 is a flowchart showing processing in the multi-value vector processing unit 14. First, the designated vector is drawn (step S700), a high resolution edge list is created (step S701), and the edge list is converted into a multi-valued edge list (step S702). When all the vectors have been drawn, the multi-value edge list is output to the synthesizing unit 18 and the process is terminated. If there is a vector that has not been drawn, step S700 and subsequent steps are repeated. Step S701 and the subsequent steps are the same as those of the multi-value character processing unit 12, and therefore only step S700 will be described here. First, the designated vector is drawn on the virtual coordinate plane according to the designated coordinate point information. Next, an area that is half the distance of the specified line width around the vector drawn on the virtual coordinate plane is set as a filled area. Then, similarly to the multi-value character processing unit 12, virtual coordinate points are set, a high-resolution edge list is created, and the edge list is converted into a multi-value edge list.
[0065]
Here, FIG. 17 is a conceptual diagram for explaining a specific example of processing in the multi-value vector processing unit 14. First, the designated vector is drawn like the center line of the hatching area shown in FIG. Next, an area that is half the distance of the specified line width from the center line is designated as a filled area as shown by a hatched area shown in FIG. At this time, since the multi-value character processing unit 12 is in the same state as when the outline data is loaded and the filled area is set, the same processing as that of the multi-value character processing unit 12 is performed after the above processing. Once the high-resolution edge list is acquired, the multi-value edge list is acquired. Table 7 shows a high-resolution edge list, and Table 8 shows a multi-valued edge list. FIG. 17C shows the result of printing the finally obtained multi-value edge list.
[Table 7]

[0066]
[Table 8]

[0067]
Next, processing in the binary vector processing unit 15 will be described. First, a specified figure is drawn to create a binary edge list. When drawing of all the figures is completed, the binary edge list is output to the combining unit 18 and the process is terminated. If there is a figure that has not been drawn, the process after drawing is repeated. Of the processes described above, the drawing of the figure is the same as that of the multi-value vector processing unit 14, and the creation of the binary edge list from the vector is the creation of the high-resolution edge list and the resolution in the multi-value character processing unit 12. In other words, only the setting of the virtual coordinate point is different and the others are the same. Table 9 shows an edge list created from the graphic shown in FIG.
[Table 9]

[0068]
Next, processing in the graphics processing unit 16 will be described. Here, the specified area is filled with the specified color and the specified fill method, but if the fill method is a method of uniformly filling the specified area, it will be explained in detail. Since there is no, there will be explained a method of not painting uniformly. In such a filling method, the pixel value at each coordinate position is determined as a function of the coordinate position on the virtual coordinate plane. As this function, for example, there is a function in which the pixel value monotonously decreases from the highest density to the lowest density in the main scanning direction within the rectangular area.
[0069]
Here, FIG. 18 is a conceptual diagram for explaining a method of not painting uniformly. FIG. 18A shows an area on the virtual coordinate plane of 14 pixels in the main scanning direction and one line in the sub-scanning direction, with the left end of the area in the main scanning direction being the highest density and the right end being the lowest density, and in the main scanning direction FIG. 18B shows an example of drawing with a function that monotonously decreases from left to right, FIG. 18B shows an example in which each pixel value is determined by setting virtual coordinate points at the same resolution as the printer resolution, and FIG. This is an example in which the area shown in FIG. 18B is printed out. Note that the upper numerical value in FIG. 18B indicates the value of the pixel immediately below the numerical value. Note that the example shown in FIG. 18 is a specific example, and it is clear that any method of filling a region can be handled by variously changing the function for determining the pixel value from the coordinate position. Here, if a function in which the pixel value is determined to be the same regardless of the coordinate position is used, it is possible to uniformly paint the specified area, so the method of uniformly painting is also included in the above-described painting method. It is. When the determination of the pixel value is completed, a multi-value edge list is created from the determined pixel greatness and output to the combining unit 18. Table 10 shows an edge list created from each pixel value in the area shown in FIG.
[Table 10]

[0070]
Next, processing in the image processing unit 17 will be described. Here, drawing is performed only in the designated area and the designated color space in the designated area according to the raster image data, but the designated resolution may not match the printer resolution. In such a case, since the virtual coordinate point set at the same resolution as the printer resolution does not match the position of each pixel of the raster image data, the value of each pixel of the raster image data near a certain coordinate point is used. Printout is enabled by calculating approximate pixel values.
[0071]
That is, printout is enabled by matching raster image data to the printer resolution by resolution conversion processing. A method for calculating the approximate pixel value at this time may be a known method. If the designated color space does not match the color space that can be printed out, or if the color represented by each pixel value of the raster image data cannot be reproduced by the printer, the pixel value of the raster image data is changed. The pixel value changing method at this time may be a known method. As described above, raster image data is converted into an image in a resolution and color space that can be printed out, and then a multi-value edge list is created from the converted image and output to the synthesis unit 18.
[0072]
Finally, processing in the synthesis unit 18 will be described. The synthesizing unit 18 receives the edge lists output from the multi-value character processing unit 12, the binary character processing unit 13, the multi-value vector processing unit 14, the binary vector processing unit 15, the graphics processing unit 16, and the raster image processing unit 17. The image is synthesized, converted into a raster image, and output to the outside through the output terminal A. This composition processing is an operation between edge lists, for example, concatenation / insertion / selection of lists, and the priority order follows the drawing order in the PDL input to the command separation unit 11. For example, when nothing is specified, the priority increases from the beginning to the end of the PDL.
[0073]
That is, the character or vector drawn later is overwritten on the character or vector drawn previously. At this time, the edge list of characters or vectors to be drawn later is forcibly inserted into the existing output edge list, and the corresponding portion of the existing output edge list is overwritten. If there is any designation, the designated drawing order is not followed. For example, when it is specified that the character or vector described at the top of the PDL is drawn with the highest priority, the edge list of the character or vector drawn later is combined with the existing output edge list. If there is no list data at the same pixel position, the lists are concatenated. If there is the same data, nothing is done.
[0074]
Next, a specific example of processing in the above-described combining unit 18 will be described. For example, it is assumed that the multi-value edge list shown in Table 5 is output from the multi-value character processing unit 12 and then the binary edge list shown in Table 11 is output from the binary vector processing unit 15.
[Table 11]

[0075]
In the PDL description, the description related to the multi-valued edge list shown in Table 5, that is, the description that multi-values the character “one” is expanded before the description related to the binary edge list shown in Table 11, that is, the image in the sub-scanning direction. It is assumed that it is ahead of the description for drawing a straight line having a search width “1”. Here, when there is no designation regarding the drawing order, a character or vector to be drawn later is given priority, so an edge list as shown in Table 12 is created. That is, for the sixth pixel, the seventh line, and the ninth line that have no data in the multi-valued edge list shown in Table 5, the data of the binary edge list shown in Table 11 is connected, and the eighth line Since there is data for the eighth pixel, the binary edge list data shown in Table 11 is inserted.
[Table 12]

[0076]
On the other hand, when the character “one” is designated to be subjected to multilevel expansion with the highest priority, even if there is data at the same pixel position, the designated character or vector is given priority. An edge list is created. That is, for the sixth pixel, the seventh line, and the ninth line that have no data in the multi-valued edge list shown in Table 5, the data of the binary edge list shown in Table 11 is connected, and the eighth line Although there is data for the eighth pixel, the data of the multi-valued edge list is given priority by designation, so that no list operation is performed.
[Table 13]

[0077]
As described above, the synthesizing unit 18 performs a list synthesizing process and finally converts it into a raster image. The conversion to the raster image only involves converting the edge list described in the run-length format into a raster image according to the description content, and thus detailed description thereof is omitted. In addition, in the synthesis unit 18, whether the multi-value edge list or the binary edge list is output from the multi-value character processing unit 12, the binary character processing unit 13, the multi-value vector processing unit 14, or the binary vector processing unit 15. And the result is output to the outside from the output terminal B for each pixel of the raster image as identification data. When creating the identification data, the processing result relating to the drawing order at the time of combining the lists is reflected.
[0078]
By the way, in the above-described list combining process, it is checked whether there is data in the list at the same pixel position. Therefore, it is possible to easily know the boundary between the images represented by the edge lists output from different processing units. . Therefore, processing for such a boundary, for example, when a black character is embedded in the graphics area, it is possible to easily perform a boundary processing that makes the registration shift inconspicuous. This boundary processing can be realized as follows.
[0079]
First, an image boundary is detected from an edge list output from any one of the multi-value character processing unit 12, the binary character processing unit 13, the multi-value vector processing unit 14, and the binary vector processing unit 15. This detection is performed by detecting the start and end points in the main scanning direction of a pixel group having a pixel value other than “0” when the edge list is converted into a raster image, that is, each list located on the same line in the edge list. The start point and end point of the data in the main scanning direction are checked, and if the end point of one list data and the start point of another list data are adjacent in the main scanning direction, the adjacent start point and end point of the two list data Can be removed from the detection target.
[0080]
Next, in the synthesis unit 18, the edge list output from the graphics processing unit 16 or the image processing unit 17, the multi-value character processing unit 12, the binary character processing unit 13, the multi-value vector processing unit 14, and the binary vector processing. When the edge list output from any of the units 15 is combined, for example, the list data in the edge list output from the graphics processing unit 16 or the image processing unit 17 is used to detect the boundary between the detected image and the start or end point. When the points coincide with each other, if the coincident point is the start point, the value is decremented. If the coincident point is the end point, the value is incremented. If boundary processing is performed in this way, registration deviation during printing can be made inconspicuous.
[0081]
In contrast to the processing described above, the list data is not changed at the time of the edge list synthesis, but is output from the output terminal B to the outside as identification data having different contents from the identification data indicating which processing unit has output that the boundary is output. You may make it output.
[0082]
As another boundary process, there is an overlay process of multi-valued characters and background. Below, the example of the said superimposition process is demonstrated with reference to FIG. Here, FIG. 19A is a conceptual diagram showing an example in which color components are ideally developed on the background as four components of YMCK, and each pixel of each color component has 256 gradations. FIGS. 19B and 19C are conceptual diagrams showing the K component and the C component, respectively, of an image obtained by developing a multivalued character in the image shown in FIG. 19A, and FIG. It is a conceptual diagram which shows the example which printed out the image shown to 19 (b) and (c). In FIG. 19, a rectangle represents a pixel, and a pixel value of a non-white rectangle is a pixel having a gradation, and follows the legend in the figure.
[0083]
Normally, when a character is overlaid on a background that is not white, as shown in FIG. 19C, the pixel at the same position as the colored pixel of the character among the background pixels is replaced with the character pixel. On the other hand, multi-valued characters are printed by a printing method that expands the resolution, such as pulse width modulation, and therefore, especially for pixels whose pixel values are halftone, as shown in FIG. , White gaps may occur between the color components. Therefore, when superimposing halftone pixels of a multivalued developed character on the background, the background pixel is not simply replaced with the pixel of the multivalued developed character. It is only necessary to store the pixel values of different color components and replace the background pixels with character pixels only for pixels of the same color component. FIG. 19E is a conceptual diagram showing an example of printout when the image shown in FIG. 19A is overlaid with characters and background by the method described above. The method for realizing the boundary processing described above is the same as the boundary processing that makes the registration shift during printing inconspicuous, and thus detailed description thereof is omitted. In the above description, the raster image is converted into an edge list by the raster image processing unit 17 and then combined with the edge list output from another processing unit. However, the raster image processing unit 17 It is also possible to extract only the data, combine the edge list output from the raster image processing unit 17 and rasterize it, and then perform the combining process with the raster data.
[0084]
C. Modified example
The embodiment of the present invention has been described above, but the embodiment of the present invention is not limited to this. For example, in the synthesis unit 18 of the above embodiment, the edge list output from a plurality of processing units is provided. The composition processing is performed before rasterizing the image, but the gist of the present invention does not depart from the present invention even if the composition processing is performed after rasterizing each edge list. In the following, the latter embodiment will be described as a modification of the present invention.
[0085]
In this modified example, only the synthesizing process in the synthesizing unit 18 is different from the above-described embodiment, and the processes in the command separating unit 11 and each processing unit are the same, and thus the description thereof is omitted. The synthesizing unit 18 first rasterizes the edge list output from each processing unit. A raster image obtained as a result of rasterization is written in a memory (not shown) in the synthesizing unit 18 in order to perform a synthesizing process. The raster image on the memory is written to the memory so that it can be output to the outside as it is. Specifically, writing is performed so that the positional relationship of the pixel to be written on the memory matches the positional relationship of the entire image.
[0086]
However, when writing, a last image obtained by rasterizing an edge list output from another processing unit may be written. Therefore, in this case, as in the above embodiment, the pixel to be written is overwritten on the existing pixel at the position to be written according to the drawing order in the PDL, or the existing pixel is not overwritten. Decide whether to save. This determination of whether or not to overwrite a pixel corresponds to the synthesis processing between the edge lists in the synthesis unit 18 of the above embodiment, but the synthesis logic does not change although there is a difference between the edge list and the raster image. Therefore, there is no difference between images output from the output terminal A to the outside.
[0087]
Further, the boundary processing between the character and the background in the combining unit 18 of the above embodiment can be performed between raster images in the same manner as the combining processing between edge lists. In the above embodiment, after detecting the boundary between the character and the background, by operating the contents of the edge list of the boundary portion, processing for preventing registration deviation at the time of printing or overlapping the multi-value expansion character and the background. Processing is performed to prevent white gaps around multi-valued characters when matching. On the other hand, in this modified example, after the boundary is detected, the above processing is performed when writing to the memory. Specifically, in order to prevent misregistration at the time of printing, before writing the pixel of the character at the boundary portion into the memory, the pixel value before or after the pixel in the main scanning direction The pixel value of the color component different from that of the character is set as the pixel value of the same color component of the pixel of the character to be written. Further, in order to prevent white space around the multi-valued character when the multi-value developed character and the background are overlapped, before the halftone pixel of the multi-value developed character is written in the memory, 1 in the main scanning direction of the pixel is detected. Of the previous or next pixel value, the pixel value of the color component different from the character is set as the pixel value of the same color component of the pixel of the character to be written.
[0088]
In either case, the processing of repeating the contents of the edge list in the above embodiment is merely replaced with the processing between raster images, and there is no difference in the content of the image finally output from the output terminal A. Note that the pixel value of the color component different from the character of the pixel to be written is the value of the same color component of the pixel immediately before or after the pixel in the main scanning direction. Which pixel to select is the same logic as the selection in the processing for preventing registration deviation during printing in the above embodiment.
[0089]
By the way, any of the above-described embodiments and modifications may be implemented in several different forms because they may be implemented between application software that outputs PDL and the print engine. For example, in a printer in which a print engine that performs printing and a print controller that performs control of the print engine, control of communication with the outside, and the like are housed in one housing, it can be realized as software that operates on the print controller. Further, even when the print engine and the print controller are separate casings, it can be realized as software operating on the print controller. Alternatively, it can be realized as software operating on a PC or WS on which application software for outputting PDL is installed. Furthermore, in order to increase the processing speed, part or all of the processing in the above embodiment may be realized by hardware, and the implementation of any embodiment does not depart from the spirit of the present invention.
[0090]
The embodiments of the image processing apparatus according to the present invention have been described above. However, the implementation of the image processing apparatus according to the present invention is not limited to the above-described contents, and the gist of the present invention is possible even with other implementations. As long as it does not deviate from the above, any implementation may be used.
[0091]
【The invention's effect】
As described above, according to the present invention, the character or line drawing part is extracted by the extracting means from the image information expressed in the page description language and input from the input means. The first intermediate code is converted into a first intermediate code having a binary gradation by one conversion means, and the first intermediate code has a multi-value gradation at a lower resolution than the first intermediate code by the second conversion means. The second intermediate code is converted into a second intermediate code, and the tag generation means determines the type of image that occupies the area for each predetermined area of the input image information based on the first intermediate code or the second intermediate code. Tag information including at least information indicating that it is a character or a line drawing part represented by the second intermediate code is generated, and image information described in a page description language is converted into a raster image It is possible to reduce the storage capacity required in that, to obtain the advantage that more can be obtained printout with high image quality.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating a configuration example of an image processing apparatus according to the present invention.
FIG. 2 is a conceptual diagram illustrating an example of PDL input to the image processing apparatus according to the present invention.
FIG. 3 is a flowchart showing an edge list generation process.
FIG. 4 is a flowchart showing details of binary edge list creation processing;
FIG. 5 is a flowchart showing details of binary edge list creation processing;
FIG. 6 is a flowchart showing details of binary edge list creation processing;
FIG. 7 is a flowchart showing details of binary edge list creation processing;
FIG. 8 is a flowchart showing details of the head line processing.
FIG. 9 is a flowchart showing details of the head line processing.
FIG. 10 is a flowchart for explaining details of final line processing;
FIG. 11 is a flowchart for explaining details of final line processing;
FIG. 12 is a flowchart for explaining an algorithm for calculating a variable n1.
FIG. 13 is a flowchart for explaining an algorithm for calculating a variable n1.
FIG. 14 is a conceptual diagram for a rasterized image and a control for comparison.
15 is a partially enlarged view of FIGS. 14 (a) and 14 (b). FIG.
16 is a flowchart showing processing in the multi-value vector processing unit 14. FIG.
FIG. 17 is a conceptual diagram for explaining a specific example of processing in the multi-value vector processing unit 14;
FIG. 18 is a conceptual diagram for explaining a method of not uniformly filling a designated area in the graphics processing unit 16;
FIG. 19 is a conceptual diagram for explaining a superimposition process of a multi-valued character and a background.
[Explanation of symbols]
11 Command separation unit (input means, extraction means)
12 Multi-valued character processing unit (first conversion means)
13 Binary character processing unit (first conversion means)
14 Multi-value vector processing unit (first conversion means)
15 Binary vector processing unit (first conversion means)
16 Graphics processing unit
17 Raster image processor
18 synthesis unit (second conversion means, tag generation means, synthesis means, boundary detection means)

Claims (8)

An input means for inputting image information described in a page description language;
Extraction means for extracting a character or line drawing part from the image information input by the input means;
First conversion means for converting the character or line drawing part extracted by the extraction means into a first intermediate code for raster development, which is a first intermediate code having a binary gradation;
Second conversion means for converting the first intermediate code converted by the first conversion means into a second intermediate code having a multi-level gradation at a lower resolution than the first intermediate code;
For each predetermined region of the image information input by the input unit based on the first intermediate code converted by the first conversion unit or the second intermediate code converted by the second conversion unit And tag generation means for generating tag information representing the type of image occupying the area,
The image processing apparatus according to claim 1, wherein the tag information generated by the tag generation unit includes at least information indicating a character or a line drawing part represented by the second intermediate code. The image processing apparatus according to claim 1, further comprising a synthesizing unit that synthesizes the second intermediate code and an intermediate code different from the second intermediate code. Boundary detection means for detecting a boundary between the character or line drawing part and a portion other than the character or line drawing part;
The image processing apparatus according to claim 2, wherein the tag information generated by the tag generation unit further includes information indicating a boundary detected by the boundary detection unit. Boundary detection means for detecting a boundary between the character or line drawing part and a portion other than the character or line drawing part;
The image processing apparatus according to claim 2, wherein the composition in the composition unit is made different depending on the detection result in the boundary detection unit. The composition unit according to claim 1, further comprising: a combining unit that combines an image obtained by raster development of the second intermediate code and an image obtained by raster development of an intermediate code different from the second intermediate code. Image processing device. Boundary detection means for detecting a boundary between the character or line drawing part and a portion other than the character or line drawing part;
The image processing apparatus according to claim 5, wherein the tag information generated by the tag generation unit further includes information indicating a boundary detected by the boundary detection unit. Boundary detection means for detecting a boundary between the character or line drawing part and a portion other than the character or line drawing part;
The image processing apparatus according to claim 5, wherein the synthesis in the synthesis unit is made different depending on a detection result in the boundary detection unit. The image processing apparatus according to any one of claims 1 to 7, wherein the tag information has different contents according to the number of gradation levels of characters or line drawing parts represented by the second intermediate code.

Images