JP4098481B2 - Information code reading method and apparatus - Google Patents

Description

【００７２】
情報コードＣ１〜Ｃ４がそれぞれ３ビットで構成されているとすれば、生成多項式は次の式で表される。
Ｆ（Ｘ）＝Ｘ³ ＋Ｘ＋１・・・（５）
【００７３】
誤りの数と位置と値とを算出する場合には、次のようになる。まず、情報コードＣ１のみに誤りが含まれている場合は、
α⁶ ・Ｓ１＝Ｓ２＝α⁶ ・Ｅ_C1・・・（６）となる。
なお、Ｅ_C1は誤りを含んだＣ１コードを示している。
また、情報コードＣ２のみに誤りが含まれている場合は、
α⁵ ・Ｓ１＝Ｓ２＝α⁵ ・Ｅ_C2・・（７）となる。
なお、Ｅ_C2は誤りを含んだＣ２コードを示している。
また、情報コードＣ３のみに誤りが含まれている場合は、
α⁴ ・Ｓ１＝Ｓ２＝α⁴ ・Ｅ_C3・・（８）となる。
なお、Ｅ_C3は誤りを含んだＣ３コードを示している。
また、情報コードＣ４のみに誤りが含まれている場合は、
α³ ・Ｓ１＝Ｓ２＝α³ ・Ｅ_C4・・（９）となる。
なお、Ｅ_C4は誤りを含んだＣ４コードを示している。
【００７４】
次に、情報コードの訂正について図１８を参照して説明する。まず、読取センサ７１によって読み取られた各コードはスタートビット検出回路８０にてスタートビットの位置が検出される。スタートビットから内部クロックによりサンプリングされた各コードは、セレクタ７９によって情報コード本体部６１と誤り検出訂正コード部６２とに分けられ、それぞれのバッファ８１，８２に一時的に格納される。シンドローム演算部８３では、各情報コード本体部６１と訂正コード部６２とからの読取データに基づきシンドロームＳ１，Ｓ２を算出する。ここで、各シンドロームＳ１，Ｓ２が「０」であれば、誤り無しであるので、何らの訂正処理も行われず、次の情報コード処理ブロックである情報読取コントローラ８６に情報コードが送られる。また、Ｓ１≠０、Ｓ２≠０である場合には、情報コード本体部を含むコード列に誤りが含まれているので、各シンドロームＳ１，Ｓ２の値により、誤り検出部８４において、誤りの値、ＲＯＭテーブルに基づき誤りの位置を演算する。誤り訂正部８５では、遅延部８７によって遅延された情報コードデータに対して誤りの位置をカウンタにセットし、カウント値が誤り位置に一致したところでゲートを開き、誤り値とのＥＯＲをとることで、誤り訂正を行う。
【００７５】
なお、誤り検出訂正コードを有する情報コードの読み取りは単独で行ってもよく、または先に説明した情報コードの読取方法と併用してもよい。
【００７６】
【発明の効果】
本発明によれば、情報コードを予備走査した後に本走査を行い、予備走査では、サンプリングデータの最大値及び最小値を求め、本走査では、前記最大値及び最小値に基づき決定した振幅範囲でサンプリングデータをメモリに記憶したから、メモリの容量を増大させることがなく、しかも精度のよいサンプリングデータを記憶することができる。
【００７７】
また、回転体の周面又は端面にスタートマークと複数個のデータマークとを円周方向に並べて情報マークを構成し、前記回転体の一方向への回転時に前記予備走査を行い、回転体の他方向への回転時に前記本走査を行うことにより効率よく情報マークを走査することができる。また、走査信号をＡ／Ｄ変換したサンプリングデータに対して、本走査では、前記決定した振幅範囲により、サンプリングデータのビット数を減らしてデータメモリに記憶することにより、データメモリの容量が少なくて済み、この分だけコストを下げることができる。
【００７８】
サンプリングデータから最大値及び最小値を抽出してスレッシュレベルを決定し、このスレッシュレベルにより、各マーク部分の長さ相当値を求め、最大長さ相当値を有するマークをスタートマークとして検出するから、各データマークの特定が容易に行える。
【００７９】
第１のスタートマークと第２のスタートマークとの間隔によりデータマークの最小長さを求め、マークのエッジ部分が存在する領域を決定し、このエッジ存在領域において、サンプリングデータがスレッシュレベルを横切るポイントと横切る方向とからデータマークのビットが「０」か又は「１」かを判定するから、各ビットの判定を確実に行うことができる。
【００８０】
また、データマークのビットが「０」か又は「１」かの判定ができないときに、読取エラーとし、この読取エラーのときに、前記スレッシュレベルを変更し、または前記エッジ存在領域を広げてデータマークを再判定することにより、読み取り精度を上げることができる。
【００８１】
また、読取エラーの時に許容ノイズの範囲を広げることにより、極大値及び極小値を再抽出し、これに基づきスレッシュレベルを変更することで、読み取り精度を上げることができる。特に、フォトセンサなどにゴミが付着して信号レベルが低下しても、これに容易に対処することができる。
【００８２】
情報コードは誤り検出訂正コード部を有し、この誤り検出訂正コード部から読み取り時の誤り率を求め、この誤り率に基づき前記読取センサのスレッシュレベルを変化させて前記情報コードを読み取るから、情報コードを安定的に読み取ることができる。
【図面の簡単な説明】
【図１】本発明の情報コード読取方法を実施したカラー感熱プリンタを示す概略図である。
【図２】情報コードの一例を示す正面図である。
【図３】データマークの記録位置を示す説明図である。
【図４】情報コードの読取処理を示すフローチャートである。
【図５】予備走査におけるサンプリングデータの種類を示すグラフである。
【図６】予備走査におけるサンプリングデータの一例を示すグラフである。
【図７】情報コードの解読処理を示すフローチャートである。
【図８】極大値決定処理を示すフローチャートである。
【図９】極小値決定処理を示すフローチャートである。
【図１０】極大値及び極小値を決定する処理の具体例を示す説明図である。
【図１１】情報コードとこれに対応するサンプリングデータとの一例を示す説明図である。
【図１２】スタートマークの検出処理の一例を示す説明図である。
【図１３】各ビットの識別処理の一例を示す説明図である。
【図１４】別の実施形態における情報コードの一例を示す拡大した正面図である。
【図１５】誤り検出訂正コードを有した情報コードの一例を示す拡大した正面図である。
【図１６】誤り検出訂正コードを有した情報コード読取装置を示す概略図である。
【図１７】同情報コード読取装置における処理手順を示すフローチャートである。
【図１８】誤り検出訂正コードを用いた誤り訂正を示す機能ブロック図である。
【符号の説明】
１０ 給紙マガジン
１１ ロール紙
１２ 記録紙
１３ 巻芯
１４，５１，６０ 情報コード
１６ 給紙ローラ
３０ フォトセンサ
６２ 誤り検出訂正コード部
７１ 読取センサ
７２ コンパレータ
ＤＭ データマーク
ＳＭ スタートマーク
ＢＡ データ記録エリア
ＳＭＡ スタートマーク記録エリア[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a method and apparatus for reading an information code, and more particularly to a method and apparatus for scanning an information code formed by arranging data marks in a scanning direction with a reading sensor and reading the information code based on the scanning signal.
[0002]
[Prior art]
Photo printers and thermal printers use roll paper obtained by winding long photographic paper or thermal recording paper in a roll shape, and for example, the paper is cut into a predetermined size after printing. Since these photographic paper and heat-sensitive recording paper are given photosensitivity and photofixability, they are stored in a roll paper feed magazine (hereinafter referred to as a paper feed magazine) having moisture retention and light shielding properties. Is set in the printer.
[0003]
This paper feed magazine is composed of a magazine main body formed of a material having moisture retention and light shielding properties, and a lid that is opened when roll paper is loaded into the magazine main body. A paper feed roller is provided in the magazine body, and the outer peripheral surface of the roll paper is pressed against this peripheral surface. When the paper feed roller is rotated by driving from the printer side, the roll paper is rotated, and the recording paper is sent out from a discharge port provided in the magazine body to a recording position in the printer.
[0004]
Incidentally, the recording density of the recording paper changes depending on the sensitivity. Also, the images to be printed are different between normal recording paper and recording paper for sticker printing. In view of this, for example, in heat-sensitive recording paper, an information code indicating information such as the type of recording paper, thermal sensitivity, and manufacturer name is provided on the end surface of the roll paper core. Further, in the thermal printer, this information code is read by a code reading device, and print conditions are set based on this information.
[0005]
The information code is composed of a start mark and a number of data marks. The start mark is set, for example, to be three times as long as the data mark, and the start mark of the data mark is specified by this start mark. The data mark is set to be half the length of the recording area of each bit, and “0” is set depending on whether the mark is recorded in the first half or the second half of the recording area. It is identified as “1”.
[0006]
[Problems to be solved by the invention]
The information code is read by scanning in the arrangement direction of each mark when the winding core is rotated by feeding the recording paper by a reading sensor that optically detects the mark. However, reading errors may occur due to increased printing noise or deterioration of the amplitude of the scanning signal due to poor printing of the information code on the core end face or the like, and dirt or dirt on the display surface of the information code. May occur.
[0007]
The present invention has been made to solve the above-described problems, and an information code reading method and apparatus capable of reliably reading an information code while suppressing occurrence of a reading error due to an increase in noise, amplitude deterioration of a scanning signal, or the like. The purpose is to provide.
[0010]
[Means for Solving the Problems]
  In order to achieve the above object, in the present invention,A start mark and a plurality of data marks are arranged in the circumferential direction on the circumferential surface or end face of the rotator, and the start mark is an information code formed longer in the circumferential direction than the data mark. In a method of reading an information code, which is sometimes scanned by a reading sensor and identifies each bit of the information code based on an output signal from the reading sensor to read information, the information code is read by the reading sensor by rotating the rotating body a plurality of times. A data sampling step of scanning, A / D converting the output signal and storing it in a data memory as sampling data, and extracting a maximum value and a minimum value from the sampling dataCalculating the allowable noise from the maximum value and the minimum value, and when the sampling data fluctuates exceeding the allowable noise, the sampling data at the changing point is determined as a maximum value or a minimum value, and the calculated maximum value and The individual threshold level corresponding to each mark is determined based on the adjacent local maximum and local minimum values.A threshold level determining step, a length of each mark portion is determined based on the determined threshold level, a mark having the longest length is detected as a start mark, a first start mark and a second mark Find the minimum length of the data mark by the distance from the start markEachData determination for determining an area where an edge portion of a mark exists, and determining whether the bit of the data mark is “0” or “1” from the point where the sampling data crosses the threshold level and the direction in which the sampling data crosses the threshold level And a process.
[0011]
  In addition,In the threshold level determination step, ½ of the difference between the maximum value and the minimum value is set to the threshold level of each mark.Has been decided.
[0012]
  In the present inventionIn the data determining step, when the edge existing area is lowered and intersects the threshold level, the bit of the data mark is determined to be “0” or “1”, and the edge existing area is raised and lowered to intersect the threshold level. Sometimes the bit of the data mark is determined to be “1” or “0” which is the opposite of the above determination, and at other times it is determined as a reading error.
[0013]
  In the present invention,A data re-determination step of re-determining a data mark based on the threshold level obtained from the maximum value and the minimum value when the reading error is determined, and obtaining the maximum value and the minimum value again by changing the width of the allowable noise. have. Note that when it is determined that the reading error has occurred, the data level may be determined again by simply changing the threshold level. Further, when it is determined that the reading error has occurred, the edge existence area may be widened and the data mark may be determined again.
[0014]
  In the present invention,In the threshold level determination step, the threshold level is determined by alternately performing a maximum value detection process and a minimum value detection process. In the local maximum value detection process, the current data of the sampling data is set as the local maximum value, the next sampling data is compared with the local maximum value, and the local maximum value is updated when the sampling data is larger than the local maximum value in this comparison. When the sampling data is smaller than the maximum value, the difference obtained by subtracting the sampling data from the maximum value is compared with the allowable noise, and the maximum value is maintained without being updated when the difference is smaller than the allowable noise. When the difference is greater than the allowable noise, the current local maximum is determined as the local maximum at this bit. Further, after determining the maximum value, the process proceeds to minimum value detection. In the local minimum value detection process, the data at the point where the local maximum value is detected is set as the local minimum value, and then the local minimum value is compared with the data at this point. In this comparison, when the sampling data is smaller than the local minimum value, the local minimum value is detected. The value is updated, and when the sampling data is larger than the minimum value, the difference obtained by subtracting the minimum value from the sampling data is compared with the allowable noise, and the minimum value is not updated when the difference is smaller than the allowable noise. And the current local minimum is determined as the local minimum at this bit when the difference is greater than the allowable noise. Next, after determining the minimum value, the data at this point is set as the maximum value, and the process proceeds to the maximum value detection process.
[0015]
In the sampling step, preliminary scanning and main scanning are performed. In the preliminary scanning, the maximum value and the minimum value of the sampling data are obtained. In the main scanning, the sampling data is measured in the amplitude range determined based on the maximum value and the minimum value. Is preferably stored in memory. In the preliminary scanning, it is preferable to rotate the rotating body in one direction, and in the main scanning, rotate the rotating body in the other direction to obtain a scanning signal. Further, with respect to the sampling data obtained by A / D converting the scanning signal, in the main scanning, it is preferable that the number of bits of the sampling data is reduced and stored in the data memory according to the determined amplitude range.
[0016]
In addition, the rotating body is constituted by a winding core obtained by winding the recording paper into a roll shape, the information code is used as identification information of the recording paper, the recording paper is sent in the paper feeding direction, and the preliminary scanning is performed. Is preferably sent in the rewind direction opposite to the paper feed direction. Each mark of the information code is preferably formed of a mark material having a reflectance or absorptivity different from that of the ground on which the mark is formed. Further, the information code preferably includes at least one of information on the type of recording paper, heat sensitivity, lot number, and production time. The information code has an error detection / correction code part, an error rate at the time of reading is obtained from the error detection / correction code part, and the information code is read by changing a threshold level of the reading sensor based on the error rate. preferable.
[0018]
  The present invention provides a rotating bodyA start mark and a plurality of data marks are arranged in the circumferential direction on the circumferential surface or end surface of the, and the start mark is an information code formed longer in the circumferential direction than the data mark,The rotating bodyIn an information code reading device that scans with a reading sensor at the time of rotation, identifies each bit of the information code based on an output signal from the reading sensor, and reads recording sheet identification information, the information code is used to rotate the rotating body. A reading sensor that scans at the same time, a sampling circuit that samples the output signal from the reading sensor by A / D conversion, data writing means for writing the sampling data into a data memory, and sampling data stored in the data memory Information decoding means for decoding the information of the information code based on the information code, wherein the information decoding means extracts the maximum value and the minimum value from the sampling dataCalculating the allowable noise from the maximum value and the minimum value, and when the sampling data fluctuates exceeding the allowable noise, the sampling data at the changing point is determined as a maximum value or a minimum value, and the calculated maximum value and The individual threshold level corresponding to each mark is determined based on the adjacent local maximum and local minimum values.A threshold level determination unit that determines a length equivalent value of each mark portion based on the threshold level, detects a mark having a maximum length equivalent value as a start mark, and the first and second The minimum length of the data mark is obtained from the start mark interval of the first and second portions, and an edge existing region determining unit that determines a region where the edge portion of the mark exists, and a point where the sampling data crosses the threshold level in the edge existing region. It is characterized by comprising a data determination unit for determining whether the bit of the data mark is “0” or “1” based on the direction.
[0019]
The winding core is rotated in a paper feeding direction and a rewinding direction opposite to the paper feeding direction, and the information code is preliminarily scanned during rotation in the paper feeding direction by the rotating means. A scanning control unit that performs main scanning at the time of rotation; in preliminary scanning, a recording amplitude range of sampling data in main scanning is determined based on the amplitude of sampling data; and in the data writing unit, bits of sampling data are determined based on the recording amplitude range It is preferable to reduce the number and write to the data memory. The information code has an error detection / correction code portion, and the sampling circuit obtains an error rate at the time of reading from the error detection / correction code portion, and changes a threshold level of the reading sensor based on the error rate. Is preferred.
[0020]
DETAILED DESCRIPTION OF THE INVENTION
In FIG. 1 showing an outline of a color thermal printer, a roll paper 11 is stored in a paper feed magazine 10. The roll paper 11 is configured by winding a strip-shaped color thermal recording paper 12 around a core 13. The core 13 is made of a paper or plastic cylinder, and an information code 14 is recorded on the end surface 13a thereof. In addition, paper holders 15 are fitted at both ends of the core 13. The paper holder 15 has a rotation shaft 15a, and the roll paper 11 is rotatably held in the paper feed magazine 10 through the rotation shaft 15a.
[0021]
The paper feed magazine 10 includes a magazine body 10a and a lid 10b. In the paper feed magazine 10, a paper feed roller 16 and a roll paper bearing 17 are provided. The paper feed roller 16 contacts the roll paper 11 and rotates in the direction of arrow A, thereby feeding the recording paper 12 from the paper feed magazine 10. Further, the recording paper 12 is wound into the paper feed magazine 10 by rotating in the direction of arrow B.
[0022]
The roll paper bearing 17 holds the rotary shaft 15a of the paper holder 15 rotatably. The roll paper bearing 17 is configured to be movable in the direction of the paper feed roller, and is configured such that the roll paper 11 always contacts the paper feed roller 16 even if the roll diameter decreases due to the use of the roll paper 11. Has been. The paper feed roller 16 is driven by a paper feed motor 20. The sheet feeding motor 20 is rotationally controlled by a system controller 22 via a driver 21.
[0023]
At the time of printing, the paper feeding motor 20 rotates in the paper feeding direction to send the recording paper 12 to the printing unit 23. As is well known, the print unit 23 performs three-color surface sequential recording with a thermal head. In order to prevent the recorded thermosensitive coloring layer from being colored by the applied thermal energy for the next layer's thermal coloring during the three-color surface sequential recording, an ultraviolet ray having a wavelength range peculiar to the thermal coloring layer. Etc., photofixing is performed. When the three-color surface sequential recording is completed, the cutter 25 is operated via the driver 24. The cutter 25 cuts the printed part of the recording paper 12 so as to separate it from the unprinted part. The paper discharge roller 26 sets the recording paper 12 at the cutting position. Then, the recording paper 12 a cut by the cutter 25 is discharged to the tray 27.
[0024]
Before printing is started, the information code 14 is read in order to detect the type of the roll paper 11 set in the paper feed magazine 10 and to set the printing conditions according to this type.
[0025]
As shown in FIG. 2, the information code 14 is formed in the circumferential direction on the end surface 13a of the core 13, and is formed of a start mark SM made up of a black rectangular print pattern and a number of data marks DM. Yes. The ground color of the end surface 13a of the core 13 is white. Since this embodiment uses a 24-bit information code, 24 data marks DM are provided. The data mark DM represents various types of information such as the type of recording paper 12, heat sensitivity, lot number, production time information, and the like. The information code 14 only needs to include one of recording paper type, thermal sensitivity, lot number, production time information, and the like.
[0026]
As shown in FIG. 3, the data mark DM is set to a length BL (= L1 / 2) which is half the length L1 of the recording area BA of each bit in the circumferential direction. Then, as shown in FIG. 3A, the data mark DM is recorded in the latter half part BA1 of the recording area BA to represent “0” information, and as shown in FIG. 3B, the first half part of the recording area BA. Information “1” is represented by recording a data mark DM in BA2. In the illustrated information code 14, a line is drawn at the boundary of each area to identify the recording area BA. However, this is for illustrative purposes, and the actual information code 14 does not have this line, and marks having a length of BL, 2BL, or 3BL are simply spaced at intervals of BL or 2BL in the circumferential direction. Has been placed.
[0027]
As shown in FIG. 2, the start mark SM has a length three times that of the data mark DM. The start mark recording area SMA is set to a length L2 (= 5 × BL) that is five times the data mark. The start mark SM is arranged with a space for one data mark DM on both sides of the start mark recording area SMA.
[0028]
In this embodiment, the diameter of the core is 50 mm, the start mark recording area SMA is 5 data marks long (BL × 5), and the data mark recording area DMA representing 24 bits is 48 data marks. Minute length (BL × 48) is used. For this reason, BL is about 3 mm. The width of the data mark DM is also about 3 mm. The length BL of the data mark DM is appropriately determined according to the diameter of the core 13 to which the information code 14 is attached and the number of bits of the information code 14. Further, the length L2 of the start mark recording area SMA, the length of the start mark SM, the interval with the data mark DM, and the like may be changed as appropriate.
[0029]
As shown in FIG. 1, the information code 14 is read by the photosensor 30. The photosensor 30 is a reflection type, and is configured by integrating a projector, a light receiver, and an amplifier. Then, the light projected by the projector and reflected by the information mark 14 is received by the light receiving sensor, and the voltage signal photoelectrically converted by the light receiving sensor is amplified by the amplifier and then sent to the information reading controller 31. As the roll diameter decreases due to the use of the roll paper 11, the roll paper bearing 17 shifts in the direction of the paper feed roller, so that the photosensor 14 cannot rotate with the roll paper bearing 17 so as to follow this movement. It is attached.
[0030]
The information reading controller 31 includes an A / D converter 32. The A / D converter 32 converts the scanning signal from the photosensor 30 into 10-bit digital data. The information reading controller 31 includes a microcomputer including a CPU 33, a ROM 34, a RAM 35, a bus 36, and the like, and decodes the information code 14 based on sampling data from the A / D converter 32. In the present embodiment, the sampling data is stored in the data memory 35a with 8 bits so that the capacity of the data memory 35a set in the RAM 35 does not increase. Then, the contents of each bit are decoded by analyzing the sampling data written in the data memory 35a. The decoded data of the information code 14 is sent to the system controller 22. The system controller 22 identifies the type, sensitivity, manufacturer, sales channel, etc. of the recording paper 12 based on this information code data. Then, print conditions are set based on the specific data. The system controller 22 is also composed of a microcomputer, and includes a keyboard 28 and a display 29 for inputting and setting various commands.
[0031]
FIG. 4 is a flowchart showing a processing procedure in the information reading controller 31. First, the paper feed motor 20 is rotated at a constant speed for a fixed time, and the winding core 13 is rotated in the paper feed direction via the paper feed roller 16, and a preliminary scan of the information code 14 is performed at the time of feeding in the paper feed direction. Do. At this time, the feeding amount of the recording paper 12 by the paper feeding roller 16 is 740 mm for the two rolls of the roll paper in an unused state (two full turns) and the reserve amount of 20 mm. After this preliminary scanning, the main scanning is performed with the paper feeding roller 16 reversed by a preliminary amount by two full turns. By rotating two full turns, sampling data in a state where the data mark DM is sandwiched between the two start marks SM can be reliably obtained, and the data can be easily decoded.
[0032]
In the preliminary scan, the maximum value PSVMax and the minimum value PSVMin are detected based on the sampling data.
[0033]
Next, using the maximum value PSVMax and the minimum value PSVMin, a range to be rounded to 8 bits is determined when writing 10-bit sampling data in the main scan to the data memory 35a. Accordingly, since the 10-bit sampling data is rounded to 8 bits and stored, the capacity of the data memory 35a does not increase.
[0034]
Note that if 10 bits are simply converted into 8 bits and stored, the data accuracy is lowered depending on the range in which the sampling data is stored, and the information code cannot be reliably decoded. For this reason, the amplitude ranges stored based on the amplitude range of the sampling data specified by the maximum value PSVMax and the minimum value PSVMin are classified into three types as shown in FIG.
[0035]
FIG. 6 shows an example of sampling data. When determining the storage range of the sampling data, as shown in FIG. 4, the amplitude of the sampling data at the time of preliminary scanning is obtained from the maximum value PSVMax, the minimum value PSVMin, and the offset voltage value OSV. Classify into: The offset voltage value OSV is stored in the EEPROM.
[0036]
First, a determination value H is obtained by adding the offset voltage value OSV twice to the difference obtained by subtracting the minimum value PSVMin from the maximum value PSVMax (PSVMax−PSVMin).
H = (PSVMax−PSVMin) + OSV × 2
Next, it is determined whether the determination value H exceeds “512”, is within the range of “256” to “512”, or is “256” or less. FIG. 5 shows an example of sampling data classified in this way.
[0037]
When 512 <H, for example, the sampling data SD1 is as shown in FIG. 5A. In this case, the lower 2 bits of the sampling data in the 10-bit main scan are deleted to obtain 8 bits. Store in the data memory 35a.
[0038]
When 256 ≦ H ≦ 512, for example, the sampling data SD2 is as shown in FIG. 5B. In this case, the upper 1 bit and the lower 1 bit are deleted, and the stored data (8 bits) is read. Data (10 bits) − (PSVMin−OSV)] / 2.
[0039]
When H <256, for example, sampling data SD3 as shown in FIG. 5C. In this case, the upper 2 bits are deleted, and the stored data (8 bits) is replaced with [read data (10 bits) − (PSVMin-OSV)] / 4.
[0040]
Thus, by rounding the sampling data to 8 bits and storing it in the data memory 35a, the capacity of the data memory 35a is not increased, and information can be decoded from the sampling data with high accuracy.
[0041]
Next, a process for decoding the contents of the information code from the sampling data stored in the data memory 35a will be described with reference to the flowcharts of FIGS. FIG. 7 shows the entire decoding process of the information code. First, the maximum value FSVMax and the minimum value FSVMin are extracted from the sampling data. Next, the allowable noise KN is calculated from the following equation using the maximum value FSVMax and the minimum value FSVMin.
KN = [(FSVMax−FSVMin) × 0.2]
“0.2” is the noise coefficient α.
[0042]
Next, as shown in FIGS. 8 and 9, the maximum value VMax and the minimum value VMin due to the marks SM and DM are detected from the sampling data stored in the data memory 35a using the obtained allowable noise KN.
[0043]
The detection processing of the maximum value VMax and the minimum value VMin is performed as follows. The sampling data SD as shown in FIG. 10 will be described as an example. First, as shown in FIG. 8, in the process of detecting the maximum value VMax, the current sampling data V (1) at the point (1) is set as the maximum value VMax. Next, the sampling data V (n) at the point (n) is compared with the maximum value VMax. When the next sampling data V (n) is larger than the maximum value VMax in this comparison, this value is determined as the maximum value. And the maximum value VMax (= V (n)) is updated (see points (3) to (5)). Further, when the sampling data V (n) is smaller than the maximum value VMax at the point (n), the difference [VMax−V (n)] obtained by subtracting the sampling data V (n) from the maximum value VMax and the allowable noise KN When the difference [VMax−V (n)] is smaller than the allowable noise KN, the maximum value is maintained without being updated (see point (6)). When the difference [VMax−V (n)] is larger than the allowable noise KN, the current maximum value VMax is determined as the maximum value VMax in this bit (see point (7)).
[0044]
Next, after determining the maximum value VMax, the process proceeds to detection of the minimum value VMin. As shown in FIG. 9, in the local minimum value detection process, the data V (n) at the point where the local maximum value is detected (see point (7)) is set as the local minimum value VMin, and then the local minimum value VMin and the data at this point are displayed. Compare V (n). In this comparison, the minimum value is updated when the sampling data V (n) is smaller than the minimum value VMin (see points (9) and (10)). When the sampling data V (n) is larger than the minimum value VMin, the difference [V (n) −VMin] obtained by subtracting the minimum value VMin from the sampling data V (n) is compared with the allowable noise KNV, When the difference is smaller than the allowable noise KNV, the minimum value VMin is maintained without being updated (see point (8)), and when the difference is larger than the allowable noise KNV, the current minimum value VMin is set to the minimum value in this bit. Determine as VMin (see point (12)). Next, after determining the minimum value VMin, the next maximum value detection process is performed with the data at this point as the maximum value VMax. Then, after detecting the maximum value VMax, the minimum value VMin is detected in the same manner as described above, and thereafter, the maximum value VMax and the minimum value VMin are detected alternately.
[0045]
Next, an intermediate value between the obtained maximum value VMax (m) and the minimum value VMin (m) is obtained by the following equation, and this intermediate value is set as a threshold level TL (m).
TL (m) = [VMax (m) -VMin (m)] / 2
In addition, m is given in order for convenience in order to identify each local maximum value and local minimum value, and becomes an arbitrary numerical value of 1 to m. In this embodiment, one start mark and 24 data marks are used to form an information code, and the recording position of each data mark is recorded in the first half or the second half and either “0” or “1” is recorded. "0" and "1" are consecutive, the two data marks DM are connected to each other. Therefore, the number recognized as a mark is not necessarily “25” of the total number of marks, and is a number lower than this. In addition, since the three data marks DM are not connected, the start mark SM and the data mark DM having a length three times that of the data mark DM are not made indistinguishable.
[0046]
FIG. 11 shows threshold levels TL (1a), TL (1b), TL (2a), TL obtained from the maximum values VMax (1) to VMax (3) and the minimum values VMin (1), VMin (2). An example of (2b) is shown.
[0047]
Next, the sections UL (1), UL (2)... Below the threshold level TL are extracted. As shown in FIG. 12, the longest one of the extracted sections UL (1), UL (2)... Is determined as the longest section LMax, and the mark of the longest section LMax is determined as the start mark SM. To do. In addition, a section having the same length as the longest section LMax is obtained. First, the longest interval LMax is multiplied by “0.1” to obtain an allowable error LMax × 0.1. Then, an interval having a length UL within the allowable error range (0.9LMax <UL <LMax) is obtained from the intervals UL (1), UL (2)... Below the threshold level obtained above. The mark in the section UL is determined as the start mark SM. In FIG. 12, since the longest section LMax, UL (1) to qth section UL (q) corresponds to this, it is determined that this section UL (q) is also the start mark SM. In this embodiment, since the roll 13 is fully scanned for the length that the core 13 rotates twice, the obtained sampling data includes at least two samples as shown in FIG. A start mark SM is included.
[0048]
Next, the number of sampling points ST corresponding to the distance between each start mark SM is obtained from the positions of the two start marks SM. Next, from this ST, the number of sampling points BL corresponding to the length of one data mark is obtained from the following equation.
BL = ST / 53
“53” is obtained by adding 5 data marks as start marks and 48 data marks as 24-bit data marks.
[0049]
Next, as shown in FIG. 13, a threshold intersection detection area TLPA is obtained in order from bit 1 with reference to the first start mark SM1. When the intermediate point of the start mark SM1 is SC1, the start position PS1 and the end position PE1 of the threshold intersection detection area TLPA are obtained from the following equations.
PS1 = SC1 + BL × 3
PE1 = SC1 + BL × 4
The threshold intersection detection area TLPA is an area where an edge portion of each data mark DM exists.
[0050]
Next, when the sampling data V (PS1) in PS1 exceeds the threshold level TL (1a) [V (PS1)> TL (1a)], it is smaller than the threshold level TL (1a) between PS1 and PE1. A point P is detected. When this point (threshold intersection point) P can be detected, this bit 1 is determined to be “0”, and when it cannot be detected, it is determined as a reading error.
[0051]
Further, when the sampling data V at the start end PS1 of the threshold intersection detection area TLPA is lower than the threshold level TL, a threshold intersection P that is larger than the threshold level TL is detected in the threshold intersection detection area TLPA. When the threshold crossing point P can be detected, the bit 1 is determined as “1”, and when it cannot be detected, it is determined as a reading error.
[0052]
In the decoding of the data after bit 2, the threshold intersection detection area is specified based on the position of the threshold intersection P of the immediately preceding bit, and whether the bit is “0” or “1” is determined in the same manner as described above. That is,
PSn = P + BL × 1.5
PEn = P + BL × 2.5
In the threshold intersection detection area TLPA specified by (1), determination similar to bit 1 is performed to determine each bit.
[0053]
When a reading error occurs, the threshold level TL is changed and the determination is repeated again. The threshold level TL is changed by raising the noise coefficient α for obtaining the allowable noise KN from “0.2” to “0.3” or lowering to “0.1”. By increasing the noise coefficient α, the number of detected local maximum points and local minimum points decreases, so that the determination accuracy of each bit is improved without determining extra local maximum points or local minimum points as local maximum values or local minimum values. . Even when dust adheres to the photosensor 30, each bit can be accurately determined by changing the noise coefficient. The noise coefficient α is not limited to “0.1”, “0.2”, and “0.3”, and may be changed as appropriate. Further, the number of changes of the noise coefficient α is not limited to one, and may be performed a plurality of times. Further, instead of changing the allowable noise KN, the threshold level TL may be simply corrected by multiplying by a coefficient, and the determination may be made again based on this.
[0054]
Even when re-determination is performed with the allowable noise KN or the threshold level TL changed, if a reading error occurs, the threshold intersection detection area TLPA is expanded and the determination is performed again. For example, in bit 1, the start position PS1 and end position PE1 are obtained from the following equations.
PS1 = SC1 + BL × 2.9
PE1 = SC1 + BL × 4.1
If bit 2 or later,
PSn = P + BL × 1.4
PEn = P + BL × 2.6
As shown, the threshold intersection detection area is expanded by about 10%. The area expansion rate is not limited to 10%, and may be changed as appropriate.
[0055]
When a read error occurs, the allowable noise KN and the threshold level TL are changed and re-determination is performed. When a read error occurs even in this re-determination, the threshold intersection detection area TLPA is changed. The coping methods and order are not limited to those described above, and may be changed as appropriate. For example, the threshold intersection detection area TLPA may be changed first, and then the allowable noise KN and the threshold level TL may be changed.
[0056]
As described above, when the determination of “0” or “1” for each bit is completed up to bit 24, the code decoding process is terminated. If a read error occurs even after re-decoding as described above, as shown in FIG. 7, the preliminary scan and the main scan are performed again, and sampling data is newly collected. Reads correctly. Even if rescanning / re-determination is performed in this manner, if a reading error occurs again, an alarm is issued to inform the operator that the reading error has occurred, and then the reading of the information code is terminated.
[0057]
In this embodiment, the feeding speed of the recording paper 12 by the paper feed roller 16 is 100 mm / sec, and the sampling interval is 2 msec. Further, the feed length of each scan of the recording paper 12 is 704 mm for two full windings + 20 mm for the spare. The number of samplings at this time is 3620, and 1 byte per sampling is stored in the data memory 35a. Therefore, the required memory number of the data memory 35a is 3620 bytes. When the recording paper 12 is full, the information code 14 is sampled at intervals of about 0.1 mm in terms of recording paper feed amount. When the recording paper is small, the information code 14 is about 0.2 mm. Sampled at
[0058]
In the above embodiment, the information code 14 is read for each printing operation. Alternatively, the information code 14 may be read for each replacement of the paper feed magazine 10. In the above embodiment, when a reading error occurs, the threshold level or the threshold intersection detection area is changed each time. Instead, after the determination of each data mark is completed, the reading error bit is changed. Then, the threshold level or the threshold intersection detection region may be changed to perform the determination again.
[0059]
In the above embodiment, the information code 14 is formed on the white background with a black print pattern. However, the information code 14 may be recorded by another method as long as the mark can be detected optically. Moreover, the combination of other color schemes is not limited to the combination of black and white. For example, as shown in FIG. 14, a start mark recorded using an ink (for example, transparent or white) that absorbs light from the photosensor on a ground (for example, white) 50 that reflects light from the photosensor. The information code 51 may be formed by the SM and the data mark DM. In this case, an infrared type photosensor is used. The ground color and ink color are not limited to those described above, and may be changed as appropriate. Further, instead of the printing pattern, for example, the information mark may be configured using a hole, a reflecting member, or the like. Each mark may have various shapes such as a rectangle, a circle, a band, and a sector. Further, the recording form of the information code 14 is not limited to the above-described one, and the present invention may be implemented for information codes of various recording forms. For example, the present invention may be implemented to identify each bar of a bar code in which the bars are arranged in the circumferential direction.
[0060]
In the above embodiment, by rotating the paper feed roller 16 by a certain amount, the winding core 13 is rotated slightly more than two rounds. However, if this rotation is two or more rounds, it is three or more rounds. May be.
[0061]
In the above embodiment, the preliminary scanning is performed by rotation in the paper feeding direction, and the main scanning is performed by rotation in the rewinding direction opposite to the paper feeding direction. However, this rotational direction is not particularly limited. For example, in the type in which the winding core 13 is not rotated by pulling out the recording paper 12 by the paper feed roller 16 but the winding core is directly rotated, the preliminary scanning is performed by first rotating the winding core in the rewinding direction. The main scanning may be performed by rotating in the paper feeding direction. Further, the preliminary scanning and the main scanning may be rotated in the same direction without changing the rotation direction, the preliminary scanning may be performed by the first rotation, and the main scanning may be performed by the rotation in the same direction after the preliminary scanning is completed.
[0062]
In the above embodiment, the information code 14 is recorded on the end surface 13a of the core 13, but in addition, it may be recorded on the peripheral surface. Further, the present invention may be implemented for reading an information code provided on the back surface of the recording paper 12 in the feeding direction of the recording paper 12.
[0063]
In the above embodiment, the present invention is applied to the thermal recording paper for the thermal printer. However, the present invention may be applied to various recording papers of other recording systems, such as photographic paper rolls of a photographic printer. Further, the present invention may be implemented not only when recording paper but also when reading an information code attached to a rotating body.
[0064]
The information code may include an error detection / correction code portion. In this case, the reading accuracy can be further improved by obtaining the error rate at the time of reading from the error detection and correction code portion, and reading the information code by changing the threshold level of the reading sensor based on the error rate. . Hereinafter, a reading method including this error detection and correction code portion will be described.
[0065]
FIG. 15 shows an example of an information code having an error detection / correction code portion. FIG. 15 shows an information code 60 in which each mark is arranged in the scanning direction. This information code 60 is formed on the outer peripheral surface or inner peripheral surface of the winding core. When the usage pattern is the end face of the winding core, they are arranged in the circumferential direction as in the above embodiment. The information code 60 is composed of an information code body portion (data mark) 61 and an error detection / correction code portion 62 provided continuously therewith. The information code main body 61 is made up to 32 bits from the start mark, for example, and the next 16 bits are used as the error detection / correction code part 62. Therefore, since the positions of the information code body 61 and the error detection / correction code 62 are determined in advance, it is possible to distinguish these by counting the number of marks from the start mark. As the error detection / correction code unit 62, for example, a Reed-Solomon code or the like is used.
[0066]
FIG. 16 shows a block diagram of the information code reader 70. A read signal from the reflection type read sensor 71 including the projector 71a and the light receiver 71b is sent to the comparator unit 72 where it is compared with a reference voltage signal (threshold level) Vref. When the read signal is larger than the reference voltage signal, the level “1” is sent to the A / D converter 73. When the read signal is smaller than the reference voltage signal, the level “0” is sent to the A / D converter 73. The A / D converter 73 sends “1” and “0” digital signals to the system controller 75 via the mechanical controller (mechanism control microcomputer) 74 in accordance with these levels.
[0067]
The system controller 75 calculates the number, position, and value of errors based on the error detection / correction code and the generator polynomial, and corrects the information code 60 based on these pieces of information. Error correction technology based on these error detection and correction codes and generator polynomials is well known in "Code Theory" (co-authored by Hiroshi Miyagawa, Yoshihiro Iwabuchi, Hideki Imai, publisher: Shoshodo), etc. Please refer. FIG. 17 shows the processing procedure at this time. First, in order to read the information code 60, the recording paper is pulled out. The information code is read by the reading sensor 71 in response to the withdrawal of the recording paper. After reading all, error detection and the number of errors are calculated using the error detection and correction code. When the number of errors is within a certain range, error correction is performed and the information code is formally captured. In this way, error correction is performed when the number of errors is within a certain range, the correction capability is not infinite, it is limited to a number within a certain range, and if this correction capability range is exceeded This is because the probability of erroneous correction increases. The read information code is sent to various processing routines, and switching of the LUT, change of the voltage applied to the thermal head, and the like are performed.
[0068]
If the number of errors exceeding a certain range is detected, the threshold level is changed depending on whether the flag is “down” or “up”. Further, it is determined whether or not the number of errors has decreased from the previous time. If the number has decreased, the process is repeated until the error falls within a certain range. That is, for example, when the flag is “down” and the number of errors is lower than the previous time, the threshold level is lowered by a certain amount via the D / A converter 76 and then the flag is set to “down”. To do. For example, when the flag is “up” and the number of errors is lower than the previous time, the threshold level is increased by a certain amount via the D / A converter 76 and then the flag is set to “up”. To do. If the number of errors does not fall below the previous time, it is determined whether or not the flag is “down”. If it is “down”, the flag is set to “down”, and if it is not “down”, the flag Set to “Up”. Thereafter, the information code is read again by returning to the tip of the information code.
[0069]
In this way, by lowering the threshold level of the comparator 72 by a certain amount via the D / A converter 76, it becomes possible to normally compare the output whose level has been lowered as a whole. Therefore, the number of detection errors of the information code is also reduced and enters the correctable range, and thereafter the information code reading process is normally performed. When the reading surface is separated from the reading sensor 71 due to the skew of the roll paper or the like, the skew may return under some conditions. In this case, since the sensor output increases, it is considered that the background noise increases and the comparator output is applied. Even in this case, the threshold level is raised by setting the flag to “up” and raising the D / A output, so that the information code can be read normally.
[0070]
Next, the contents of the error detection and correction code and the generator polynomial will be described. Consider an information code system composed of two check words P and Q attached to information codes C1, C2, C3 and C4. The check words P and Q are defined as solutions of the following simultaneous equations. In the case of 2 parity, error detection and erasure correction up to 2 words can be performed, and when no erasure is used, the correction capability is up to 1 word.
A + B + C + D + P + Q = 0 (1)
α⁶・ A + α^Five・ B + α^Four・ C + α^Three・ D + α²・ P + α ・ Q = 0 (2)
Note that “+” in the above formulas 1 and 2 and the following formulas 3 and 4 represents modulo-2 addition. A to D indicate codes in byte units of the information codes C1 to C4, and α to α.⁶Indicates a primitive root on the Galois field, and “′” such as A ′ and B ′ indicates an information code including an error.
[0071]
Therefore, syndromes S1 and S2 are as follows.

[0072]
If the information codes C1 to C4 are each composed of 3 bits, the generator polynomial is expressed by the following equation.
F (X) = X^Three+ X + 1 (5)
[0073]
When calculating the number, position and value of errors, it is as follows. First, if only the information code C1 contains an error,
α⁶・ S1 = S2 = α⁶・ E_C1(6)
E_C1Indicates a C1 code including an error.
In addition, if only the information code C2 contains an error,
α^Five・ S1 = S2 = α^Five・ E_C2(7)
E_C2Indicates a C2 code including an error.
Also, if only the information code C3 contains an error,
α^Four・ S1 = S2 = α^Four・ E_C3(8)
E_C3Indicates a C3 code including an error.
In addition, when only the information code C4 contains an error,
α^Three・ S1 = S2 = α^Three・ E_C4(9)
E_C4Indicates a C4 code including an error.
[0074]
Next, information code correction will be described with reference to FIG. First, the start bit position of each code read by the reading sensor 71 is detected by the start bit detection circuit 80. Each code sampled from the start bit by the internal clock is divided into an information code main body 61 and an error detection / correction code 62 by the selector 79 and temporarily stored in the respective buffers 81 and 82. The syndrome calculation unit 83 calculates syndromes S1 and S2 based on the read data from each information code main body 61 and the correction code unit 62. If each of the syndromes S1 and S2 is “0”, there is no error, so no correction processing is performed, and an information code is sent to the information reading controller 86, which is the next information code processing block. If S1 ≠ 0 and S2 ≠ 0, an error is included in the code string including the information code main body, so that the error detection unit 84 determines the error value based on the values of the syndromes S1 and S2. The error position is calculated based on the ROM table. The error correction unit 85 sets the error position for the information code data delayed by the delay unit 87 in the counter, opens the gate when the count value matches the error position, and takes EOR with the error value. Perform error correction.
[0075]
Note that the information code having the error detection and correction code may be read alone or in combination with the information code reading method described above.
[0076]
【The invention's effect】
According to the present invention, after the information code is pre-scanned, the main scan is performed. In the pre-scan, the maximum value and the minimum value of the sampling data are obtained. In the main scan, the amplitude is determined based on the maximum value and the minimum value. Since the sampling data is stored in the memory, it is possible to store the sampling data with high accuracy without increasing the capacity of the memory.
[0077]
Further, an information mark is formed by arranging a start mark and a plurality of data marks in a circumferential direction on a circumferential surface or an end surface of the rotating body, and the preliminary scanning is performed when the rotating body is rotated in one direction. By performing the main scanning at the time of rotation in the other direction, the information mark can be efficiently scanned. In contrast to the sampling data obtained by A / D converting the scanning signal, in the main scanning, the number of bits of the sampling data is reduced and stored in the data memory according to the determined amplitude range, thereby reducing the capacity of the data memory. That's it, you can reduce the cost by this amount.
[0078]
The maximum value and the minimum value are extracted from the sampling data to determine the threshold level, and by using this threshold level, the length equivalent value of each mark portion is obtained, and the mark having the maximum length equivalent value is detected as a start mark. Each data mark can be easily identified.
[0079]
The minimum length of the data mark is determined by the distance between the first start mark and the second start mark, the area where the edge portion of the mark exists is determined, and the sampling data crosses the threshold level in this edge existing area Since the bit of the data mark is determined to be “0” or “1” from the crossing direction, each bit can be determined reliably.
[0080]
In addition, when it is not possible to determine whether the bit of the data mark is “0” or “1”, a read error is generated, and at the time of this read error, the threshold level is changed, or the edge existing area is expanded and data is read. By re-determining the mark, the reading accuracy can be increased.
[0081]
Further, the reading accuracy can be improved by expanding the allowable noise range at the time of reading error, re-extracting the maximum value and the minimum value, and changing the threshold level based on this. In particular, even if dust adheres to a photo sensor or the like and the signal level is lowered, this can be easily dealt with.
[0082]
The information code has an error detection / correction code part, an error rate at the time of reading is obtained from the error detection / correction code part, and the information code is read by changing a threshold level of the reading sensor based on the error rate. The code can be read stably.
[Brief description of the drawings]
FIG. 1 is a schematic diagram showing a color thermal printer that implements an information code reading method of the present invention.
FIG. 2 is a front view showing an example of an information code.
FIG. 3 is an explanatory diagram showing recording positions of data marks.
FIG. 4 is a flowchart showing information code reading processing;
FIG. 5 is a graph showing types of sampling data in preliminary scanning.
FIG. 6 is a graph showing an example of sampling data in preliminary scanning.
FIG. 7 is a flowchart showing information code decoding processing;
FIG. 8 is a flowchart showing local maximum value determination processing;
FIG. 9 is a flowchart showing a minimum value determination process.
FIG. 10 is an explanatory diagram illustrating a specific example of processing for determining a maximum value and a minimum value.
FIG. 11 is an explanatory diagram showing an example of an information code and sampling data corresponding to the information code.
FIG. 12 is an explanatory diagram illustrating an example of start mark detection processing;
FIG. 13 is an explanatory diagram illustrating an example of identification processing of each bit.
FIG. 14 is an enlarged front view showing an example of an information code in another embodiment.
FIG. 15 is an enlarged front view showing an example of an information code having an error detection / correction code.
FIG. 16 is a schematic diagram showing an information code reader having an error detection and correction code.
FIG. 17 is a flowchart showing a processing procedure in the information code reader.
FIG. 18 is a functional block diagram showing error correction using an error detection / correction code.
[Explanation of symbols]
10 Paper feed magazine
11 Roll paper
12 Recording paper
13 core
14, 51, 60 Information code
16 Paper feed roller
30 Photosensor
62 Error detection and correction code part
71 Reading sensor
72 Comparator
DM data mark
SM start mark
BA data recording area
SMA start mark recording area

Claims (16)

A start mark and a plurality of data marks are arranged in the circumferential direction on the circumferential surface or end face of the rotator, and the start mark is an information code formed longer in the circumferential direction than the data mark. In a method of reading an information code, which is sometimes scanned by a reading sensor, and each bit of the information code is identified based on an output signal from the reading sensor to read information.
A data sampling step of rotating the rotating body a plurality of times, scanning the information code with a reading sensor, A / D converting the output signal and storing it in a data memory as sampling data;
The maximum value and the minimum value are extracted from the sampling data, the allowable noise is calculated from the maximum value and the minimum value, and when the sampling data fluctuates beyond the allowable noise, the sampling data at the changing point is the maximum value. Alternatively, a threshold level determining step for determining an individual threshold level corresponding to each mark based on a local maximum value and a local minimum value adjacent to each other from the local maximum value and the local minimum value obtained as a local minimum value ;
According to the determined threshold level, the length of each mark portion is obtained, and a start mark detection step of detecting a mark having the longest length as a start mark;
The minimum length of the data mark is obtained by the interval between the first start mark and the second start mark to determine the area where the edge portion of each mark exists, and the sampling data has a threshold level in this edge existing area. A data reading method comprising: a data determination step of determining whether a bit of a data mark is “0” or “1” from a crossing point and a crossing direction. 2. The data reading method according to claim 1 , wherein the threshold level determination step of each mark determines 1/2 of the difference between the maximum value and the minimum value as the threshold level of each mark. The data determination step determines the bit of the data mark as “0” or “1” when the sampling data descends the edge existing area and crosses the threshold level, and the sampling data moves up and down the edge existing area. and a determination of "1" or "0" in the contrary to the determination of the bit data mark when crossing the threshold level, according to claim 1 or and judging an error reading at other times 2. A method for reading the information code described in 2 . A data re-determination step of re-determining a data mark based on the threshold level obtained from the maximum value and the minimum value when the reading error is determined, and obtaining the maximum value and the minimum value again by changing the width of the allowable noise. The data reading method according to claim 3, further comprising : 5. The data reading method according to claim 3 , further comprising a data re-determination step of re-determining a data mark by changing a threshold level of each mark when the reading error is determined. 6. The data reading method according to claim 3 , further comprising a data re-determination step of re-determining a data mark by expanding the edge existence area when the reading error is determined. The threshold level determination step alternately performs local maximum value detection processing and local minimum value detection processing,
In the maximum value detection process,
The current data of the sampling data and the maximum value is compared with the next sampling data the maximum value, in this comparison, and updates the maximum value when the sampling data is greater than the maximum value, the sampling data than the maximum value When the difference is less than the allowable noise, the difference obtained by subtracting the sampling data from the maximum value is compared with the allowable noise, and when the difference is smaller than the allowable noise, the maximum value is maintained without being updated. If it is, determine the current local maximum as the local maximum at this bit,
After determining this local maximum, we move on to local minimum detection.
In the local minimum value detection process, the data at the point where the local maximum value is detected is set as the local minimum value, and then the local minimum value is compared with the data at this point. When the sampling data is larger than the minimum value, the difference between the sampling data minus the minimum value is compared with the allowable noise, and when the difference is smaller than the allowable noise, the minimum value is not updated. And when the difference is greater than the allowable noise, determine the current local minimum as the local minimum at this bit;
7. The information code reading method according to claim 1 , wherein after determining the minimum value, the data at this point is set as a maximum value, and the process proceeds to the maximum value detection process. The sampling step performs preliminary scanning and main scanning. In the preliminary scanning, the maximum value and the minimum value of the sampling data are obtained. In the main scanning, sampling data is obtained based on the amplitude range determined based on the maximum value and the minimum value. 8. The information code reading method according to claim 1, wherein the number of bits is reduced and the data is stored in a memory. 9. The information code reading method according to claim 8, wherein in the preliminary scanning, the rotating body is rotated in one direction, and in the main scanning, the rotating body is rotated in the other direction to obtain an output signal. The rotating body is constituted by a winding core obtained by winding a recording paper into a roll shape, the information code is used as identification information of the recording paper, the recording paper is sent in a paper feeding direction, and the preliminary scanning is performed, and the recording paper The information code reading method according to any one of claims 1 to 9, wherein the main scanning is performed in a rewinding direction opposite to the paper feeding direction. 11. The information code reading method according to claim 1, wherein each mark of the information code is formed of a mark material having a reflectivity or absorptivity different from that of a ground on which the mark is formed. . 12. The information code reading method according to claim 10 or 11 , wherein the information code includes at least one of recording paper type, heat sensitivity, lot number, and production time information. The information code has an error detection / correction code part, obtains an error rate at the time of reading from the error detection / correction code part, and reads the information code by changing a threshold level of the reading sensor based on the error rate. 13. The information code reading method according to claim 1, wherein the information code is read. Is composed of a start mark and a plurality of data marks on the peripheral surface or end surface of the rotary body are arranged in the circumferential direction, the information code start mark which is formed longer in the circumferential direction than the data marks of the rotating body In an information code reader that scans with a reading sensor during rotation, identifies each bit of the information code based on an output signal from the reading sensor, and reads recording paper identification information,
A reading sensor that scans the information code when the rotating body rotates;
A sampling circuit for A / D converting and sampling the output signal from the reading sensor;
Data writing means for writing the sampling data into a data memory;
Information decoding means for decoding information of the information code based on the sampling data stored in the data memory,
The information decoding means includes
Extracting maximum and minimum values from the sampled data, the maximum value and the minimum value and to calculate the allowable noise, maximum sampling data at change points of Tokiniko that the sampled data exceeds this allowable noise fluctuates A threshold level determining unit that determines an individual threshold level corresponding to each mark based on a local maximum value and a local minimum value that are adjacent to each other based on the local maximum value and the local minimum value obtained as a value or a local minimum value ;
A start mark detection unit for obtaining a length equivalent value of each mark portion based on the threshold level, and detecting a mark having a maximum length equivalent value as a start mark;
An edge presence region determining unit for determining a minimum length of a data mark according to an interval between the first and second start marks and determining a region where an edge portion of the mark exists;
In the edge presence region, the data determination unit is configured to determine whether the bit of the data mark is “0” or “1” from the point where the sampling data crosses the threshold level and the crossing direction. Information code reader. The information code has an error detection and correction code part, and the data writing means obtains an error rate at the time of reading from the error detection and correction code part, and changes a threshold level of the reading sensor based on the error rate. 15. The information code reader according to claim 14 . The rotating body is constituted by a winding core obtained by winding a recording paper in a roll shape, the information code is used as identification information of the recording paper, and the winding core is wound in the direction opposite to the paper feeding direction and the paper feeding direction. Rotating means for rotating in the returning direction, and a scanning control unit for performing preliminary scanning of the information code when rotating the winding core in the feeding direction and performing main scanning when rotating in the rewinding direction,
  In the preliminary scan, the recording amplitude range of the sampling data in the main scan is determined based on the amplitude of the sampling data, and the data writing unit reduces the number of bits of the sampling data based on the recording amplitude range and writes it in the data memory. 16. The information code reader according to claim 14 or 15.

Images