Deprecated: The each() function is deprecated. This message will be suppressed on further calls in /home/zhenxiangba/zhenxiangba.com/public_html/phproxy-improved-master/index.php on line 456
JP6926846B2 - How to sort citrus fruits with irregularities on the surface - Google Patents
[go: Go Back, main page]

JP6926846B2 - How to sort citrus fruits with irregularities on the surface - Google Patents

How to sort citrus fruits with irregularities on the surface Download PDF

Info

Publication number
JP6926846B2
JP6926846B2 JP2017168408A JP2017168408A JP6926846B2 JP 6926846 B2 JP6926846 B2 JP 6926846B2 JP 2017168408 A JP2017168408 A JP 2017168408A JP 2017168408 A JP2017168408 A JP 2017168408A JP 6926846 B2 JP6926846 B2 JP 6926846B2
Authority
JP
Japan
Prior art keywords
unevenness
fruit
face
threshold value
setting
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
JP2017168408A
Other languages
Japanese (ja)
Other versions
JP2019045292A (en
Inventor
天池 寛武
寛武 天池
誠 垣内
誠 垣内
Original Assignee
一般財団法人雑賀技術研究所
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 一般財団法人雑賀技術研究所 filed Critical 一般財団法人雑賀技術研究所
Priority to JP2017168408A priority Critical patent/JP6926846B2/en
Publication of JP2019045292A publication Critical patent/JP2019045292A/en
Application granted granted Critical
Publication of JP6926846B2 publication Critical patent/JP6926846B2/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Length Measuring Devices By Optical Means (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
  • Sorting Of Articles (AREA)

Description

本発明は、柑橘類の果実をコンベアで搬送しながら、表面に凹凸がある形状異常果を非接触でオンライン選別する選別方法に関する。 The present invention relates to a sorting method for non-contact online sorting of citrus fruits having irregular shapes on the surface while being conveyed by a conveyor.

柑橘類の果実において、果肉よりも果皮が成長して果肉から果皮が著しく分離し、表面に凹凸がある浮皮果になると、貯蔵又は輸送中に腐敗しやすくなり、正常果よりも数段早く腐敗してしまう。その上、浮皮果になると、味が淡泊になって品質が低下し、浮きが酷いものは加工用果実となるので収益性が著しく低下する。
また、果肉に対して果皮が浮いていない果実であっても、表面に浮皮果と同様の凹凸があるものもあり、このような表面に凹凸がある果実は、見た目が悪いため、浮皮果と同様の低い等級に分類される。
以上より、表面に凹凸がある形状異常果の選別を精度良く行う方法及び装置についてのニーズは非常に高い。
In citrus fruits, when the pericarp grows more than the flesh and the pericarp separates significantly from the flesh and becomes a floating fruit with an uneven surface, it easily rots during storage or transportation and rots several times faster than normal fruits. It ends up. In addition, when it comes to floating fruit, the taste becomes light and the quality deteriorates, and when it floats badly, it becomes a fruit for processing, so profitability is significantly reduced.
In addition, even if the fruit has no skin floating on the flesh, some fruits have the same unevenness as the floating skin fruit, and such a fruit with uneven surface looks bad, so it is called a floating skin fruit. It is classified into a similar low grade.
From the above, there is a great need for a method and an apparatus for accurately selecting abnormally shaped fruits having irregularities on the surface.

柑橘類の果実の選果を行う選果場で用いる選果機により前記果実をコンベアで搬送しながら全数検査を行って選別する場合、コンベア速度は、例えば30m/min〜60m/min程度である。したがって、搬送方向前後の果実間に設ける間隔についての要求仕様等により、選別に要求される処理速度は、例えば40ms/個〜60ms/個程度になる。
よって、柑橘類の果実のオンライン選別は高速に処理を行う必要がある。
When the fruits are 100% inspected and sorted by a fruit sorter used in a fruit sorting plant for sorting citrus fruits while being conveyed by a conveyor, the conveyor speed is, for example, about 30 m / min to 60 m / min. Therefore, the processing speed required for sorting is, for example, about 40 ms / piece to 60 ms / piece, depending on the required specifications for the interval provided between the fruits before and after the transport direction.
Therefore, online sorting of citrus fruits requires high-speed processing.

柑橘類の果実をコンベアで搬送しながら表面に凹凸がある形状異常果を非接触でオンライン選別する選別方法として、搬送ライン(22)に沿って測定対象物である果実(12)をコンベアで搬送しながら、果実(12)に対し照射部(24)から線状光(28)を照射し、果実(12)に照射された線状光(28)を撮像装置(26)で撮像し、撮像装置(26)により撮像された画像データを数値化し、演算装置で予め用意されている基準となる形状正常果の形状選別データとを比較することにより形状を判別するものがある(例えば、特許文献1の図20参照)。 As a sorting method for non-contact online sorting of abnormally shaped fruits having irregularities on the surface while transporting citrus fruits on a conveyor, the fruits (12) to be measured are transported on a conveyor along a transport line (22). While irradiating the fruit (12) with linear light (28) from the irradiation unit (24), the linear light (28) irradiated to the fruit (12) is imaged by the image pickup device (26), and the image pickup device is used. There is one that discriminates the shape by digitizing the image data captured by (26) and comparing it with the shape selection data of the standard shape normal fruit prepared in advance by the arithmetic unit (for example, Patent Document 1). See FIG. 20).

特許文献1の表面に凹凸がある柑橘類の果実の選別方法は、前記演算装置が、撮像装置(26)により撮像された画像のうち、果実(12)に照射された線状光部分のみを抽出し、前記画像から抽出された線状光部分からなる表面撮像線(32)を一定間隔ごとに区切り、前記表面撮像線(32)のそれぞれの傾斜角度を求めるとともに、前記各傾斜角度について隣接区間の傾斜角度の角度差データを得、前記角度差データと予め用意されている形状正常果の形状データの角度差データとを比較して、その差が所定の閾値から外れた場合に、果実(12)は表面に凹凸がある形状異常果であると判断している。 In the method for selecting citrus fruits having irregularities on the surface of Patent Document 1, the calculation device extracts only the linear light portion irradiated on the fruit (12) from the images captured by the image pickup device (26). Then, the surface imaging line (32) composed of the linear light portion extracted from the image is divided at regular intervals, the inclination angles of the surface imaging lines (32) are obtained, and the adjacent sections are obtained for each inclination angle. The angle difference data of the inclination angle of the above is obtained, the angle difference data is compared with the angle difference data of the shape data of the shape normal fruit prepared in advance, and when the difference deviates from a predetermined threshold, the fruit ( 12) is determined to be an abnormal shape fruit with irregularities on the surface.

また、柑橘類の果実をコンベアで搬送しながら所定の項目(例えば糖度、酸度、成熟度、傷、腐り、長さ、色、外径、形状、高さ、幅、体積、偏平度、腐り具合、浮き皮具合、規格外等の等階級)を検査及び判定する検査装置として、搬送ラインに沿って果実(A)をコンベア(2)で搬送しながら、検査部(b)の第1検査領域で果実(A)に光を照射して反射光と透過光をカメラ(32,33)で受光して、その受光データを判定装置(4a)に記憶し、回転装置(40)により果実(A)を上下逆向きとなる状態に回転し、検査部(b)の第2検査領域で果実(A)に光を照射して透過光をカメラ(34)で受光して、その受光データを判定装置(4a)に記憶し、判定装置(4a)により前記受光データと予め格納しておいた等階級データとを比較して前記所定の項目を検査及び判別するものがある(例えば、特許文献2参照)。 In addition, while transporting citrus fruits on a conveyor, certain items (for example, sugar content, acidity, maturity, scratches, rot, length, color, outer diameter, shape, height, width, volume, flatness, rot condition, etc. As an inspection device that inspects and judges the floating skin condition, nonstandard grade, etc.), while transporting the fruit (A) on the conveyor (2) along the transport line, in the first inspection area of the inspection unit (b). The fruit (A) is irradiated with light, the reflected light and the transmitted light are received by the camera (32, 33), the received data is stored in the determination device (4a), and the fruit (A) is stored by the rotating device (40). Is rotated upside down, the fruit (A) is irradiated with light in the second inspection area of the inspection unit (b), the transmitted light is received by the camera (34), and the received data is determined by the determination device. There is one that is stored in (4a) and inspected and discriminated from the predetermined item by comparing the received light data with the equiclass data stored in advance by the determination device (4a) (see, for example, Patent Document 2). ).

特開2006−162393号公報Japanese Unexamined Patent Publication No. 2006-162393 特開2001−097551号公報Japanese Unexamined Patent Publication No. 2001-097551

柑橘類の一例として、みかんについて、本願の発明者らは実験を行って鋭意検討を行った。その結果、表面に凹凸があるみかんの果実について、表側(ヘタ側)と裏側とで、以下のように表面性状の特徴に差異があることが分かった。
(表側の表面性状の特徴)
表側は、細かいしわのような凹凸(高周波状の凹凸)が多く、大きなくぼみ状の凹凸(低周波状の凹凸)は少ない。
(裏側の表面性状の特徴)
裏側は、細かいしわのような凹凸(高周波状の凹凸)は少なく、大きなくぼみ状の凹凸(低周波状の凹凸)が多い。
このような表面に凹凸があるみかんの果実における表面性状の特徴は、みかん以外の柑橘類の果実についても当てはまる。
As an example of citrus fruits, the inventors of the present application conducted experiments and diligently studied mandarin oranges. As a result, it was found that the surface texture of the mandarin orange fruit having irregularities on the surface was different between the front side (septal side) and the back side as follows.
(Characteristics of surface texture on the front side)
On the front side, there are many fine wrinkle-like irregularities (high-frequency irregularities) and few large dent-like irregularities (low-frequency irregularities).
(Characteristics of surface texture on the back side)
On the back side, there are few irregularities such as fine wrinkles (high frequency irregularities), and many large depressions (low frequency irregularities).
The surface texture characteristics of mandarin orange fruits having such uneven surfaces also apply to citrus fruits other than mandarin oranges.

特許文献1の表面に凹凸がある柑橘類の果実の選別方法におけるコンベアの詳細は不明であるが、選果機のコンベアで柑橘類の果実が搬送される際、表向き、裏向きのどちらになるかはランダムであるのが一般的である。
よって、果実の表側の表面性状の特徴に合わせて判別条件を整えた際において、コンベア上を果実が裏向きで搬送されてきた場合、及び果実の裏側の表面性状の特徴に合わせて判別条件を整えた際において、コンベア上を果実が裏向きで搬送されてきた場合には、誤判別する確率が高くなるので、表面に凹凸がある柑橘類の果実の選別精度が悪化する。
The details of the conveyor in the method for selecting citrus fruits having irregularities on the surface of Patent Document 1 are unknown, but when the citrus fruits are transported by the conveyor of the fruit sorter, it is determined whether the citrus fruits are face up or face down. It is generally random.
Therefore, when the discrimination conditions are adjusted according to the surface texture characteristics of the front side of the fruit, the discrimination conditions are set according to the case where the fruit is transported face down on the conveyor and the surface texture characteristics of the back side of the fruit. When the fruits are transported face down on the conveyor at the time of preparation, the probability of erroneous discrimination increases, so that the selection accuracy of citrus fruits having irregularities on the surface deteriorates.

特許文献2のような果実を上下逆向きとなる状態に回転できる回転装置を備えて、果実の表裏両面を撮影するようにすれば選別精度を高めることができる。
しかしながら、回転装置を備えることによりコンベア装置の構成が複雑になるので製造コスト及び保守コストが増大する。
また、果実の表側の表面性状の特徴に合わせて判別条件を整えた際において果実を表向きで搬送するように、あるいは果実の裏側の表面性状の特徴に合わせて判別条件を整えた際において果実を裏向きで搬送するように整列させれば、誤判別する確率が高くならないので、表面に凹凸がある柑橘類の果実の選別精度が悪化しない。
しかしながら、果実の表裏の判別を行って特許文献2の回転装置等の装置で果実を上下逆向きに反転させて整列させる場合、前記装置を備えることによりコンベア装置の構成が複雑になるので製造コスト及び保守コストが増大する。
Sorting accuracy can be improved by providing a rotating device capable of rotating the fruit upside down as in Patent Document 2 and photographing both the front and back sides of the fruit.
However, the provision of the rotating device complicates the configuration of the conveyor device, which increases the manufacturing cost and the maintenance cost.
In addition, when the discrimination conditions are adjusted according to the surface texture characteristics of the front side of the fruit, the fruit is transported face up, or when the discrimination conditions are adjusted according to the surface texture characteristics of the back side of the fruit, the fruit is subjected to. If the fruits are aligned so as to be transported face down, the probability of erroneous discrimination does not increase, so that the selection accuracy of citrus fruits having irregularities on the surface does not deteriorate.
However, when the front and back of the fruit are discriminated and the fruits are turned upside down and aligned by a device such as a rotating device of Patent Document 2, the provision of the device complicates the configuration of the conveyor device, so that the manufacturing cost And maintenance costs increase.

その上、特許文献1の表面に凹凸がある柑橘類の果実の選別方法のように、前記角度差データの偏差が一定の閾値から外れた場合に表面に凹凸がある形状異常果であると判定する方法では、現実の果実では、前記角度差データの偏差が1ヶ所だけ大きくても浮皮果等であることは少なく、1本の線の中にプラスの角度偏差とマイナスの角度偏差が何回か現れて凹凸を形成していることが多い。
よって、特許文献1の表面に凹凸がある柑橘類の果実の選別方法は、表面に凹凸がある形状異常果の選別をより精度良く行うという観点からは改良の余地がある。
Further, as in the method for selecting citrus fruits having irregularities on the surface of Patent Document 1, when the deviation of the angle difference data deviates from a certain threshold value, it is determined that the fruit has irregular shapes on the surface. In the method, in the actual fruit, even if the deviation of the angle difference data is large only in one place, it is rarely a floating fruit, etc., and there are several positive and negative angle deviations in one line. It often appears and forms irregularities.
Therefore, the method for selecting citrus fruits having irregularities on the surface of Patent Document 1 has room for improvement from the viewpoint of more accurately selecting irregularly shaped fruits having irregularities on the surface.

その上さらに、特許文献1のような、柑橘類に照射された線状光を撮像装置で受光し、撮像装置により受光された測定用光を数値化し、演算装置で予め用意されている形状正常果の形状データとを比較することにより形状を判別する方法では、線状光を画像処理で抽出する際、果実表面の凹凸が激しいと抽出線が複雑に分断されてしまい数値化に支障が出る場合がある。
よって、果実表面の凹凸が激しい場合には、表面に凹凸がある柑橘類の果実の選別精度が悪化する。
Furthermore, as in Patent Document 1, the linear light radiated to citrus fruits is received by the image pickup device, the measurement light received by the image pickup device is quantified, and the shape normal result prepared in advance by the arithmetic unit is obtained. In the method of discriminating the shape by comparing with the shape data of, when extracting linear light by image processing, if the surface of the fruit is severely uneven, the extracted line is divided in a complicated manner, which hinders quantification. There is.
Therefore, when the surface of the fruit is extremely uneven, the accuracy of sorting citrus fruits having an uneven surface deteriorates.

なお、ライン状のレーザ光源を果実に照射し、その反射光を高さデータとして取得して三次元形状を得る光切断法を用いることも考えられるが、前記のとおり高速に処理を行う必要がある果実のオンライン選別には不適であるとともに、製造コストが嵩んでしまう。 It is also conceivable to use a light cutting method in which a line-shaped laser light source is applied to the fruit and the reflected light is acquired as height data to obtain a three-dimensional shape, but it is necessary to perform the processing at high speed as described above. Not only is it unsuitable for online sorting of certain fruits, it also increases manufacturing costs.

そこで、本発明が前述の状況に鑑み、解決しようとするところは、コンベア装置の構成を複雑にすることなく、表面に凹凸がある形状異常果の選別精度を向上できる、表面に凹凸がある柑橘類の果実の選別方法を提供する点にある。 Therefore, what the present invention seeks to solve in view of the above-mentioned situation is that citrus fruits having irregularities on the surface can improve the sorting accuracy of irregularly shaped fruits having irregularities on the surface without complicating the configuration of the conveyor device. The point is to provide a method for selecting citrus fruits.

請求項1に係る発明の表面に凹凸がある柑橘類の果実の選別方法は、前記課題解決のために、柑橘類の果実をコンベアで搬送しながら全数検査を行い、表面に凹凸がある形状異常果を非接触でオンライン選別する選別方法であって、
前記コンベアは、前記果実を表向き又は裏向きに整列することなく、表裏がランダムである前記果実を一列に搬送するものであり、
前記コンベアの搬送経路の適宜位置に設けた光源から前記果実に線状光を照射し、前記光源と異なる角度で設置したカメラで前記果実を撮影する画像撮影工程と、
前記画像撮影工程で撮影した画像を画像処理した画像データから、前記果実に照射された前記線状光による線を抽出し、
前記線を表向き果実に対する実験により予め定めた等間隔で区切り、
区切られた前記間隔毎の点における前記線の角度を求め、
前記点の隣り合うものの角度差の絶対値が表向き果実に対する実験により予め定めた角度差閾値を超える回数を数えて表向き用設定の凹凸度Cfを算出し、
前記線を裏向き果実に対する実験により予め定めた等間隔で区切り、
区切られた前記間隔毎の点における前記線の角度を求め、
前記点の隣り合うものの角度差の絶対値が裏向き果実に対する実験により予め定めた角度差閾値を超える回数を数えて裏向き用設定の凹凸度Crを算出し、
前記裏向き用設定の凹凸度Crの分布が前記表向き用設定の凹凸度Cfの分布と同一尺度になるように実験により予め定めた変換式により、前記裏向き用設定の凹凸度Crを裏向き用設定の補正凹凸度Cr’に変換し、
前記表向き用設定の凹凸度Cf及び前記裏向き用設定の補正凹凸度Cr’の大きい方を凹凸度とする凹凸度算出工程、又は、
前記表向き用設定の凹凸度Cfの分布が前記裏向き用設定の凹凸度Crの分布と同一尺度になるように実験により予め定めた変換式により、前記表向き用設定の凹凸度Cfを表向き用設定の補正凹凸度Cf’に変換し、
前記表向き用設定の補正凹凸度Cf’及び前記裏向き設定の凹凸度Crの大きい方を凹凸度とする凹凸度算出工程と、
前記凹凸度算出工程で算出した前記凹凸度を実験により予め定めた凹凸度閾値と比較し、
前記凹凸度が前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、前記果実は前記形状異常果であると判断し、
前記凹凸度が前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、前記果実は形状正常果であると判断する形状判断工程と、
を含むことを特徴とする。
In the method for selecting citrus fruits having irregularities on the surface of the invention according to claim 1, in order to solve the above-mentioned problems, 100% of the citrus fruits are inspected while being transported by a conveyor, and irregularly shaped fruits having irregularities on the surface are found. It is a sorting method that sorts online without contact.
The conveyor transports the fruits, whose front and back are random, in a row without arranging the fruits face up or face down.
An imaging step of irradiating the fruit with linear light from a light source provided at an appropriate position on the conveyor's transport path and photographing the fruit with a camera installed at an angle different from the light source.
From the image data obtained by processing the image taken in the image taking step, a line due to the linear light applied to the fruit is extracted.
The lines are separated by experiments on ostensibly fruits at equal intervals.
Find the angle of the line at each of the separated points.
The degree of unevenness Cf set for ostensibly is calculated by counting the number of times that the absolute value of the angle difference between adjacent points exceeds the predetermined angle difference threshold value in the experiment on the ostensibly fruit.
Divide the line at equal intervals predetermined by experiments on face-down fruits.
Find the angle of the line at each of the separated points.
The degree of unevenness Cr of the face-down setting was calculated by counting the number of times that the absolute value of the angle difference of the adjacent points exceeded the angle difference threshold value predetermined by the experiment on the face-down fruit.
The unevenness Cr of the face-down setting is turned face down by a conversion formula predetermined by an experiment so that the distribution of the unevenness Cr of the face-down setting is on the same scale as the distribution of the unevenness Cf of the face-up setting. Converted to the correction unevenness Cr'of the setting for
The unevenness calculation step in which the larger of the unevenness Cf of the front-facing setting and the corrected unevenness Cr'of the back-facing setting is the unevenness, or
The unevenness Cf of the face-up setting is set to face-up by a conversion formula predetermined by an experiment so that the distribution of the unevenness Cf of the face-up setting becomes the same scale as the distribution of the unevenness Cr of the face-down setting. Converted to the correction unevenness Cf'of
The unevenness calculation step in which the larger of the correction unevenness Cf'of the front-facing setting and the unevenness Cr of the back-facing setting is the unevenness.
The unevenness calculated in the unevenness calculation step is compared with a predetermined unevenness threshold value by an experiment.
When the unevenness is equal to or higher than the unevenness threshold value or larger than the unevenness threshold value, it is determined that the fruit is the abnormal shape fruit.
When the unevenness is smaller than the unevenness threshold value or is equal to or less than the unevenness threshold value, the shape determination step of determining that the fruit is a normal shape fruit and
It is characterized by including.

また、請求項に係る発明の表面に凹凸がある柑橘類の果実の選別方法は、前記課題解決のために、柑橘類の果実をコンベアで搬送しながら全数検査を行い、表面に凹凸がある形状異常果を非接触でオンライン選別する選別方法であって、
前記コンベアは、前記果実を表向き又は裏向きに整列することなく、表裏がランダムである前記果実を一列に搬送するものであり、
前記コンベアの搬送経路の適宜位置で前記果実を撮影し、所定エリア内におけるヘタの有無を画像処理で識別して、ヘタが有る場合には表向き果実であると判断し、ヘタが無い場合には裏向き果実であると判断する表裏判断工程と、
前記コンベアの搬送経路の適宜位置に設けた光源から前記果実に線状光を照射し、前記光源と異なる角度で設置したカメラで前記果実を撮影する画像撮影工程と、
前記画像撮影工程で撮影した画像を画像処理した画像データから、前記果実に照射された前記線状光による線を抽出し、
前記表裏判断工程で、表向き果実であると判断した場合、
前記線を表向き果実に対する実験により予め定めた等間隔で区切り、
区切られた前記間隔毎の点における前記線の角度を求め、
前記点の隣り合うものの角度差の絶対値が表向き果実に対する実験により予め定めた角度差閾値を超える回数を数えて表向き用設定の凹凸度Cfを算出する凹凸度算出工程と、
前記凹凸度算出工程で算出した前記凹凸度Cfを実験により予め定めた凹凸度閾値と比較し、
前記凹凸度Cfが前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、前記果実は前記形状異常果であると判断し、
前記凹凸度Cfが前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、前記果実は形状正常果であると判断する形状判断工程と、
前記表裏判断工程で、裏向き果実であると判断した場合、
前記線を裏向き果実に対する実験により予め定めた等間隔で区切り、
区切られた前記間隔毎の点における前記線の角度を求め、
前記点の隣り合うものの角度差の絶対値が裏向き果実に対する実験により予め定めた角度差閾値を超える回数を数えて裏向き用設定の凹凸度Crを算出する凹凸度算出工程と、
前記裏向き用設定の凹凸度Crの分布が前記表向き用設定の凹凸度Cfの分布と同一尺度になるように実験により予め定めた変換式により、前記裏向き用設定の凹凸度Crを裏向き用設定の補正凹凸度Cr’に変換し、
裏向き用設定の補正凹凸度Cr’を実験により予め定めた凹凸度閾値と比較し、
前記補正凹凸度Cr’が前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、前記果実は前記形状異常果であると判断し、
前記補正凹凸度Cr’が前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、前記果実は形状正常果であると判断する形状判断工程と、
を含むことを特徴とする。
Further , in the method for selecting citrus fruits having irregularities on the surface of the invention according to claim 2 , in order to solve the above-mentioned problem, 100% inspection is performed while transporting the citrus fruits on a conveyor, and the shape abnormality having irregularities on the surface is performed. It is a sorting method that sorts fruits online in a non-contact manner.
The conveyor transports the fruits, whose front and back are random, in a row without arranging the fruits face up or face down.
The fruit is photographed at an appropriate position on the transport path of the conveyor, and the presence or absence of calyx in a predetermined area is identified by image processing. The front and back judgment process to judge that the fruit is face down,
An imaging step of irradiating the fruit with linear light from a light source provided at an appropriate position on the conveyor's transport path and photographing the fruit with a camera installed at an angle different from the light source.
From the image data obtained by processing the image taken in the image taking step, a line due to the linear light applied to the fruit is extracted.
When it is determined that the fruit is ostensibly in the front and back determination process,
The lines are separated by experiments on ostensibly fruits at equal intervals.
Find the angle of the line at each of the separated points.
The unevenness calculation step of calculating the unevenness Cf of the outward setting by counting the number of times that the absolute value of the angle difference of the adjacent points exceeds the angle difference threshold value predetermined by the experiment on the ostensibly fruit.
The unevenness Cf calculated in the unevenness calculation step is compared with a predetermined unevenness threshold value by an experiment.
When the unevenness Cf is equal to or higher than the unevenness threshold value or larger than the unevenness threshold value, it is determined that the fruit is the abnormal shape fruit.
When the unevenness Cf is smaller than the unevenness threshold value or equal to or lower than the unevenness threshold value, the shape determination step of determining that the fruit is a normal shape fruit and
When it is determined that the fruit is face-down in the front-back judgment step,
Divide the line at equal intervals predetermined by experiments on face-down fruits.
Find the angle of the line at each of the separated points.
The unevenness calculation step of calculating the unevenness Cr of the reverse setting by counting the number of times that the absolute value of the angle difference of the adjacent points exceeds the angle difference threshold value predetermined by the experiment on the face-down fruit.
The unevenness Cr of the face-down setting is turned face down by a conversion formula predetermined by an experiment so that the distribution of the unevenness Cr of the face-down setting is on the same scale as the distribution of the unevenness Cf of the face-up setting. Converted to the correction unevenness Cr'of the setting for
Comparing the correction unevenness Cr'of the face-down setting with the unevenness threshold value predetermined by the experiment,
When the corrected unevenness Cr'is equal to or higher than the unevenness threshold value or larger than the unevenness degree threshold value, it is determined that the fruit is the abnormal shape fruit.
When the corrected unevenness Cr'is smaller than the unevenness threshold value or is equal to or less than the unevenness degree threshold value, the shape determination step of determining that the fruit is a normal shape fruit and the shape determination step.
It is characterized by including.

さらにまた、請求項に係る発明の表面に凹凸がある柑橘類の果実の選別方法は、前記課題解決のために、柑橘類の果実をコンベアで搬送しながら全数検査を行い、表面に凹凸がある形状異常果を非接触でオンライン選別する選別方法であって、
前記コンベアは、前記果実を表向き又は裏向きに整列することなく、表裏がランダムである前記果実を一列に搬送するものであり、
前記コンベアの搬送経路の適宜位置で前記果実を撮影し、所定エリア内におけるヘタの有無を画像処理で識別して、ヘタが有る場合には表向き果実であると判断し、ヘタが無い場合には裏向き果実であると判断する表裏判断工程と、
前記コンベアの搬送経路の適宜位置に設けた光源から前記果実に線状光を照射し、前記光源と異なる角度で設置したカメラで前記果実を撮影する画像撮影工程と、
前記画像撮影工程で撮影した画像を画像処理した画像データから、前記果実に照射された前記線状光による線を抽出し、
前記表裏判断工程で、裏向き果実であると判断した場合、
前記線を裏向き果実に対する実験により予め定めた等間隔で区切り、
区切られた前記間隔毎の点における前記線の角度を求め、
前記点の隣り合うものの角度差の絶対値が裏向き果実に対する実験により予め定めた角度差閾値を超える回数を数えて裏向き用設定の凹凸度Crを算出する凹凸度算出工程と、
前記凹凸度算出工程で算出した前記凹凸度Crを実験により予め定めた凹凸度閾値と比較し、
前記凹凸度Crが前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、前記果実は前記形状異常果であると判断し、
前記凹凸度Crが前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、前記果実は形状正常果であると判断する形状判断工程と、
前記表裏判断工程で、表向き果実であると判断した場合、
前記線を表向き果実に対する実験により予め定めた等間隔で区切り、
区切られた前記間隔毎の点における前記線の角度を求め、
前記点の隣り合うものの角度差の絶対値が表向き果実に対する実験により予め定めた角度差閾値を超える回数を数えて表向き用設定の凹凸度Cfを算出する凹凸度算出工程と、
前記表向き用設定の凹凸度Cfの分布が前記裏向き用設定の凹凸度Crの分布と同一尺度になるように実験により予め定めた変換式により、前記表向き用設定の凹凸度Cfを表向き用設定の補正凹凸度Cf’に変換し、
表向き用設定の補正凹凸度Cf’を実験により予め定めた凹凸度閾値と比較し、
前記補正凹凸度Cf’が前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、前記果実は前記形状異常果であると判断し、
前記補正凹凸度Cf’が前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、前記果実は形状正常果であると判断する形状判断工程と、
を含むことを特徴とする。
Furthermore , in the method for selecting citrus fruits having irregularities on the surface of the invention according to claim 3 , in order to solve the above-mentioned problems, 100% inspection is performed while transporting the citrus fruits on a conveyor, and the shape has irregularities on the surface. It is a sorting method that sorts abnormal fruits online in a non-contact manner.
The conveyor transports the fruits, whose front and back are random, in a row without arranging the fruits face up or face down.
The fruit is photographed at an appropriate position on the transport path of the conveyor, and the presence or absence of calyx in a predetermined area is identified by image processing. The front and back judgment process to judge that the fruit is face down,
An imaging step of irradiating the fruit with linear light from a light source provided at an appropriate position on the conveyor's transport path and photographing the fruit with a camera installed at an angle different from the light source.
From the image data obtained by processing the image taken in the image taking step, a line due to the linear light applied to the fruit is extracted.
When it is determined that the fruit is face-down in the front-back judgment step,
Divide the line at equal intervals predetermined by experiments on face-down fruits.
Find the angle of the line at each of the separated points.
The unevenness calculation step of calculating the unevenness Cr of the reverse setting by counting the number of times that the absolute value of the angle difference of the adjacent points exceeds the angle difference threshold value predetermined by the experiment on the face-down fruit.
The unevenness Cr calculated in the unevenness calculation step is compared with a predetermined unevenness threshold value by an experiment.
When the unevenness Cr is equal to or higher than the unevenness threshold value or larger than the unevenness threshold value, it is determined that the fruit is the abnormal shape fruit.
When the unevenness Cr is smaller than the unevenness threshold value or equal to or less than the unevenness threshold value, the shape determination step of determining that the fruit is a normal shape fruit and
When it is determined that the fruit is ostensibly in the front and back determination process,
The lines are separated by experiments on ostensibly fruits at equal intervals.
Find the angle of the line at each of the separated points.
The unevenness calculation step of calculating the unevenness Cf of the outward setting by counting the number of times that the absolute value of the angle difference of the adjacent points exceeds the angle difference threshold value predetermined by the experiment on the ostensibly fruit.
The unevenness Cf of the face-up setting is set to face-up by a conversion formula predetermined by an experiment so that the distribution of the unevenness Cf of the face-up setting becomes the same scale as the distribution of the unevenness Cr of the face-down setting. Converted to the correction unevenness Cf'of
Comparing the corrected unevenness Cf'of the face-up setting with the unevenness threshold value predetermined by the experiment,
When the corrected unevenness Cf'is equal to or higher than the unevenness threshold value or larger than the unevenness degree threshold value, it is determined that the fruit is the abnormal shape fruit.
When the corrected unevenness Cf'is smaller than the unevenness threshold value or is equal to or less than the unevenness degree threshold value, the shape determination step of determining that the fruit is a normal shape fruit and the shape determination step.
It is characterized by including.

請求項ないしに係る発明の表面に凹凸がある柑橘類の果実の選別方法によれば、柑橘類の果実を搬送するコンベアは、前記果実を表向き又は裏向きに整列することなく、表裏がランダムである前記果実を一列に搬送するものであることから、前記回転装置又は整列装置等を備えない。
よって、コンベア装置の構成が複雑にならないので、製造コスト及び保守コストが増大しない。
その上、特許文献1の表面に凹凸がある柑橘類の果実の選別方法のように、前記角度差データの偏差が一定の閾値から外れた場合に表面に凹凸がある形状異常果であると判断するのではなく、前記角度差の絶対値が表向き果実に対する実験により予め定めた角度差閾値を超える回数を数えて凹凸度を算出し、この凹凸度が実験により予め定めた凹凸度閾値以上又は前記凹凸度閾値よりも大きい場合に表面に凹凸がある形状異常果であると判断する。
よって、現実の果実における表面に凹凸がある形状異常果により適合した選別方法であるので、形状異常果の選別精度を向上できる。
その上、表裏がランダムに搬送されてくる果実に対して、表向き用設定の凹凸度Cf、及び裏向き用設定の凹凸度Crについて、表裏で値のレベルと散らばり具合が異なることから、裏向き用設定の凹凸度Crの分布が表向き用設定の凹凸度Cfの分布と同一尺度になるように実験により予め定めた変換式により、裏向き用設定の凹凸度Crを裏向き用設定の補正凹凸度Cr’に変換、又は、表向き用設定の凹凸度Cfの分布が裏向き用設定の凹凸度Crの分布と同一尺度になるように実験により予め定めた変換式により、表向き用設定の凹凸度Cfを表向き用設定の補正凹凸度Cf’に変換し、表裏の値のレベルと散らばり具合を揃えている。
そして、請求項に係る発明においては、表向き用設定の凹凸度Cf及び裏向き用設定の補正凹凸度Cr’の大きい方を凹凸度とし、又は、表向き用設定の補正凹凸度Cf’及び裏向き用設定の凹凸度Crの大きい方を凹凸度として、実験により予め定めた凹凸度閾値と比較し、形状異常果か否かの判断を行っている。
また、請求項に係る発明においては、表裏判断工程で表向き果実であると判断した場合には、表向き用設定の凹凸度Cfを実験により予め定めた凹凸度閾値と比較し、形状異常果か否かの判断を行い、前記表裏判断工程で裏向き果実であると判断した場合には、裏向き用設定の補正凹凸度Cr’を実験により予め定めた凹凸度閾値と比較し、形状異常果か否かの判断を行っている。
さらに、請求項に係る発明においては、表裏判断工程で裏向き果実であると判断した場合には、裏向き用設定の凹凸度Crを実験により予め定めた凹凸度閾値と比較し、形状異常果か否かの判断を行い、前記表裏判断工程で表向き果実であると判断した場合には、表向き用設定の補正凹凸度Cf’を実験により予め定めた凹凸度閾値と比較し、形状異常果か否かの判断を行っている。
よって、表面性状の特徴に差異がある表側(ヘタ側)と裏側を考慮し、ランダムに搬送されてくる表向き果実又は表向き果実に合わせた調整を行って凹凸度を算出し、凹凸度閾値と比較して前記判断を行うので、形状異常果の選別精度を一層向上できる。
According to the method for selecting citrus fruits having irregularities on the surface of the invention according to claims 1 to 3 , the conveyor for transporting the citrus fruits does not arrange the fruits face up or face down, and the front and back sides are randomly arranged. Since the fruit is transported in a row, it is not provided with the rotating device, the aligning device, or the like.
Therefore, since the configuration of the conveyor device is not complicated, the manufacturing cost and the maintenance cost do not increase.
Further, as in the method for selecting citrus fruits having irregularities on the surface of Patent Document 1, when the deviation of the angle difference data deviates from a certain threshold value, it is determined that the fruit has irregular shapes on the surface. Instead, the degree of unevenness is calculated by counting the number of times that the absolute value of the angle difference exceeds the angle difference threshold value predetermined by the experiment on the ostensibly fruit, and the degree of unevenness is equal to or more than the degree of unevenness threshold value predetermined by the experiment or the unevenness. If it is larger than the degree threshold value, it is judged that the fruit has an irregular shape on the surface.
Therefore, since the sorting method is more suitable for the irregularly shaped fruit having irregularities on the surface of the actual fruit, the sorting accuracy of the abnormally shaped fruit can be improved.
In addition, for fruits whose front and back are randomly transported, the level of value and the degree of dispersion are different between the front and back for the unevenness Cf set for the front side and the unevenness Cr for the back side, so that the front and back sides are face down. The unevenness Cr of the face-down setting is corrected by the conversion formula predetermined by the experiment so that the distribution of the unevenness Cr of the face-up setting is on the same scale as the distribution of the unevenness Cf of the face-up setting. Converted to degree Cr'or, according to a conversion formula predetermined by an experiment so that the distribution of unevenness Cf set for face-up is the same scale as the distribution of unevenness Cr set for face-up, the degree of unevenness set for face-up Cf is converted to the correction unevenness Cf'of the face-up setting, and the level of the values on the front and back and the degree of dispersion are aligned.
Then, in the invention according to claim 1 , the larger of the unevenness Cf of the face-up setting and the correction unevenness Cr'of the face-up setting is set as the unevenness, or the correction unevenness Cf'and the back side of the face-up setting. The larger of the unevenness Cr of the orientation setting is set as the unevenness, and the unevenness is compared with the unevenness threshold value predetermined by the experiment to determine whether or not the result is an abnormal shape.
Further, in the invention according to claim 2 , when it is determined in the front / back determination step that the fruit is ostensibly fruit, the unevenness Cf set for ostensibly is compared with the unevenness threshold value predetermined by the experiment, and the fruit has an abnormal shape. When it is judged whether or not the fruit is face-down in the front-back judgment step, the corrected unevenness Cr'of the face-down setting is compared with the unevenness threshold value predetermined by the experiment, and the shape abnormal fruit is obtained. We are making a judgment as to whether or not it is.
Further, in the invention according to claim 3 , when it is determined that the fruit is face-down in the front-back determination step, the unevenness Cr set for face-down is compared with the unevenness threshold value predetermined by the experiment, and the shape is abnormal. When it is judged whether or not the fruit is fruit or not, and if it is determined to be a face-up fruit in the front-back judgment step, the corrected unevenness Cf'of the face-up setting is compared with the unevenness threshold value predetermined by the experiment, and the shape abnormal fruit is obtained. We are making a judgment as to whether or not it is.
Therefore, considering the front side (septal side) and the back side, which have different surface texture characteristics, the degree of unevenness is calculated by adjusting according to the face-up fruit or the face-up fruit that is randomly transported, and compared with the degree of unevenness threshold. Since the above determination is made, the accuracy of sorting the abnormally shaped fruits can be further improved.

さらにまた、請求項に係る発明の表面に凹凸がある柑橘類の果実の選別方法は、請求項1〜に係る発明の何れかの構成に加え、前記画像撮影工程で撮影した画像を画像処理した画像データから前記線を抽出する際に、
離間している2本の抽出線について、
(a)前記2本の抽出線の端点間の距離が実験により予め定めた閾値よりも近いこと、
(b)前記2本の抽出線の端点の方向差が実験により予め定めた閾値よりも小さいこと、
(c)前記2本の抽出線の端点の垂直方向の距離が実験により予め定めた閾値よりも小さいこと、
の(a)ないし(c)の条件を全て満たす場合、
前記2本の抽出線の端点間を補間して前記2本の抽出線を接続する。
Furthermore, in the method for selecting citrus fruits having irregularities on the surface of the invention according to claim 4 , in addition to the configuration of any one of the inventions according to claims 1 to 3 , the image taken in the image taking step is image-processed. when extracting the wire from the image data,
About two separated extraction lines
(A) The distance between the end points of the two extraction lines is closer than the threshold value predetermined by the experiment.
(B) The direction difference between the end points of the two extraction lines is smaller than the threshold value predetermined by the experiment.
(C) The vertical distance between the end points of the two extraction lines is smaller than the threshold value predetermined by the experiment.
When all the conditions (a) to (c) are satisfied,
The two extraction lines are connected by interpolating between the end points of the two extraction lines.

請求項に係る発明の表面に凹凸がある柑橘類の果実の選別方法によれば、前記請求項1ないしに係る発明の表面に凹凸がある柑橘類の果実の選別方法の作用効果に加え、所定の条件を満たす場合に、離間している2本の抽出線の端点を接続することから、分断された線では確認できなかった表面の凹凸の特徴が現れる可能性が高くなるので、形状異常果の選別精度をより一層向上できる。 According to the method for selecting citrus fruits having irregularities on the surface of the invention according to claim 4 , in addition to the effects of the method for selecting citrus fruits having irregularities on the surface according to claims 1 to 3, a predetermined method is used. Since the end points of the two separated extraction lines are connected when the condition of The sorting accuracy of citrus fruit can be further improved.

以上のとおり、本発明に係る表面に凹凸がある柑橘類の果実の選別方法によれば、コンベア装置の構成を複雑にすることなく、表面に凹凸がある形状異常果の選別精度を向上できるという顕著な効果を奏する。 As described above, according to the method for selecting citrus fruits having irregularities on the surface according to the present invention, it is remarkable that the accuracy of selecting irregularly shaped fruits having irregularities on the surface can be improved without complicating the configuration of the conveyor device. It has a great effect.

本発明の実施の形態に係る表面に凹凸がある柑橘類の果実の選別方法を行う装置の一例を示す斜視図である。It is a perspective view which shows an example of the apparatus which performs the method of sorting the citrus fruit which has the unevenness on the surface which concerns on embodiment of this invention. 画像撮影工程の概略説明図である。It is the schematic explanatory drawing of the image taking process. 第1の方法の凹凸度算出工程を示す概略説明図であり、(a)は表向き果実の場合、(b)は裏向き果実である場合を示している。It is a schematic explanatory drawing which shows the unevenness calculation process of 1st method, (a) shows the case of a face-up fruit, (b) shows the case of a face-down fruit. 第2の方法の凹凸度算出工程を示す概略説明図である。It is the schematic explanatory drawing which shows the unevenness calculation process of 2nd method. 第3の方法の凹凸度算出工程を示す概略説明図であり、(a)は表向き果実の場合、(b)は裏向き果実である場合を示している。It is a schematic explanatory drawing which shows the unevenness calculation process of 3rd method, (a) shows the case of a face-up fruit, (b) shows the case of a face-down fruit.

以下、本発明に係る実施形態を図面に基づいて説明する。 Hereinafter, embodiments according to the present invention will be described with reference to the drawings.

<コンベア及び画像撮影部>
図1の斜視図に示すように、柑橘類の果実1は、コンベア2により搬送方向Tへ一定速度で搬送される。
コンベア2は、果実1,1,…を表向き又は裏向きに整列することなく、表裏がランダムである果実1,1,…を一列に搬送する。
搬送経路の適宜位置には、画像撮影部Aを備え、画像撮影部A内には、光源3、及びカメラ4を所定位置に配置する。
光源3は、コンベア2に向かって測定用光(例えば、平行な直線の線状光)を照射し、コンベア2上を運ばれてきた果実1が測定用光照射エリアに入るとセンサが感知し、カメラ4が撮影を行う。
ここで、図1における画像撮影部Aは、内部の構成部品を視認できるように図示しない手前側のカバーを外している。実稼働時には、前記カバーを取り付けて内部への外光の侵入を抑制する。
<Conveyor and imaging section>
As shown in the perspective view of FIG. 1, the citrus fruit 1 is conveyed by the conveyor 2 in the conveying direction T at a constant speed.
The conveyor 2 conveys the fruits 1, 1, ... With random front and back in a row without arranging the fruits 1, 1, ... Face up or face down.
An image capturing unit A is provided at an appropriate position on the transport path, and a light source 3 and a camera 4 are arranged at predetermined positions in the image capturing unit A.
The light source 3 irradiates the measurement light (for example, parallel straight linear light) toward the conveyor 2, and the sensor detects that the fruit 1 carried on the conveyor 2 enters the measurement light irradiation area. , The camera 4 takes a picture.
Here, the image capturing unit A in FIG. 1 has a cover on the front side (not shown) removed so that the internal components can be visually recognized. During actual operation, the cover is attached to prevent outside light from entering the inside.

<画像撮影工程>
図1の斜視図、及び図2の概略説明図に示すように、画像撮影部Aにおけるカメラ4は、光源3と異なる角度で設置しているため、果実の凹凸を撮影できる。
光源3から測定用光5を果実1に照射し、カメラ4は果実1を撮影する。
ここで、測定用光5は、平行な直線の線状光に限定されるものではなく、同心円状の円状光、多数の円状光を並置したもの、正方形を並置した格子状光等であってもよい。
<Image shooting process>
As shown in the perspective view of FIG. 1 and the schematic explanatory view of FIG. 2, since the camera 4 in the image capturing unit A is installed at an angle different from that of the light source 3, the unevenness of the fruit can be photographed.
The light source 3 irradiates the fruit 1 with the measurement light 5, and the camera 4 photographs the fruit 1.
Here, the measurement light 5 is not limited to parallel straight linear light, but is concentric circular light, a large number of circular lights juxtaposed, a grid light in which squares are juxtaposed, and the like. There may be.

以下において、第1の方法、第2の方法、及び第3の方法について、それらの凹凸度段出工程、及び形状判断工程を説明する。
以下の説明において、コンベア2により搬送されてくる柑橘類の果実1について、表向き果実をF、裏向き果実をRとする。
以下における、間隔Df、角度差閾値θfは、表向き果実Fだけで値を少しずつ変えながら実験を行い、最も精度が高くなるように定めた値であり、間隔Dr、角度差閾値θrは、裏向き果実Rだけで値を少しずつ変えながら実験を行い、最も精度が高くなるように定めた値である。
Hereinafter, with respect to the first method, the second method, and the third method, the unevenness step-up step and the shape determination step thereof will be described.
In the following description, for the citrus fruit 1 transported by the conveyor 2, the face-up fruit is F and the face-down fruit is R.
In the following, the interval Df and the angle difference threshold θf are values determined so as to have the highest accuracy by conducting an experiment while changing the values little by little only for the ostensibly fruit F, and the interval Dr and the angle difference threshold θr are the back values. It is a value determined so that the accuracy is the highest by conducting an experiment while changing the value little by little only with the orientation fruit R.

[第1の方法]
図3(a)及び図3(b)の概略説明図を参照して第1の方法について説明する。
[First method]
The first method will be described with reference to the schematic explanatory views of FIGS. 3 (a) and 3 (b).

<凹凸度算出工程>
第1の方法では、前記画像撮影工程で撮影した画像を画像処理した画像データ6から線7を抽出し、線7を表向き果実Fに対する実験により予め定めた等間隔Dfで区切る。
そして、区切られた間隔Df毎の点Piにおける線7の角度を求め、点Piの隣り合うものの角度差の絶対値が表向き果実Fに対する実験により予め定めた角度差閾値θfを超える回数を数えて凹凸度Cを算出する。
<Concavity and convexity calculation process>
In the first method, the line 7 is extracted from the image data 6 obtained by image-processing the image taken in the image taking step, and the line 7 is divided by the equidistant Df predetermined by the experiment on the ostensibly fruit F.
Then, the angle of the line 7 at the point Pi for each of the divided intervals Df is obtained, and the number of times that the absolute value of the angle difference between the adjacent points Pi exceeds the angle difference threshold θf predetermined by the experiment on the ostensibly fruit F is counted. The degree of unevenness C is calculated.

<形状判断工程>
前記凹凸度算出工程で算出した凹凸度Cを実験により予め定めた凹凸度閾値と比較し、凹凸度Cが前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、果実1は表面に凹凸がある形状異常果であると判断する。
凹凸度Cが前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、果実1は形状正常果であると判断する。
<Shape judgment process>
The unevenness C calculated in the unevenness calculation step is compared with the unevenness threshold value predetermined by the experiment, and when the unevenness C is equal to or higher than the unevenness threshold value or larger than the unevenness threshold value, the fruit 1 Is judged to be an abnormal shape fruit with irregularities on the surface.
When the unevenness C is smaller than the unevenness threshold value or equal to or lower than the unevenness degree threshold value, the fruit 1 is judged to be a fruit having a normal shape.

[第2の方法]
図4の概略説明図を参照して第2の方法について説明する。
[Second method]
The second method will be described with reference to the schematic explanatory view of FIG.

<凹凸度算出工程>
第2の方法では、前記画像撮影工程で撮影した画像を画像処理した画像データ6から線7を抽出し、線7を表向き果実Fに対する実験により予め定めた等間隔Dfで区切る。
そして、区切られた間隔Df毎の点Piにおける線7の角度を求め、点Piの隣り合うものの角度差の絶対値が表向き果実Fに対する実験により予め定めた角度差閾値θfを超える回数を数えて表向き用設定の凹凸度Cfを算出する。
<Concavity and convexity calculation process>
In the second method, the line 7 is extracted from the image data 6 obtained by image-processing the image taken in the image taking step, and the line 7 is divided by the equidistant Df predetermined by the experiment on the ostensibly fruit F.
Then, the angle of the line 7 at the point Pi for each of the divided intervals Df is obtained, and the number of times that the absolute value of the angle difference between the adjacent points Pi exceeds the angle difference threshold θf predetermined by the experiment on the ostensibly fruit F is counted. The unevenness Cf of the face-up setting is calculated.

同様に、線7を裏向き果実Rに対する実験により予め定めた等間隔Drで区切る。
そして、区切られた間隔Dr毎の点Piにおける線7の角度を求め、点Piの隣り合うものの角度差の絶対値が裏向き果実Rに対する実験により予め定めた角度差閾値θrを超える回数を数えて裏向き用設定の凹凸度Crを算出する。
Similarly, the line 7 is separated by experimentally defined equidistant Drs on the face-down fruit R.
Then, the angle of the line 7 at the point Pi for each divided interval Dr is obtained, and the number of times that the absolute value of the angle difference between adjacent points Pi exceeds the angle difference threshold θr predetermined by the experiment for the face-down fruit R is counted. The unevenness Cr of the face-down setting is calculated.

表向き用設定の凹凸度Cfと裏向き用設定の凹凸度Crでは、値のレベルと散らばり具合が異なる。よって、表裏の値のレベルと散らばり具合を揃えるための補正を行う。
すなわち、裏向き用設定の凹凸度Crの分布が表向き用設定の凹凸度Cfの分布と同一尺度になるように実験により予め定めた変換式により、裏向き用設定の凹凸度Crを裏向き用設定の補正凹凸度Cr’に変換する。
例えば、実験で予め求めた、表向き用設定の凹凸度Cfの平均をEf、標準偏差をσf、裏向き用設定の凹凸度Crの平均をEr、標準偏差をσrとし、裏向き用設定の補正凹凸度Cr’を、式(1)で算出する。
Cr’=(Cr−Er)・σf/σr+Ef (1)
The level of the value and the degree of dispersion are different between the unevenness Cf set for the front side and the unevenness Cr of the setting for the back side. Therefore, a correction is made to make the level of the values on the front and back and the degree of dispersion uniform.
That is, the unevenness Cr of the face-down setting is used for face-down by a conversion formula predetermined by an experiment so that the distribution of the unevenness Cr of the face-down setting becomes the same scale as the distribution of the unevenness Cf of the face-up setting. Convert to the set correction unevenness Cr'.
For example, the average of the unevenness Cf of the face-up setting is Ef, the standard deviation is σf, the average of the unevenness Cr of the face-down setting is Er, and the standard deviation is σr, which are obtained in advance in the experiment. The degree of unevenness Cr'is calculated by the equation (1).
Cr'= (Cr-Er) · σf / σr + Ef (1)

そして、表向き用設定の凹凸度Cf及び裏向き用設定の補正凹凸度Cr’の大きい方を凹凸度Cとする。 Then, the larger of the unevenness Cf of the front-facing setting and the corrected unevenness Cr'of the back-facing setting is defined as the unevenness C.

あるいは、表向き用設定の凹凸度Cfの分布が裏向き用設定の凹凸度Crの分布と同一尺度になるように実験により予め定めた変換式により、表向き用設定の凹凸度Cfを表向き用設定の補正凹凸度Cf’に変換してもよい。
その場合は、表向き用設定の補正凹凸度Cf’及び裏向き設定の凹凸度Crの大きい方を凹凸度Cとする。
Alternatively, the unevenness Cf of the face-up setting is set to the face-up setting by a conversion formula predetermined by an experiment so that the distribution of the unevenness Cf of the face-up setting becomes the same scale as the distribution of the unevenness Cr of the face-down setting. It may be converted to the correction unevenness Cf'.
In that case, the larger of the correction unevenness Cf'of the face-up setting and the unevenness Cr of the face-up setting is defined as the unevenness C.

<形状判断工程>
前記凹凸度算出工程で算出した凹凸度Cを実験により予め定めた凹凸度閾値と比較し、凹凸度Cが前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、果実1は前記形状異常果であると判断する。
凹凸度Cが前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、果実1は形状正常果であると判断する。
<Shape judgment process>
The unevenness C calculated in the unevenness calculation step is compared with the unevenness threshold value predetermined by the experiment, and when the unevenness C is equal to or higher than the unevenness threshold value or larger than the unevenness threshold value, the fruit 1 Is determined to be the abnormal shape fruit.
When the unevenness C is smaller than the unevenness threshold value or equal to or lower than the unevenness degree threshold value, the fruit 1 is judged to be a fruit having a normal shape.

[第3の方法]
図5(a)及び図5(b)の概略説明図を参照して第3の方法について説明する。
[Third method]
The third method will be described with reference to the schematic explanatory views of FIGS. 5 (a) and 5 (b).

<表裏判断工程>
第3の方法では、コンベア2の搬送経路の適宜位置で果実1を撮影し、所定エリア内におけるヘタ1Aの有無を画像処理で識別して、ヘタ1Aが有る場合には表向き果実Fであると判断し、ヘタ1Aが無い場合には裏向き果実Rであると判断する。
このような果実1の表裏の判断は、例えば図1の画像撮影部Aの上流側で行うが、構成によっては、画像撮影部A内で行ってもよい。
<Front and back judgment process>
In the third method, the fruit 1 is photographed at an appropriate position on the transport path of the conveyor 2, the presence or absence of the calyx 1A in the predetermined area is identified by image processing, and if the calyx 1A is present, the fruit F is ostensibly. Judgment is made, and if there is no calyx 1A, it is judged to be a face-down fruit R.
Such determination of the front and back of the fruit 1 is performed, for example, on the upstream side of the image capturing unit A in FIG. 1, but may be performed within the image capturing unit A depending on the configuration.

<表向き果実であると判断した場合の凹凸度算出工程>
図5(a)に示すように、前記画像撮影工程で撮影した画像を画像処理した画像データ6から線7を抽出し、線7を表向き果実Fに対する実験により予め定めた等間隔Dfで区切る。
そして、区切られた間隔Df毎の点Piにおける線7の角度を求め、点Piの隣り合うものの角度差の絶対値が表向き果実Fに対する実験により予め定めた角度差閾値θfを超える回数を数えて表向き用設定の凹凸度Cfを算出する。
<Process of calculating unevenness when it is judged to be an ostensibly fruit>
As shown in FIG. 5A, a line 7 is extracted from the image data 6 obtained by image processing the image taken in the image taking step, and the line 7 is divided by an equal interval Df predetermined by an experiment on the ostensibly fruit F.
Then, the angle of the line 7 at the point Pi for each of the divided intervals Df is obtained, and the number of times that the absolute value of the angle difference between the adjacent points Pi exceeds the angle difference threshold θf predetermined by the experiment on the ostensibly fruit F is counted. The unevenness Cf of the face-up setting is calculated.

<表向き果実であると判断した場合の形状判断工程>
前記凹凸度算出工程で算出した表向き用設定の凹凸度Cfを実験により予め定めた凹凸度閾値と比較し、凹凸度Cfが前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、果実1は前記形状異常果であると判断する。
凹凸度Cfが前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、果実1は形状正常果であると判断する。
<Shape determination process when it is determined that the fruit is ostensibly>
When the unevenness Cf of the face-up setting calculated in the unevenness calculation step is compared with the unevenness threshold value predetermined by the experiment, and the unevenness degree Cf is equal to or more than the unevenness threshold value or larger than the unevenness degree threshold value. Determines that the fruit 1 is the abnormally shaped fruit.
When the unevenness Cf is smaller than the unevenness threshold value or equal to or lower than the unevenness degree threshold value, the fruit 1 is judged to be a fruit having a normal shape.

<裏向き果実であると判断した場合の凹凸度算出工程>
図5(b)に示すように、前記画像撮影工程で撮影した画像を画像処理した画像データ6から線7を抽出し、線7を裏向き果実Rに対する実験により予め定めた等間隔Drで区切る。
そして、区切られた間隔Dr毎の点Piにおける線7の角度を求め、点Piの隣り合うものの角度差の絶対値が表向き果実Fに対する実験により予め定めた角度差閾値θrを超える回数を数えて裏向き用設定の凹凸度Crを算出する。
前記第2の方法と同様に、表裏の値のレベルと散らばり具合を揃えるために、裏向き用設定の凹凸度Crの分布が表向き用設定の凹凸度Cfの分布と同一尺度になるように実験により予め定めた変換式により、裏向き用設定の凹凸度Crを裏向き用設定の補正凹凸度Cr’に変換する。例えば、裏向き用設定の補正凹凸度Cr’を、式(1)で算出する。
<Process of calculating unevenness when it is judged to be a face-down fruit>
As shown in FIG. 5B, a line 7 is extracted from the image data 6 obtained by image processing the image taken in the image taking step, and the line 7 is divided by a predetermined equal interval Dr by an experiment on the face-down fruit R. ..
Then, the angle of the line 7 at the point Pi for each divided interval Dr is obtained, and the number of times that the absolute value of the angle difference between the adjacent points Pi exceeds the angle difference threshold θr predetermined by the experiment on the ostensibly fruit F is counted. The unevenness Cr of the face-down setting is calculated.
Similar to the second method, in order to make the level of the values on the front and back and the degree of dispersion uniform, the experiment was performed so that the distribution of the unevenness Cr of the face-down setting was the same scale as the distribution of the unevenness Cf of the face-up setting. The unevenness Cr of the face-down setting is converted to the correction unevenness Cr'of the face-down setting by the conversion formula predetermined by. For example, the correction unevenness Cr'of the face-down setting is calculated by the equation (1).

<裏向き果実であると判断した場合の形状判断工程>
前記凹凸度算出工程で算出した裏向き用設定の補正凹凸度Cr’を実験により予め定めた凹凸度閾値と比較し、補正凹凸度Cr’が前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、果実1は前記形状異常果であると判断する。
補正凹凸度Cr’が前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、果実1は形状正常果であると判断する。
<Shape determination process when determining that the fruit is face down>
The corrected unevenness Cr'of the face-down setting calculated in the unevenness calculation step is compared with the unevenness threshold value predetermined by the experiment, and when the corrected unevenness Cr'is equal to or more than the unevenness threshold value or the unevenness degree threshold value is obtained. If it is larger than, it is determined that the fruit 1 is the abnormally shaped fruit.
When the corrected unevenness Cr'is smaller than the unevenness threshold value or is equal to or less than the unevenness degree threshold value, it is determined that the fruit 1 is a fruit having a normal shape.

以上の第3の方法において、表裏の値のレベルと散らばり具合を揃えるために、表向き用設定の凹凸度Cfの分布が裏向き用設定の凹凸度Crの分布と同一尺度になるように実験により予め定めた変換式により、表向き用設定の凹凸度Cfを表向き用設定の補正凹凸度Cf’に変換してもよい。
その場合の凹凸度算出工程及び形状判断工程は以下のようになる。
In the above third method, in order to make the level of the values on the front and back and the degree of dispersion uniform, the distribution of the unevenness Cf set for the front side is the same scale as the distribution of the unevenness Cr of the setting for the back side. The unevenness Cf of the face-up setting may be converted to the correction unevenness Cf'of the face-up setting by a predetermined conversion formula.
In that case, the unevenness calculation step and the shape determination step are as follows.

<裏向き果実であると判断した場合の凹凸度算出工程>
図5(b)に示すように、前記画像撮影工程で撮影した画像を画像処理した画像データ6から線7を抽出し、線7を裏向き果実Rに対する実験により予め定めた等間隔Drで区切る。
そして、区切られた間隔Dr毎の点Piにおける線7の角度を求め、点Piの隣り合うものの角度差の絶対値が表向き果実Fに対する実験により予め定めた角度差閾値θrを超える回数を数えて裏向き用設定の凹凸度Crを算出する。
<Process of calculating unevenness when it is judged to be a face-down fruit>
As shown in FIG. 5B, a line 7 is extracted from the image data 6 obtained by image processing the image taken in the image taking step, and the line 7 is divided by a predetermined equal interval Dr by an experiment on the face-down fruit R. ..
Then, the angle of the line 7 at the point Pi for each divided interval Dr is obtained, and the number of times that the absolute value of the angle difference between the adjacent points Pi exceeds the angle difference threshold θr predetermined by the experiment on the ostensibly fruit F is counted. The unevenness Cr of the face-down setting is calculated.

<裏向き果実であると判断した場合の形状判断工程>
前記凹凸度算出工程で算出した裏向き用設定の凹凸度Crを実験により予め定めた凹凸度閾値と比較し、凹凸度Crが前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、果実1は前記形状異常果であると判断する。
凹凸度Crが前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、果実1は形状正常果であると判断する。
<Shape determination process when determining that the fruit is face down>
When the unevenness Cr of the face-down setting calculated in the unevenness calculation step is compared with the unevenness threshold value predetermined by the experiment, and the unevenness Cr is equal to or higher than the unevenness threshold value or larger than the unevenness degree threshold value. It is determined that the fruit 1 is the abnormally shaped fruit.
When the unevenness Cr is smaller than the unevenness threshold value or equal to or lower than the unevenness degree threshold value, the fruit 1 is judged to be a fruit having a normal shape.

<表向き果実であると判断した場合の凹凸度算出工程>
図5(a)に示すように、前記画像撮影工程で撮影した画像を画像処理した画像データ6から線7を抽出し、線7を表向き果実Fに対する実験により予め定めた等間隔Dfで区切る。
そして、区切られた間隔Df毎の点Piにおける線7の角度を求め、点Piの隣り合うものの角度差の絶対値が表向き果実Fに対する実験により予め定めた角度差閾値θfを超える回数を数えて表向き用設定の凹凸度Cfを算出する。
前記第2の方法と同様に、表裏の値のレベルと散らばり具合を揃えるために、表向き用設定の凹凸度Cfの分布が裏向き用設定の凹凸度Crの分布と同一尺度になるように実験により予め定めた変換式により、表向き用設定の凹凸度Cfを表向き用設定の補正凹凸度Cf’に変換する。
<Process of calculating unevenness when it is judged to be an ostensibly fruit>
As shown in FIG. 5A, a line 7 is extracted from the image data 6 obtained by image processing the image taken in the image taking step, and the line 7 is divided by an equal interval Df predetermined by an experiment on the ostensibly fruit F.
Then, the angle of the line 7 at the point Pi for each of the divided intervals Df is obtained, and the number of times that the absolute value of the angle difference between the adjacent points Pi exceeds the angle difference threshold θf predetermined by the experiment on the ostensibly fruit F is counted. The unevenness Cf of the face-up setting is calculated.
Similar to the second method, in order to make the level of the values on the front and back and the degree of dispersion uniform, the experiment was performed so that the distribution of the unevenness Cf in the front setting was the same scale as the distribution of the unevenness Cr in the face setting. Converts the unevenness Cf of the face-up setting to the correction unevenness Cf'of the face-up setting by the conversion formula predetermined by.

<表向き果実であると判断した場合の形状判断工程>
前記凹凸度算出工程で算出した表向き用設定の補正凹凸度Cf’を実験により予め定めた凹凸度閾値と比較し、補正凹凸度Cf’が前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、果実1は前記形状異常果であると判断する。
補正凹凸度Cf’が前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、果実1は形状正常果であると判断する。
<Shape determination process when it is determined that the fruit is ostensibly>
The corrected unevenness Cf'of the face-up setting calculated in the unevenness calculation step is compared with the unevenness threshold value predetermined by the experiment, and when the corrected unevenness degree Cf'is equal to or more than the unevenness threshold value or from the unevenness degree threshold value. If it is also large, it is determined that the fruit 1 is the abnormally shaped fruit.
When the corrected unevenness Cf'is smaller than the unevenness threshold value or is equal to or less than the unevenness degree threshold value, the fruit 1 is judged to be a fruit having a normal shape.

<線の分断への対策>
前記画像撮影工程で撮影した画像を画像処理した画像データ6から線7を抽出する際に、果実1の表面の凹凸が激しいと抽出線が複雑に分断されてしまい数値化に支障が出る場合がある。
そこで、離間している2本の抽出線について、
(a)前記2本の抽出線の端点間の距離が実験により予め定めた閾値よりも近いこと、
(b)前記2本の抽出線の端点の方向差が実験により予め定めた閾値よりも小さいこと、
(c)前記2本の抽出線の端点の垂直方向の距離が実験により予め定めた閾値よりも小さいこと、
の(a)ないし(c)の条件を全て満たす場合、前記2本の抽出線の端点間を補間して前記2本の抽出線を接続する。
補間手法は特定の方法に限定されるものではなく、例えば端点間を単純に直線で接続してもよいし、端点からの数点の座標を使ってスプラインやラグランジュなどの一般的な補間方法で接続してもよい。
<Countermeasures against line breaks>
When the line 7 is extracted from the image data 6 obtained by image-processing the image taken in the image taking step, if the surface of the fruit 1 is severely uneven, the extracted line may be complicatedly divided and quantification may be hindered. be.
Therefore, regarding the two extraction lines that are separated from each other,
(A) The distance between the end points of the two extraction lines is closer than the threshold value predetermined by the experiment.
(B) The direction difference between the end points of the two extraction lines is smaller than the threshold value predetermined by the experiment.
(C) The vertical distance between the end points of the two extraction lines is smaller than the threshold value predetermined by the experiment.
When all the conditions (a) to (c) are satisfied, the two extraction lines are connected by interpolating between the end points of the two extraction lines.
The interpolation method is not limited to a specific method. For example, the endpoints may be simply connected by a straight line, or a general interpolation method such as spline or lagrange may be used using the coordinates of several points from the endpoints. You may connect.

<形状判断工程後の工程>
前記第1の方法、前記第2の方法、又は前記第3の方法で柑橘類の果実1の形状判断を行った後、必要に応じて画像撮影部Aの下流側に設けた近赤外分光部でも、近赤外分光により前記形状異常果であるか否かの形状判別を行う。それにより前記形状異常果の選別精度を向上できる。
前記第1の方法、前記第2の方法、若しくは前記第3の方法の前記形状判断工程、及び近赤外分光部による形状判断を行った結果を選果機に送信し、対象の果実1を選果機が排果する。
なお、近赤外分光部を設けない場合は、前記形状判断工程で形状判断を行った結果のみを選果機に送信し、対象の果実1を選果機が排果する。
<Process after shape judgment process>
After determining the shape of the citrus fruit 1 by the first method, the second method, or the third method, a near-infrared spectroscopic unit provided on the downstream side of the imaging unit A as necessary. However, near-infrared spectroscopy is used to determine the shape of the abnormal shape. Thereby, the sorting accuracy of the abnormally shaped fruit can be improved.
The result of the shape determination step of the first method, the second method, or the third method, and the shape determination by the near-infrared spectroscope is transmitted to the fruit sorter, and the target fruit 1 is selected. The fruit sorter discharges the fruit.
When the near-infrared spectroscope is not provided, only the result of shape determination in the shape determination step is transmitted to the fruit sorter, and the fruit sorter discharges the target fruit 1.

<形状異常果の選別精度の比較実験>
(実験方法)
柑橘類の果実1をみかんの果実とし、果実1,1,…の搬送方向T(図1)の間隔を最短で50mm、コンベア2の搬送速度を60m/minとし、形状正常果76個、形状異常果28について、形状異常果の選別精度を求める。
<Comparison experiment of sorting accuracy of abnormally shaped fruits>
(experimental method)
Citrus fruit 1 is used as mandarin orange fruit, the distance between the fruits 1, 1, ... In the transport direction T (FIG. 1) is 50 mm at the shortest, the transport speed of the conveyor 2 is 60 m / min, 76 normal-shaped fruits, abnormal shape. For the fruit 28, the sorting accuracy of the abnormally shaped fruit is obtained.

(比較例及び実施例)
比較例は特許文献1の方法とし、実施例は以下(1)ないし(5)とした。
(1)実施例1:前記第1の方法
(2)実施例2:前記第2の方法
(3)実施例3:前記第2の方法で前記線の分断への対策を行ったもの
(4)実施例4:前記第3の方法
(5)実施例5:前記第3の方法で前記線の分断への対策を行ったもの
(Comparative Examples and Examples)
The comparative example is the method of Patent Document 1, and the examples are as follows (1) to (5).
(1) Example 1: The first method (2) Example 2: The second method (3) Example 3: The second method is used to take measures against the division of the line (4). ) Example 4: The third method (5) Example 5: The third method is used to take measures against the division of the line.

(実験結果)
実験結果を表1に示す。
実施例の形状異常果の選別精度は、比較例に対して15%〜34%向上している。
また、実施例1から実施例5まで形状異常果の選別精度が向上しているとともに、前記線の分断への対策は効果があり、特に前記第3の方法に前記線の分断への対策を行った場合の効果が大きいことが分かる。
(Experimental result)
The experimental results are shown in Table 1.
The accuracy of sorting the abnormally shaped fruits of the examples is improved by 15% to 34% as compared with the comparative examples.
Further, from Example 1 to Example 5, the accuracy of sorting the abnormally shaped fruits is improved, and the countermeasure against the division of the line is effective. In particular, the third method is provided with the countermeasure against the division of the line. It can be seen that the effect of doing so is great.

Figure 0006926846
Figure 0006926846

以上のような本発明の実施の形態に係る表面に凹凸がある柑橘類の果実の選別方法によれば、柑橘類の果実1,1,…を搬送するコンベア2は、果実1,1,…を表向き又は裏向きに整列することなく、表裏がランダムである果実1,1,…を一列に搬送するものであることから、前記回転装置又は整列装置等を備えない。
よって、コンベア装置2の構成が複雑にならないので、製造コスト及び保守コストが増大しない。
According to the method for selecting citrus fruits having irregularities on the surface according to the embodiment of the present invention as described above, the conveyor 2 for transporting the citrus fruits 1, 1, ... Faces the fruits 1, 1, ... Alternatively, the rotating device, the aligning device, or the like is not provided because the fruits 1, 1, ... Which are random on the front and back are transported in a row without being aligned face down.
Therefore, since the configuration of the conveyor device 2 is not complicated, the manufacturing cost and the maintenance cost do not increase.

本発明の実施の形態に係る第1の方法を用いる表面に凹凸がある柑橘類の果実の選別方法によれば、特許文献1の表面に凹凸がある柑橘類の果実の選別方法のように、前記角度差データの偏差が一定の閾値から外れた場合に表面に凹凸がある形状異常果であると判断するのではなく、前記角度差の絶対値が表向き果実Fに対する実験により予め定めた角度差閾値を超える回数を数えて凹凸度Cを算出し、この凹凸度Cが実験により予め定めた凹凸度閾値以上又は前記凹凸度閾値よりも大きい場合に表面に凹凸がある形状異常果であると判断する。
よって、現実の果実における表面に凹凸がある形状異常果により適合した選別方法であるので、形状異常果の選別精度を向上できる。
According to the method for selecting citrus fruits having irregularities on the surface using the first method according to the first embodiment of the present invention, the angle is the same as the method for selecting citrus fruits having irregularities on the surface of Patent Document 1. When the deviation of the difference data deviates from a certain threshold value, it is not judged that the fruit has an uneven shape on the surface, but the absolute value of the angle difference is the angle difference threshold value predetermined by the experiment on the ostensibly fruit F. The degree of unevenness C is calculated by counting the number of times the number of times is exceeded, and when the degree of unevenness C is equal to or greater than the degree of unevenness threshold set in advance by an experiment or larger than the degree of unevenness threshold, it is determined that the fruit has irregularities on the surface.
Therefore, since the sorting method is more suitable for the irregularly shaped fruit having irregularities on the surface of the actual fruit, the sorting accuracy of the abnormally shaped fruit can be improved.

本発明の実施の形態に係る第2の方法を用いる表面に凹凸がある柑橘類の果実の選別方法、又は本発明の実施の形態に係る第3の方法を用いる表面に凹凸がある柑橘類の果実の選別方法によれば、表裏がランダムに搬送されてくる果実1,1,…に対して、表向き用設定の凹凸度Cf、及び裏向き用設定の凹凸度Crについて、表裏で値のレベルと散らばり具合が異なることから、裏向き用設定の凹凸度Crの分布が表向き用設定の凹凸度Cfの分布と同一尺度になるように実験により予め定めた変換式により、裏向き用設定の凹凸度Crを裏向き用設定の補正凹凸度Cr’に変換、又は、表向き用設定の凹凸度Cfの分布が裏向き用設定の凹凸度Crの分布と同一尺度になるように実験により予め定めた変換式により、表向き用設定の凹凸度Cfを表向き用設定の補正凹凸度Cf’に変換し、表裏の値のレベルと散らばり具合を揃えている。
そして、第2の方法においては、表向き用設定の凹凸度Cf及び裏向き用設定の補正凹凸度Cr’の大きい方を凹凸度とし、又は、表向き用設定の補正凹凸度Cf’及び裏向き用設定の凹凸度Crの大きい方を凹凸度として、実験により予め定めた凹凸度閾値と比較し、形状異常果か否かの判断を行っている。
また、第3の方法においては、表裏判断工程で表向き果実Fであると判断した場合には、表向き用設定の凹凸度Cfを実験により予め定めた凹凸度閾値と比較し、形状異常果か否かの判断を行い、前記表裏判断工程で裏向き果実Rであると判断した場合には、裏向き用設定の補正凹凸度Cr’を実験により予め定めた凹凸度閾値と比較し、形状異常果か否かの判断を行っている。
あるいは、第3の方法においては、表裏判断工程で裏向き果実Rであると判断した場合には、裏向き用設定の凹凸度Crを実験により予め定めた凹凸度閾値と比較し、形状異常果か否かの判断を行い、前記表裏判断工程で表向き果実Fであると判断した場合には、表向き用設定の補正凹凸度Cf’を実験により予め定めた凹凸度閾値と比較し、形状異常果か否かの判断を行っている。
よって、表面性状の特徴に差異がある表側(ヘタ側)と裏側を考慮し、ランダムに搬送されてくる表向き果実F又は表向き果実Rに合わせた調整を行って凹凸度を算出し、凹凸度閾値と比較して前記判断を行うので、形状異常果の選別精度を一層向上できる。
A method for selecting citrus fruits having irregularities on the surface using the second method according to the embodiment of the present invention, or a method for selecting citrus fruits having irregularities on the surface using the third method according to the embodiment of the present invention. According to the sorting method, for the fruits 1, 1, ... Since the conditions are different, the unevenness Cr of the face-down setting is set according to a conversion formula predetermined by an experiment so that the distribution of the unevenness Cr of the face-up setting is on the same scale as the distribution of the unevenness Cf of the face-up setting. Is converted to the correction unevenness Cr'of the face-down setting, or a conversion formula predetermined by an experiment so that the distribution of the unevenness Cf of the face-up setting becomes the same scale as the distribution of the unevenness Cr of the face-down setting. Therefore, the unevenness Cf of the face-up setting is converted to the correction unevenness Cf'of the face-up setting, and the level of the values on the front and back sides and the degree of dispersion are made uniform.
Then, in the second method, the larger of the unevenness Cf of the face-up setting and the correction unevenness Cr'of the face-up setting is set as the unevenness, or the correction unevenness Cf'of the face-up setting and the face-down setting are used. The larger of the set unevenness Cr is set as the unevenness, and the unevenness is compared with the unevenness threshold value predetermined by the experiment to determine whether or not the shape is abnormal.
Further, in the third method, when it is determined in the front / back determination step that the fruit F is ostensibly, the unevenness Cf set for ostensibly is compared with the unevenness threshold value predetermined by the experiment, and whether the fruit is abnormal in shape or not. If it is determined that the fruit R is face-down in the front-back judgment step, the corrected unevenness Cr'of the face-down setting is compared with the unevenness threshold value predetermined by the experiment, and the shape abnormal fruit is obtained. We are making a judgment as to whether or not it is.
Alternatively, in the third method, when it is determined that the fruit R is face-down in the front-back determination step, the unevenness Cr set for face-down is compared with the unevenness threshold value predetermined by the experiment, and the shape abnormal fruit is obtained. When it is determined whether or not the fruit is face-up fruit F in the front-back judgment step, the corrected unevenness Cf'of the face-up setting is compared with the unevenness threshold value predetermined by the experiment, and the shape abnormal fruit is determined. We are making a judgment as to whether or not it is.
Therefore, considering the front side (septal side) and the back side, which have different surface texture characteristics, the unevenness is calculated by adjusting according to the face-up fruit F or the face-up fruit R that are randomly transported, and the unevenness threshold is calculated. Since the above-mentioned determination is made in comparison with the above, the selection accuracy of the abnormally shaped fruit can be further improved.

さらに、前記線の分断への対策を行うことにより、所定の条件を満たす場合に、離間している2本の抽出線の端点が接続されることから、分断された線では確認できなかった表面の凹凸の特徴が現れる可能性が高くなるので、形状異常果の選別精度をより一層向上できる。 Further, by taking measures against the division of the line, the end points of the two separated extraction lines are connected when a predetermined condition is satisfied, so that the surface that cannot be confirmed by the divided line cannot be confirmed. Since there is a high possibility that the unevenness of the shape will appear, the accuracy of sorting abnormally shaped fruits can be further improved.

以上の実施の形態の記載はすべてすべて例示であり、これに制限されるものではない。本発明の範囲から逸脱することなく種々の改良及び変更を施すことができる。 All the descriptions of the above embodiments are examples, and the present invention is not limited thereto. Various improvements and modifications can be made without departing from the scope of the present invention.

1 柑橘類の果実
1A ヘタ
2 コンベア
3 光源
4 カメラ
5 測定用光
6 画像データ
7 線
A 画像撮影部
C,Cf,Cr 凹凸度
Cr’ 補正凹凸度
Df,Dr 間隔
F 表向き果実
Pi 点
R 裏向き果実
T 搬送方向
θf,θr 角度差閾値
1 Citrus fruit 1A Heta 2 Conveyor 3 Light source 4 Camera 5 Measurement light 6 Image data 7 Line A Imaging unit C, Cf, Cr Concavity and convexity Cr'Correction unevenness Df, Dr Interval F Face-up fruit Pi Point R Face-down fruit T Transport direction θf, θr Angle difference threshold

Claims (4)

柑橘類の果実をコンベアで搬送しながら全数検査を行い、表面に凹凸がある形状異常果を非接触でオンライン選別する選別方法であって、
前記コンベアは、前記果実を表向き又は裏向きに整列することなく、表裏がランダムである前記果実を一列に搬送するものであり、
前記コンベアの搬送経路の適宜位置に設けた光源から前記果実に線状光を照射し、前記光源と異なる角度で設置したカメラで前記果実を撮影する画像撮影工程と、
前記画像撮影工程で撮影した画像を画像処理した画像データから、前記果実に照射された前記線状光による線を抽出し、
前記線を表向き果実に対する実験により予め定めた等間隔で区切り、
区切られた前記間隔毎の点における前記線の角度を求め、
前記点の隣り合うものの角度差の絶対値が表向き果実に対する実験により予め定めた角度差閾値を超える回数を数えて表向き用設定の凹凸度Cfを算出し、
前記線を裏向き果実に対する実験により予め定めた等間隔で区切り、
区切られた前記間隔毎の点における前記線の角度を求め、
前記点の隣り合うものの角度差の絶対値が裏向き果実に対する実験により予め定めた角度差閾値を超える回数を数えて裏向き用設定の凹凸度Crを算出し、
前記裏向き用設定の凹凸度Crの分布が前記表向き用設定の凹凸度Cfの分布と同一尺度になるように実験により予め定めた変換式により、前記裏向き用設定の凹凸度Crを裏向き用設定の補正凹凸度Cr’に変換し、
前記表向き用設定の凹凸度Cf及び前記裏向き用設定の補正凹凸度Cr’の大きい方を凹凸度とする凹凸度算出工程、又は、
前記表向き用設定の凹凸度Cfの分布が前記裏向き用設定の凹凸度Crの分布と同一尺度になるように実験により予め定めた変換式により、前記表向き用設定の凹凸度Cfを表向き用設定の補正凹凸度Cf’に変換し、
前記表向き用設定の補正凹凸度Cf’及び前記裏向き設定の凹凸度Crの大きい方を凹凸度とする凹凸度算出工程と、
前記凹凸度算出工程で算出した前記凹凸度を実験により予め定めた凹凸度閾値と比較し、
前記凹凸度が前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、前記果実は前記形状異常果であると判断し、
前記凹凸度が前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、前記果実は形状正常果であると判断する形状判断工程と、
を含むことを特徴とする表面に凹凸がある柑橘類の果実の選別方法。
This is a sorting method in which 100% of citrus fruits are inspected while being transported by a conveyor, and irregularly shaped fruits with irregularities on the surface are sorted online without contact.
The conveyor transports the fruits, whose front and back are random, in a row without arranging the fruits face up or face down.
An imaging step of irradiating the fruit with linear light from a light source provided at an appropriate position on the conveyor's transport path and photographing the fruit with a camera installed at an angle different from the light source.
From the image data obtained by processing the image taken in the image taking step, a line due to the linear light applied to the fruit is extracted.
The lines are separated by experiments on ostensibly fruits at equal intervals.
Find the angle of the line at each of the separated points.
The degree of unevenness Cf set for ostensibly is calculated by counting the number of times that the absolute value of the angle difference between adjacent points exceeds the predetermined angle difference threshold value in the experiment on the ostensibly fruit.
Divide the line at equal intervals predetermined by experiments on face-down fruits.
Find the angle of the line at each of the separated points.
The degree of unevenness Cr of the face-down setting was calculated by counting the number of times that the absolute value of the angle difference of the adjacent points exceeded the angle difference threshold value predetermined by the experiment on the face-down fruit.
The unevenness Cr of the face-down setting is turned face down by a conversion formula predetermined by an experiment so that the distribution of the unevenness Cr of the face-down setting is on the same scale as the distribution of the unevenness Cf of the face-up setting. Converted to the correction unevenness Cr'of the setting for
The unevenness calculation step in which the larger of the unevenness Cf of the front-facing setting and the corrected unevenness Cr'of the back-facing setting is the unevenness, or
The unevenness Cf of the face-up setting is set to face-up by a conversion formula predetermined by an experiment so that the distribution of the unevenness Cf of the face-up setting becomes the same scale as the distribution of the unevenness Cr of the face-down setting. Converted to the correction unevenness Cf'of
The unevenness calculation step in which the larger of the correction unevenness Cf'of the front-facing setting and the unevenness Cr of the back-facing setting is the unevenness.
The unevenness calculated in the unevenness calculation step is compared with a predetermined unevenness threshold value by an experiment.
When the unevenness is equal to or higher than the unevenness threshold value or larger than the unevenness threshold value, it is determined that the fruit is the abnormal shape fruit.
When the unevenness is smaller than the unevenness threshold value or is equal to or less than the unevenness threshold value, the shape determination step of determining that the fruit is a normal shape fruit and
A method for selecting citrus fruits having an uneven surface, which is characterized by containing.
柑橘類の果実をコンベアで搬送しながら全数検査を行い、表面に凹凸がある形状異常果を非接触でオンライン選別する選別方法であって、
前記コンベアは、前記果実を表向き又は裏向きに整列することなく、表裏がランダムである前記果実を一列に搬送するものであり、
前記コンベアの搬送経路の適宜位置で前記果実を撮影し、所定エリア内におけるヘタの有無を画像処理で識別して、ヘタが有る場合には表向き果実であると判断し、ヘタが無い場合には裏向き果実であると判断する表裏判断工程と、
前記コンベアの搬送経路の適宜位置に設けた光源から前記果実に線状光を照射し、前記光源と異なる角度で設置したカメラで前記果実を撮影する画像撮影工程と、
前記画像撮影工程で撮影した画像を画像処理した画像データから、前記果実に照射された前記線状光による線を抽出し、
前記表裏判断工程で、表向き果実であると判断した場合、
前記線を表向き果実に対する実験により予め定めた等間隔で区切り、
区切られた前記間隔毎の点における前記線の角度を求め、
前記点の隣り合うものの角度差の絶対値が表向き果実に対する実験により予め定めた角度差閾値を超える回数を数えて表向き用設定の凹凸度Cfを算出する凹凸度算出工程と、
前記凹凸度算出工程で算出した前記凹凸度Cfを実験により予め定めた凹凸度閾値と比較し、
前記凹凸度Cfが前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、前記果実は前記形状異常果であると判断し、
前記凹凸度Cfが前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、前記果実は形状正常果であると判断する形状判断工程と、
前記表裏判断工程で、裏向き果実であると判断した場合、
前記線を裏向き果実に対する実験により予め定めた等間隔で区切り、
区切られた前記間隔毎の点における前記線の角度を求め、
前記点の隣り合うものの角度差の絶対値が裏向き果実に対する実験により予め定めた角度差閾値を超える回数を数えて裏向き用設定の凹凸度Crを算出する凹凸度算出工程と、
前記裏向き用設定の凹凸度Crの分布が前記表向き用設定の凹凸度Cfの分布と同一尺度になるように実験により予め定めた変換式により、前記裏向き用設定の凹凸度Crを裏向き用設定の補正凹凸度Cr’に変換し、
裏向き用設定の補正凹凸度Cr’を実験により予め定めた凹凸度閾値と比較し、
前記補正凹凸度Cr’が前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、前記果実は前記形状異常果であると判断し、
前記補正凹凸度Cr’が前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、前記果実は形状正常果であると判断する形状判断工程と、
を含むことを特徴とする表面に凹凸がある柑橘類の果実の選別方法。
This is a sorting method in which 100% of citrus fruits are inspected while being transported by a conveyor, and irregularly shaped fruits with irregularities on the surface are sorted online without contact.
The conveyor transports the fruits, whose front and back are random, in a row without arranging the fruits face up or face down.
The fruit is photographed at an appropriate position on the transport path of the conveyor, and the presence or absence of calyx in a predetermined area is identified by image processing. The front and back judgment process to judge that the fruit is face down,
An imaging step of irradiating the fruit with linear light from a light source provided at an appropriate position on the conveyor's transport path and photographing the fruit with a camera installed at an angle different from the light source.
From the image data obtained by processing the image taken in the image taking step, a line due to the linear light applied to the fruit is extracted.
When it is determined that the fruit is ostensibly in the front and back determination process,
The lines are separated by experiments on ostensibly fruits at equal intervals.
Find the angle of the line at each of the separated points.
The unevenness calculation step of calculating the unevenness Cf of the outward setting by counting the number of times that the absolute value of the angle difference of the adjacent points exceeds the angle difference threshold value predetermined by the experiment on the ostensibly fruit.
The unevenness Cf calculated in the unevenness calculation step is compared with a predetermined unevenness threshold value by an experiment.
When the unevenness Cf is equal to or higher than the unevenness threshold value or larger than the unevenness threshold value, it is determined that the fruit is the abnormal shape fruit.
When the unevenness Cf is smaller than the unevenness threshold value or equal to or lower than the unevenness threshold value, the shape determination step of determining that the fruit is a normal shape fruit and
When it is determined that the fruit is face-down in the front-back judgment step,
Divide the line at equal intervals predetermined by experiments on face-down fruits.
Find the angle of the line at each of the separated points.
The unevenness calculation step of calculating the unevenness Cr of the reverse setting by counting the number of times that the absolute value of the angle difference of the adjacent points exceeds the angle difference threshold value predetermined by the experiment on the face-down fruit.
The unevenness Cr of the face-down setting is turned face down by a conversion formula predetermined by an experiment so that the distribution of the unevenness Cr of the face-down setting is on the same scale as the distribution of the unevenness Cf of the face-up setting. Converted to the correction unevenness Cr'of the setting for
Comparing the correction unevenness Cr'of the face-down setting with the unevenness threshold value predetermined by the experiment,
When the corrected unevenness Cr'is equal to or higher than the unevenness threshold value or larger than the unevenness degree threshold value, it is determined that the fruit is the abnormal shape fruit.
When the corrected unevenness Cr'is smaller than the unevenness threshold value or is equal to or less than the unevenness degree threshold value, the shape determination step of determining that the fruit is a normal shape fruit and the shape determination step.
A method for selecting citrus fruits having an uneven surface, which is characterized by containing.
柑橘類の果実をコンベアで搬送しながら全数検査を行い、表面に凹凸がある形状異常果を非接触でオンライン選別する選別方法であって、
前記コンベアは、前記果実を表向き又は裏向きに整列することなく、表裏がランダムである前記果実を一列に搬送するものであり、
前記コンベアの搬送経路の適宜位置で前記果実を撮影し、所定エリア内におけるヘタの有無を画像処理で識別して、ヘタが有る場合には表向き果実であると判断し、ヘタが無い場合には裏向き果実であると判断する表裏判断工程と、
前記コンベアの搬送経路の適宜位置に設けた光源から前記果実に線状光を照射し、前記光源と異なる角度で設置したカメラで前記果実を撮影する画像撮影工程と、
前記画像撮影工程で撮影した画像を画像処理した画像データから、前記果実に照射された前記線状光による線を抽出し、
前記表裏判断工程で、裏向き果実であると判断した場合、
前記線を裏向き果実に対する実験により予め定めた等間隔で区切り、
区切られた前記間隔毎の点における前記線の角度を求め、
前記点の隣り合うものの角度差の絶対値が裏向き果実に対する実験により予め定めた角度差閾値を超える回数を数えて裏向き用設定の凹凸度Crを算出する凹凸度算出工程と、
前記凹凸度算出工程で算出した前記凹凸度Crを実験により予め定めた凹凸度閾値と比較し、
前記凹凸度Crが前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、前記果実は前記形状異常果であると判断し、
前記凹凸度Crが前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、前記果実は形状正常果であると判断する形状判断工程と、
前記表裏判断工程で、表向き果実であると判断した場合、
前記線を表向き果実に対する実験により予め定めた等間隔で区切り、
区切られた前記間隔毎の点における前記線の角度を求め、
前記点の隣り合うものの角度差の絶対値が表向き果実に対する実験により予め定めた角度差閾値を超える回数を数えて表向き用設定の凹凸度Cfを算出する凹凸度算出工程と、
前記表向き用設定の凹凸度Cfの分布が前記裏向き用設定の凹凸度Crの分布と同一尺度になるように実験により予め定めた変換式により、前記表向き用設定の凹凸度Cfを表向き用設定の補正凹凸度Cf’に変換し、
表向き用設定の補正凹凸度Cf’を実験により予め定めた凹凸度閾値と比較し、
前記補正凹凸度Cf’が前記凹凸度閾値以上である場合又は前記凹凸度閾値よりも大きい場合には、前記果実は前記形状異常果であると判断し、
前記補正凹凸度Cf’が前記凹凸度閾値よりも小さい場合又は前記凹凸度閾値以下である場合には、前記果実は形状正常果であると判断する形状判断工程と、
を含むことを特徴とする表面に凹凸がある柑橘類の果実の選別方法。
This is a sorting method in which 100% of citrus fruits are inspected while being transported by a conveyor, and irregularly shaped fruits with irregularities on the surface are sorted online without contact.
The conveyor transports the fruits, whose front and back are random, in a row without arranging the fruits face up or face down.
The fruit is photographed at an appropriate position on the transport path of the conveyor, and the presence or absence of calyx in a predetermined area is identified by image processing. The front and back judgment process to judge that the fruit is face down,
An imaging step of irradiating the fruit with linear light from a light source provided at an appropriate position on the conveyor's transport path and photographing the fruit with a camera installed at an angle different from the light source.
From the image data obtained by processing the image taken in the image taking step, a line due to the linear light applied to the fruit is extracted.
When it is determined that the fruit is face-down in the front-back judgment step,
Divide the line at equal intervals predetermined by experiments on face-down fruits.
Find the angle of the line at each of the separated points.
The unevenness calculation step of calculating the unevenness Cr of the reverse setting by counting the number of times that the absolute value of the angle difference of the adjacent points exceeds the angle difference threshold value predetermined by the experiment on the face-down fruit.
The unevenness Cr calculated in the unevenness calculation step is compared with a predetermined unevenness threshold value by an experiment.
When the unevenness Cr is equal to or higher than the unevenness threshold value or larger than the unevenness threshold value, it is determined that the fruit is the abnormal shape fruit.
When the unevenness Cr is smaller than the unevenness threshold value or equal to or less than the unevenness threshold value, the shape determination step of determining that the fruit is a normal shape fruit and
When it is determined that the fruit is ostensibly in the front and back determination process,
The lines are separated by experiments on ostensibly fruits at equal intervals.
Find the angle of the line at each of the separated points.
The unevenness calculation step of calculating the unevenness Cf of the outward setting by counting the number of times that the absolute value of the angle difference of the adjacent points exceeds the angle difference threshold value predetermined by the experiment on the ostensibly fruit.
The unevenness Cf of the face-up setting is set to face-up by a conversion formula predetermined by an experiment so that the distribution of the unevenness Cf of the face-up setting becomes the same scale as the distribution of the unevenness Cr of the face-down setting. Converted to the correction unevenness Cf'of
Comparing the corrected unevenness Cf'of the face-up setting with the unevenness threshold value predetermined by the experiment,
When the corrected unevenness Cf'is equal to or higher than the unevenness threshold value or larger than the unevenness degree threshold value, it is determined that the fruit is the abnormal shape fruit.
When the corrected unevenness Cf'is smaller than the unevenness threshold value or is equal to or less than the unevenness degree threshold value, the shape determination step of determining that the fruit is a normal shape fruit and the shape determination step.
A method for selecting citrus fruits having an uneven surface, which is characterized by containing.
前記画像撮影工程で撮影した画像を画像処理した画像データから前記線を抽出する際に、
離間している2本の抽出線について、
(a)前記2本の抽出線の端点間の距離が実験により予め定めた閾値よりも近いこと、
(b)前記2本の抽出線の端点の方向差が実験により予め定めた閾値よりも小さいこと、
(c)前記2本の抽出線の端点の垂直方向の距離が実験により予め定めた閾値よりも小さいこと、
の(a)ないし(c)の条件を全て満たす場合、
前記2本の抽出線の端点間を補間して前記2本の抽出線を接続する、
請求項1〜の何れか1項に記載の表面に凹凸がある柑橘類の果実の選別方法。
When extracting the line images captured by the image capturing process from the image data obtained by image processing,
About two separated extraction lines
(A) The distance between the end points of the two extraction lines is closer than the threshold value predetermined by the experiment.
(B) The direction difference between the end points of the two extraction lines is smaller than the threshold value predetermined by the experiment.
(C) The vertical distance between the end points of the two extraction lines is smaller than the threshold value predetermined by the experiment.
When all the conditions (a) to (c) are satisfied,
Interpolate between the endpoints of the two extraction lines to connect the two extraction lines.
The method for selecting citrus fruits having irregularities on the surface according to any one of claims 1 to 3.
JP2017168408A 2017-09-01 2017-09-01 How to sort citrus fruits with irregularities on the surface Active JP6926846B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2017168408A JP6926846B2 (en) 2017-09-01 2017-09-01 How to sort citrus fruits with irregularities on the surface

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2017168408A JP6926846B2 (en) 2017-09-01 2017-09-01 How to sort citrus fruits with irregularities on the surface

Publications (2)

Publication Number Publication Date
JP2019045292A JP2019045292A (en) 2019-03-22
JP6926846B2 true JP6926846B2 (en) 2021-08-25

Family

ID=65816376

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2017168408A Active JP6926846B2 (en) 2017-09-01 2017-09-01 How to sort citrus fruits with irregularities on the surface

Country Status (1)

Country Link
JP (1) JP6926846B2 (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113182202B (en) * 2021-04-25 2022-05-27 新疆维吾尔自治区阿克苏职业技术学院 Categorised auxiliary device of fragrant pear sex

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4005228B2 (en) * 1998-07-27 2007-11-07 ペンタックス株式会社 Optical member inspection apparatus, image processing apparatus, and computer-readable medium
JP2006162393A (en) * 2004-12-06 2006-06-22 Mitsui Mining & Smelting Co Ltd Fruit and vegetable shape sorting apparatus and shape sorting method
EP2167408B1 (en) * 2007-06-06 2012-11-14 Visiomatics NV System for the automatic determination of the position and orientation of a fruit
JP4835713B2 (en) * 2009-03-24 2011-12-14 カシオ計算機株式会社 Image processing apparatus and computer program
JP5351673B2 (en) * 2009-09-09 2013-11-27 パナソニック株式会社 Appearance inspection device, appearance inspection method

Also Published As

Publication number Publication date
JP2019045292A (en) 2019-03-22

Similar Documents

Publication Publication Date Title
US8284248B2 (en) Method for real time detection of defects in a food product
CN107533012A (en) Melt the surface defect inspection apparatus and surface defect inspection method of coated steel sheet
KR20080104000A (en) Optical measuring device and measuring method of grain grain
EP2135048A2 (en) Method and system for inspecting manufactured parts and sorting the inspected parts
US20180333752A1 (en) Imaging system for sorting poultry meat with wooden breast myopathy
CN111602047B (en) Tablet inspection method and tablet inspection device
JP2015203586A (en) Inspection method
JP2015114226A (en) Visual inspection device and visual inspection method
JP2012251983A (en) Wrap film wrinkle inspection method and device
JP6926846B2 (en) How to sort citrus fruits with irregularities on the surface
KR101096790B1 (en) Agricultural product volume measuring device using multi channel camera
AU2008243688B2 (en) Method and apparatus for three dimensional image processing and analysis
JP2016075660A (en) Boiled egg quality judgment device
JP5004478B2 (en) Object discrimination method and discrimination device
JP6997495B1 (en) Pistachio sorting device
JP7003786B2 (en) Particle size measuring device and particle size measuring method
JP2009074967A (en) Wood inspection method, apparatus and program
CN118096779B (en) Method and device for measuring shaddock fruit edibility based on machine vision and X-ray imaging
JP2006162393A (en) Fruit and vegetable shape sorting apparatus and shape sorting method
CN102215704A (en) Method for assigning a stonefruit to a predetermined class and a device therefor
JP2001239221A (en) Image forming device and assorting/classifying/judging device used for assorting of agricultural product
US20210383145A1 (en) A food processing device and a method of providing images of food objects in a food processing device
WO2018217077A1 (en) Method and apparatus for detecting imperfections in glass products
KR20160059623A (en) sorting method for grains
JP2011149699A (en) Method and device for determining surface quality of steel plate

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20200512

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20210325

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20210406

A521 Request for written amendment filed

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20210506

TRDD Decision of grant or rejection written
A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

Effective date: 20210706

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20210719

R150 Certificate of patent or registration of utility model

Ref document number: 6926846

Country of ref document: JP

Free format text: JAPANESE INTERMEDIATE CODE: R150

R250 Receipt of annual fees

Free format text: JAPANESE INTERMEDIATE CODE: R250