KR100290764B1 - Method for selecting a apple with a shape and color - Google Patents
Method for selecting a apple with a shape and color Download PDFInfo
- Publication number
- KR100290764B1 KR100290764B1 KR1019990008846A KR19990008846A KR100290764B1 KR 100290764 B1 KR100290764 B1 KR 100290764B1 KR 1019990008846 A KR1019990008846 A KR 1019990008846A KR 19990008846 A KR19990008846 A KR 19990008846A KR 100290764 B1 KR100290764 B1 KR 100290764B1
- Authority
- KR
- South Korea
- Prior art keywords
- fruit
- color
- value
- grade
- camera
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 235000013399 edible fruits Nutrition 0.000 claims abstract description 165
- 238000004040 coloring Methods 0.000 claims abstract description 15
- 238000012216 screening Methods 0.000 claims abstract description 8
- 230000036244 malformation Effects 0.000 claims abstract description 5
- 230000001186 cumulative effect Effects 0.000 claims description 8
- 238000013528 artificial neural network Methods 0.000 claims description 5
- 230000005856 abnormality Effects 0.000 claims description 2
- 238000006243 chemical reaction Methods 0.000 claims 1
- 238000001514 detection method Methods 0.000 claims 1
- 239000000123 paper Substances 0.000 description 16
- 239000003086 colorant Substances 0.000 description 1
- 239000011087 paperboard Substances 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
- G01J3/46—Measurement of colour; Colour measuring devices, e.g. colorimeters
- G01J3/52—Measurement of colour; Colour measuring devices, e.g. colorimeters using colour charts
- G01J3/522—Measurement of colour; Colour measuring devices, e.g. colorimeters using colour charts circular colour charts
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/04—Sorting according to size
- B07C5/08—Sorting according to size measured electrically or electronically
- B07C5/083—Mechanically moved sensing devices, the movement of which influences the electric contact members
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01J—MEASUREMENT OF INTENSITY, VELOCITY, SPECTRAL CONTENT, POLARISATION, PHASE OR PULSE CHARACTERISTICS OF INFRARED, VISIBLE OR ULTRAVIOLET LIGHT; COLORIMETRY; RADIATION PYROMETRY
- G01J3/00—Spectrometry; Spectrophotometry; Monochromators; Measuring colours
- G01J3/46—Measurement of colour; Colour measuring devices, e.g. colorimeters
- G01J3/463—Colour matching
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/02—Food
- G01N33/025—Fruits or vegetables
Landscapes
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Food Science & Technology (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Analytical Chemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Medicinal Chemistry (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Sorting Of Articles (AREA)
- Image Analysis (AREA)
Abstract
과일을 자동으로 선별하는 자동선별장치의 이송장치에 놓여진 과일을 카메라를 이용하여 촬영하여 과일의 색상과 과일의 외형상태를 판별하여 과일의 상품등급을 자동으로 선별할 수 있도록 하는 과일의 색상 및 외형상태 선별방법이 개시되어 있다. 먼저, 선별에 들어가기 앞서 표준색지에 의해 카메라의 RGB신호를 보정한다. 이어서 과일선별장치의 과일접시가 센서에 의해 감지되면 카메라를 이용하여 과일접시에 놓여진 과일을 촬영하고, 카메라에 의해 촬영된 과일의 화상신호로부터 과일의 색상신호인 RGB신호를 산출해낸다. 상기 RGB신호를 표준색지와 비교하여 RGB신호를 보정한다. 상기 RGB신호를 보정하여 산출된 이론적인 NTSC RGB를 얻고, 이 신호를 바탕으로 메모리부에 저장되어 있는 L*a*b좌표계의 a*값을 얻는다. 상기 L*a*b좌표계의 a*값이 착색화소인지를 판정한다. 이와 같은 작업을 모든 사과화소에 대해 수행하여 과일의 내부의 총 착색화소수를 구하여 착색비율을 얻고 과일의 등급을 결정하게 된다. 또한, 과일의 화상신호로부터 장축과 단축의 길이를 측정하여 장축길이와 단축길이의 비율이 설정된 값보다 크게되면 기형과일로 판단하게 되므로 자동으로 신속하고 정확하게 과일의 색상등급 및 기형여부를 판단하여 과일의 품질등급을 결정한다.The color of the fruit and the appearance of the fruit can be automatically selected by determining the color of the fruit and the appearance of the fruit by capturing the fruit placed in the transfer device of the automatic sorting device that automatically sorts the fruit. A state screening method is disclosed. First, the RGB signal of the camera is corrected by the standard color paper prior to the selection. Subsequently, when the fruit plate of the fruit sorting device is detected by the sensor, the fruit placed on the fruit plate is photographed using a camera, and an RGB signal which is a color signal of the fruit is calculated from the image signal of the fruit photographed by the camera. The RGB signal is corrected by comparing the RGB signal with standard color paper. The theoretical NTSC RGB calculated by correcting the RGB signal is obtained, and the a * value of the L * a * b coordinate system stored in the memory unit is obtained based on this signal. It is determined whether the a * value of the L * a * b coordinate system is a colored pixel. This operation is performed for all apple pixels to obtain the total number of colored pixels inside the fruit to obtain a coloring ratio and to determine the grade of the fruit. In addition, by measuring the length of the long axis and short axis from the image signal of the fruit, if the ratio of the long axis length and short axis length is larger than the set value, it is judged as malformed fruit, so it is automatically and quickly and accurately judges the color grade and malformation of the fruit. Determine the quality grade of
Description
본 발명은 과일 선별방법에 관한 것으로, 보다 상세하게는 과일을 자동으로 선별하는 자동선별장치의 이송장치의 접시위에 놓여진 과일을 카메라를 이용하여 촬영하여 과일의 색상과 과일의 외형상태를 판별하여 과일의 상품등급을 자동으로 부여할 수 있도록 하는 과일의 색상 및 외형상태 선별방법에 관한 것이다.The present invention relates to a fruit sorting method, and more particularly, by photographing the fruit placed on the plate of the transfer device of the automatic sorting device for automatically sorting the fruit using a camera to determine the fruit color and the appearance of the fruit The present invention relates to a method of sorting fruit color and appearance so that the product grade can be automatically assigned.
일반적으로 과일선별장치는 과일을 중량별로 분류하도록 된 중량식 과일 선별장치가 많이 선택되어 사용되고 있다. 상기 과일선별장치의 과일접시에 자동으로 과일이 적재되도록 과일접시는 과일적재대에 인접하도록 이동을 한다. 상기 과일접시가 과일적재대에 인접하면서 이동하는 중에 과일적재대에 있는 과일이 자동으로 과일접시에 놓여지게 되며, 상기 과일접시에 과일이 놓여진 상태에서 과일접시는 이동하여 과일의 무게를 재는 스프링저울의 원리로 과일을 선별한다. 이와같은 중량식 과일선별장치를 이용하여 과일을 선별하면 과일의 색상 및 외형상태는 판별할 수 없으므로, 과일의 상품성 판단에 큰 기준이 되는 색상 및 외형상태는 사람이 직접 판별해야 하는 불편함이 있다.In general, the fruit sorting device has been selected and used a lot of weight-type fruit sorting device to sort the fruit by weight. The fruit plate moves so as to be adjacent to the fruit loading table so that the fruit is automatically loaded on the fruit plate of the fruit sorting device. While the fruit plate is moving adjacent to the fruit loading zone, the fruit in the fruit loading zone is automatically placed on the fruit plate, and in the state where the fruit is placed on the fruit plate, the spring balance moves to weigh the fruit. Selecting fruit on the principle of When the fruit is selected using such a weighted fruit sorting device, the color and appearance of the fruit cannot be discriminated. Therefore, the color and the appearance of the fruit are inconvenient to be directly determined. .
본 발명은 상기와 같은 문제점을 해결하기 위하여 안출된 것으로, 본 발명의 목적은 이송장치의 과일접시에 놓여진 과일을 카메라로 촬영하여 과일의 화상신호를 분석하여 과일의 색상과 과일의 외형상태를 판별하여 과일의 상품등급을 자동으로 선별할 수 있도록 하는 과일의 색상 및 외형상태 선별방법을 제공하는데 있다.The present invention has been made to solve the above problems, an object of the present invention is to take a picture of the fruit placed on the fruit plate of the transfer device with a camera to analyze the image signal of the fruit to determine the color of the fruit and the appearance of the fruit The present invention provides a method of sorting fruit color and appearance so that the product grade of fruit can be automatically selected.
도 1은 본 발명에 따른 과일의 색상 및 외형상태 선별방법을 개략적으로 나타낸 흐름도이다.1 is a flow chart schematically showing a color and appearance of the fruit sorting method according to the invention.
도 2는 도 1의 제2단계의 동작을 상세히 나타낸 흐름도이다.FIG. 2 is a flowchart illustrating the operation of the second step of FIG. 1 in detail.
도 3은 도 1의 제3단계의 동작을 상세히 나타낸 흐름도이다.3 is a flowchart illustrating the operation of the third step of FIG. 1 in detail.
도 4는 도 1의 제4단계의 동작을 상세히 나타낸 흐름도이다.4 is a flowchart illustrating an operation of the fourth step of FIG. 1 in detail.
도 5는 도 1의 제5단계의 동작을 상세히 나타낸 흐름도이다.5 is a flowchart illustrating an operation of a fifth step of FIG. 1 in detail.
상기 목적을 달성하기 위하여 본 발명은, 과일의 색상등급을 비교판단할 기준 색좌표인 L*a*b*값(이중에서 a*값)을 측정하기 위한 기준설정단계; 카메라에 의해 촬영된 표준색지의 화상신호로부터 기준색좌표인 L*a*b*좌표의 a*값을 컴퓨터에 저장시키는 저장단계; 자동선별장치의 과일접시에 놓여진 과일을 카메라에 의해 촬영하고 입력된 화상신호를 분석하여 과일이 과일접시에 놓여져 있는지를 판단하는 감지단계; 상기 과일접시에 놓여진 과일의 화상신호를 기준색좌표인 L*a*b*좌표의 a*값으로 변환한 후 사과의 총 착색화소수를 측정하여 과일의 색상등급(1등급, 2등급, 3등급 그리고 등외등급)을 판별하는 색상등급판별단계; 카메라로부터 인가되는 화상신호를 체인코딩기법을 이용하여 과일의 장축길이와 단축길이를 산출하여 과일의 외형상태가 기형인가를 판단하는 외형등급판단단계; 그리고 과일의 색상등급과 기형여부를 비교판단하여 과일의 품질등급을 결정하는 품질등급판단단계를 포함하는 과일의 색상 및 외형상태 선별방법을 제공한다.In order to achieve the above object, the present invention, a reference setting step for measuring the L * a * b * value (double a * value) which is a reference color coordinate to determine the color grade of the fruit; A storage step of storing, on a computer, the a * value of the L * a * b * coordinate as the reference color coordinate from the image signal of the standard color paper photographed by the camera; A sensing step of photographing the fruit placed on the fruit dish of the automatic sorting apparatus by a camera and analyzing the input image signal to determine whether the fruit is placed on the fruit dish; After converting the image signal of the fruit placed on the fruit plate to the a * value of the L * a * b * coordinate, which is the reference color coordinate, the total color pixel number of the apple is measured and the fruit's color grade And a color grading step of judging an outlier grade; An appearance grade judging step of judging whether the appearance state of the fruit is deformed by calculating the long axis length and the short axis length of the fruit using a chain coding method on the image signal applied from the camera; In addition, the present invention provides a method for screening the color and appearance of the fruit, including a quality grade determining step of determining the fruit's quality grade by comparing the fruit's color grade and abnormality.
본 발명에 따르면, 먼저 카메라를 이용하여 과일의 색 측정을 위해 표준색지를 사용하여 카메라의 기준색도값을 설정한 후, 센서에 의해 과일접시의 위치가 감지되면 카메라를 이용하여 과일접시에 놓여진 과일을 촬영하고, 카메라에 의해 촬영된 과일의 화상신호로부터 과일의 색상신호인 RGB신호를 산출해낸다. 상기 RGB신호를 보정하여 이론적인 NTSC RGB신호를 얻는다. 이 신호를 바탕으로 메모리부에 저장되어 있는 L*a*b*좌표계의 a*값을 얻고, 이 a*값이 착색화소인지를 판정한다. 이와 같은 작업을 모든 사과화소에 대해 수행하여 과일의 내부의 총 착색화소수를 구하여 착색비율을 얻고 과일의 등급을 결정하게 된다. 또한, 과일의 화상신호로부터 장축과 단축의 길이를 측정하여 장축길이와 단축길이의 비율이 설정된 값보다 크게되면 기형과일로 판단하게 되므로 자동으로 신속하고 정확하게 과일의 색상등급 및 기형여부를 판단하여 과일의 품질등급을 결정한다.According to the present invention, first, after setting the reference chromaticity value of the camera using a standard color paper for measuring the color of the fruit using the camera, if the position of the fruit plate is detected by the sensor, the fruit placed on the fruit plate using the camera The image is taken by the camera and the RGB signal which is the color signal of the fruit is calculated from the image signal of the fruit. The RGB signal is corrected to obtain a theoretical NTSC RGB signal. Based on this signal, the a * value of the L * a * b * coordinate system stored in the memory unit is obtained, and it is determined whether this a * value is a colored pixel. This operation is performed for all apple pixels to obtain the total number of colored pixels inside the fruit to obtain a coloring ratio and to determine the grade of the fruit. In addition, by measuring the length of the long axis and short axis from the image signal of the fruit, if the ratio of the long axis length and short axis length is larger than the set value, it is judged as malformed fruit, so it is automatically and quickly and accurately judges the color grade and malformation of the fruit. Determine the quality grade of
이하, 첨부된 도면을 참조하여 본 발명을 설명하면 다음과 같다.Hereinafter, the present invention will be described with reference to the accompanying drawings.
도 1은 본 발명에 따른 과일의 색상 및 외형상태 선별방법을 개략적으로 나타낸 흐름도이고, 도 2는 도 1의 제2단계의 동작을 상세히 나타낸 흐름도이고, 도 3은 도 1의 제3단계의 동작을 상세히 나타낸 흐름도이고, 도 4는 도 1의 제4단계의 동작을 상세히 나타낸 흐름도이다.1 is a flow chart schematically showing the color and appearance of the fruit sorting method according to the invention, Figure 2 is a flow chart illustrating in detail the operation of the second step of Figure 1, Figure 3 is the operation of the third step of Figure 1 4 is a flowchart illustrating the operation of the fourth step of FIG. 1 in detail.
도 1 내지 도 4를 참조하여 과일의 색상 및 외형상태 선별방법을 설명하면, 먼저, 과일의 색상 및 기형상태에 따른 품질등급을 결정하여 자동으로 선별하는 자동선별장치(도시 안됨)를 과일의 색상등급을 비교판단할 기준 색좌표인 L*a*b*좌표중의 a*값을 설정하기 위한 모드로 전환(S100)시킨다. 카메라에 의해 촬영된 표준색지(도시 안됨)의 화상신호로부터 기준색좌표인 L*a*b*좌표의 a*값을 설정하여 컴퓨터(도시 안됨)에 저장시키게 된다(S200). 자동판별장치의 과일접시(도시 안됨)에 놓여진 과일을 촬영하는 카메라(도시 안됨)에 의해 촬영되는 화상신호를 분석하여 과일이 과일접시에 놓여져 있는지를 판단(S300)하게 된다. 상기 과일접시에 놓여진 과일의 화상신호를 기준색좌표인 L*a*b*좌표의 a*값과 비교하여 과일의 색상등급(1등급 내지 3등급, 그리고 등외등급)을 판별하게 된다(S400). 카메라로부터 인가되는 화상신호를 체인코딩기법을 이용하여 과일의 장축길이와 단축길이를 산출하여 과일의 외형상태가 기형인가를 판단하게 된다(S500). 과일의 색상등급과 기형여부를 비교판단하여 과일의 품질등급을 결정하게 된다(S600).1 to 4, the fruit color and appearance state selection method will be described. First, an automatic screening device (not shown) that automatically selects the quality grade according to the fruit color and malformation state (not shown) to color the fruit. In operation S100, the mode is set to a mode for setting the a * value of the L * a * b * coordinates, which are the reference color coordinates to be compared. The a * value of the L * a * b * coordinate, which is the reference color coordinate, is set from the image signal of the standard color paper (not shown) photographed by the camera and stored in the computer (not shown) (S200). The image signal photographed by the camera (not shown) photographing the fruit placed on the fruit plate (not shown) of the automatic discriminating device is analyzed to determine whether the fruit is placed on the fruit plate (S300). The image signal of the fruit placed on the fruit plate is compared with the a * value of the L * a * b * coordinate, which is the reference color coordinate, to determine the color grade (grades 1 to 3, and isotropic grade) of the fruit (S400). The long-axis length and short-length length of the fruit are calculated by using the chain coding method of the image signal applied from the camera to determine whether the fruit is malformed (S500). The quality grade of the fruit is determined by comparing the fruit's color grade and deformity (S600).
좀더 상세하게 설명하면, 상기 저장단계(S200)에서는 다수개의 색지로 구성된 표준색지판(예를 들면 : 96개의 색지를 하나의 판에 구성시킴)을 이루는 각각의 표준색지를 측색계(colorimeter)를 이용하여 촬영하고, 상기 측색계에 의해 촬영되어 얻어진 각각의 표준색지가 갖는 색도값을 모두 컴퓨터(도시 안됨)에 저장한다(S210)(이때 표준색지판의 빨간색 색지가 255의 색도값을 갖을 수 있고, 검은색은 0 의 색도값을 갖을 수 있다).In more detail, in the storage step (S200), each standard color paper constituting a standard color paper plate (for example, 96 color papers is configured on one plate) consisting of a plurality of color papers using a colorimeter. Photographing, and stores all the chromaticity values of each of the standard color paper obtained by the colorimeter on the computer (not shown) (S210) (at this time, the red color paper of the standard color board may have a chromaticity value of 255, black May have a chromaticity value of 0).
상기 96개의 표준색지를 표준색지판에 구성시킨 후 카메라를 이용하여 촬영한다. 상기 카메라에 의해 촬영된 표준색지판의 화상신호로부터 각 표준색지의 기준색도값을 산출해낸다(이때 상기 카메라에 의해 촬영된 표준색지의 빨간색 색지가 250의 색도값을 갖을 수 있고, 검은색은 2의 색도값을 갖을 수 있다).The 96 standard color papers are constructed on a standard color board and photographed using a camera. The reference chromaticity value of each standard color paper is calculated from the image signal of the standard color paper plate photographed by the camera (at this time, the red color paper of the standard color paper photographed by the camera may have a chromaticity value of 250, and black is 2 May have chromaticity values).
상기 카메라에 의해 촬영된 표준색지판의 화상신호로부터 얻어진 각 표준색지의 색도값을 컴퓨터에 입력시킨다(S220). 상기 컴퓨터는 카메라에 의해 촬영된 화상신호로부터 얻어진 표준색지의 기준색도값과 저장된 측색계에 의해 얻어진 표준색도값을 비교하여 기준색도값이 표준색도값을 갖도록 보정한다(S230). 따라서, 컴퓨터는 카메라로부터 입력되는 빨간색의 기준색도값이 250일 때 상기 기준색도값 250은 255로 간주를 하게 된다. 또한, 상기 컴퓨터는 카메라로부터 입력되는 빨간색의 기준색도값이 250보다 낮은 값을 갖게 되면 신경회로망의 학습에 의해 자동으로 보정되어 255보다 낮은 임의의 값을 갖게 된다(S240)(예를 들면, 카메라로부터 입력되는 빨간색의 기준색도값이 250보다 낮은 245의 값을 갖게 되면 신경회로망의 학습에 의해 자동으로 보정되어 255보다 낮은 250의 값을 갖게 된다).A chromaticity value of each standard color paper obtained from the image signal of the standard color paper board photographed by the camera is input to the computer (S220). The computer compares the reference chromaticity value of the standard color paper obtained from the image signal photographed by the camera with the standard chromaticity value obtained by the stored colorimeter to correct the reference chromaticity value to have the standard chromaticity value (S230). Thus, when the red reference chromaticity value input from the camera is 250, the reference chromaticity value 250 is regarded as 255. In addition, when the reference color value of red input from the camera has a value lower than 250, the computer is automatically corrected by learning of the neural network to have an arbitrary value lower than 255 (S240) (for example, the camera). When the reference chromaticity value of red input from 245 is lower than 250, the value of 245 is automatically corrected by the neural network learning and has a value of 250 lower than 255).
상기 신경회로망의 학습에 의해 보정된 각 색도값은 이론적인 NTSC색도값에 해당하며, 상기 NTSC색도값을 색변환공식(color coordinate system)에 적용함으로써 색도값에 해당하는 참조표(LUT : Look up table)(도시 안됨)를 만들어 컴퓨터에 저장한다(S250). 상기 컴퓨터에 저장된 참조표로부터 L*a*b*좌표의 a*값(Red 및 Green에 해당하는 색)을 산출하여 컴퓨터에 저장시킨다(S260). 상기 L*a*b*좌표계의 L*은 밝기에 관계된 값을 나타내고, a*는 Red 및 Green에 관계된 값을 나타내며, b*는 blue 및 yellow에 관계된 값을 나타낸다.Each chromaticity value corrected by the learning of the neural network corresponds to a theoretical NTSC chromaticity value, and applies the NTSC chromaticity value to a color coordinate system (LUT: Look up) table) (not shown) is created and stored in the computer (S250). The a * value (color corresponding to Red and Green) of L * a * b * coordinates is calculated from the reference table stored in the computer and stored in the computer (S260). L * of the L * a * b * coordinate system indicates a value related to brightness, a * indicates a value related to Red and Green, and b * indicates a value related to blue and yellow.
상기 감지단계(S300)에서는 컴퓨터는 카메라에 의해 촬영된 화면의 화상신호가 입력되면 화상신호인 화면의 중앙상단에서부터 하단으로 빨간색(Red)에 해당하는 화면의 화소값을 스캐닝(scanning)한다(S310). 상기 화면의 중앙상단에서부터 스캐닝한 빨간색에 해당하는 화소값이 설정값(threshold)보다 큰가를 비교판단한다(S320). 만약에 화면의 중앙상단에서부터 스캐닝한 빨간색에 해당하는 화소값이 설정값보다 크면 누적화소값을 1증가시킨다(S330). 상기 화면의 중앙상단에서부터 하단으로 스캐닝하여 스캐닝 위치가 화면의 중앙에 해당하면 누적화소값이 설정값(예; 20)보다 큰가를 비교판단한다(S340). 상기 누적화소값이 설정값보다 크면 과일접시위에 과일(예; 사과)이 놓여진 것으로 판정(S350)하고, 반대로 누적화소값이 설정값보다 작으면 과일접시위에 과일이 없는 것으로 판정한다(S360).In the detecting step (S300), when the image signal of the screen photographed by the camera is input, the computer scans the pixel value of the screen corresponding to the red (Red) from the top of the center of the screen that is the image signal (S310) (S310) ). In operation S320, it is determined whether the pixel value corresponding to the red color scanned from the upper center of the screen is larger than a threshold. If the pixel value corresponding to the red color scanned from the upper center of the screen is larger than the set value, the cumulative pixel value is increased by one (S330). When the scanning position corresponds to the center of the screen by scanning from the top of the center to the bottom of the screen, it is determined whether the cumulative pixel value is larger than the set value (eg, 20) (S340). If the cumulative pixel value is larger than the set value, it is determined that the fruit (for example, an apple) is placed on the fruit plate (S350). On the contrary, if the cumulative pixel value is smaller than the set value, it is determined that there is no fruit on the fruit plate (S360).
상기 색상등급판별단계(S400)에서는 카메라로부터 입력되는 화상신호의 하나의 화소에 대한 색도값을 산출(S410)하여 저장된 참조표(LUT)(도시 안됨)에서 해당하는 L*a*b*좌표계의 a*값을 구한 후 설정값과 L*a*b*좌표계의 a*값의 크기를 비교판단한다(S420).In the color grading step S400, a chromaticity value of one pixel of an image signal input from a camera is calculated (S410), and a corresponding L * a * b * coordinate system is stored in a stored reference table (LUT) (not shown). After the a * value is obtained, a comparison between the set value and the a * value of the L * a * b * coordinate system is made (S420).
만약에 L*a*b*좌표계의 a*값이 설정값보다 크게되면 착색화소수를 1증가시키고, L*a*b*좌표계의 a*값이 설정값보다 작으면 착색화소수를 증가시키지 않는다(S430).If the a * value of the L * a * b * coordinate system is larger than the set value, the number of colored pixels is increased by one. If the a * value of the L * a * b * coordinate system is smaller than the set value, the color pixel number is not increased. Do not (S430).
상기 L*a*b*좌표계의 a*값이 설정값보다 크면 착색화소수를 1증가시킨 후 과일의 화상신호의 모든 화소에 대해 착색정도를 검사하였는가를 판단한다(S440).If the a * value of the L * a * b * coordinate system is larger than the set value, it is determined whether or not the degree of coloring is examined for all pixels of the image signal of the fruit after increasing the number of colored pixels by one (S440).
상기 과일의 화상신호의 모든 화소에 대해 착색정도를 검사하였으면 착색화소수와 화소수를 비교하여 화소수에 대한 착색화소수의 비율에 해당하는 착색비율을 산출해낸다(S450).When the degree of coloring is examined for all the pixels of the image signal of the fruit, the number of colors is compared with the number of pixels to calculate a color ratio corresponding to the ratio of the number of pixels to the number of pixels (S450).
상기 착색비율이 제1설정값보다 큰가를 판단(S460)하여 착색비율이 제1설정값보다 크면 색1등급으로 결정하고(S465), 착색비율이 제1설정값보다 작으면 제2설정값보다 큰가를 판단(S470)하여 착색비율이 제2설정값보다 크면 색2등급으로 결정하고(S475), 착색비율이 제2설정값보다 작으면 제3설정값보다 큰가를 판단(S480)하여 착색비율이 제3설정값보다 크면 색3등급으로 결정하고(S485), 착색비율이 제3설정값보다 작으면 등외등급으로 결정(S490)한다.It is determined whether the coloring ratio is greater than the first set value (S460). If the coloring ratio is larger than the first set value, the color ratio is determined as color first grade (S465). If the color ratio is greater than the second set value (S470) to determine whether it is larger than the second set value (S475), if the color ratio is smaller than the second set value is determined to be greater than the third set value (S480) If it is larger than the third set value, it is determined as color third grade (S485). If the coloring ratio is smaller than the third set value, it is determined as isotropic grade (S490).
상기 외형등급판단단계(S500)에서는 카메라로부터 인가되는 과일의 화상신호중에서 과일의 외곽선에 해당하는 하나의 화소를 선택한다(S510). 선택된 화소를 기준으로 반시계방향으로 이동을 하면서 체인코딩기법을 이용하여 과일의 외곽선을 표시하게 된다(S520). 상기 체인코딩기법을 이용하여 과일의 외곽선을 표시한 후 최초 시작위치인가를 판단(S530)하여 최초시작위치가 아니면 과일의 외곽선을 표시하는 단계(S520)를 반복실행하고, 화상신호의 화소가 최초 시작위치이면 외곽선을 기준으로 과일의 중심을 계산하여 결정한다(S540). 상기 외곽선의 일측점들로부터 과일의 중심을 통과하는 다수개의 직선을 만든후 각각의 직선의 길이를 계산한다(S550). 상기 과일의 중심을 지나는 직선들의 길이를 산출하여 가장 긴 직선인 장축길이와 가장 짧은 직선인 단축길이를 비교하여 장단축비를 결정(S560)한다. 상기 장축길이와 단축길이의 비에 해당하는 장단축비가 설정된 값보다 큰가를 비교판단(S570)하여 장단축비가 설정된 값보다 크면 기형과일로 판단(S580)하고, 장단축비가 설정된 값보다 작으면 정상적인 과일로 판단한다(S560).In the appearance grade determining step S500, one pixel corresponding to the outline of the fruit is selected from the image signals of the fruit applied from the camera (S510). While moving in a counterclockwise direction based on the selected pixel, the fruit outline is displayed using a chain coding method (S520). After displaying the outline of the fruit using the chain coding technique, it is determined whether it is the first starting position (S530), and repeatedly displaying the outline of the fruit if it is not the initial starting position (S520). If the starting position is determined by calculating the center of the fruit on the basis of the outline (S540). After making a plurality of straight lines passing through the center of the fruit from one side points of the outline is calculated the length of each straight line (S550). The length of the straight line passing through the center of the fruit is calculated to compare the long axis length, which is the longest straight line, and the short axis length, which is the shortest straight line, to determine a long and short axis ratio (S560). The comparison between the long and short axis ratios corresponding to the ratio between the long and short axis lengths is greater than the set value (S570). If the long and short axis ratios are larger than the set value, the mold is determined as malformed fruit (S580). Judging by the fruit (S560).
상기 품질등급판단단계(S600)에서는 색상등급판별단계(S400)와 외형등급판단단계(S500)를 비교하여 과일의 품질등급을 판정한다. 즉, 상기 색상등급판별단계(S400)에서 산출되는 과일의 색등급(색1등급, 색2등급, 색3등급 및 등외등급)과 외형등급판단단계(S500)에서 산출되는 과일의 외형상태(기형과일 및 정상과일)를 혼합비교함으로써 과일의 품질등급을 판정한다. 예를 들면, 과일의 색등급이 색1등급이면서 외형이 정상이면 특상품으로 판정하고, 과일의 색등급이 색2등급이면서 외형이 정상이면 상품으로 판정하며, 과일의 색등급이 색1등급이면서 외형이 기형과일이면 중품으로 판정하게 된다.In the quality grade judging step (S600), the quality grade of the fruit is determined by comparing the color judging step (S400) and the appearance grade judging step (S500). That is, the color grade (color 1 grade, color 2 grade, color 3 grade, and outlier grade) of the fruit calculated in the color grading step (S400) and the appearance state of the fruit calculated in the shape grading step (S500) (deformation) The quality grade of the fruit is determined by mixing and comparing fruits and normal fruits. For example, if the fruit's color grade is color 1 and the appearance is normal, the product is judged as a special product. If the fruit's color grade is color 2 and the appearance is normal, the product is judged as a commodity. If this malformed fruit is judged as intermediate goods.
이상에서 설명한 바와같이, 본 발명은 과일선별장치의 과일접시에 과일이 놓여진 것이 감지되면 카메라를 이용하여 과일접시에 놓여진 과일을 촬영하고, 카메라에 의해 촬영된 과일의 화상신호로부터 과일의 색상신호인 RGB신호를 산출해낸다. 상기 RGB신호를 보정하여 산출된 NTSC RGB를 이용하여 L*a*b*좌표계의 a*값을 얻는다. 상기 L*a*b*좌표계의 a*값을 설정된 착색화소의 a*값과 비교하여 과일의 착색비율을 산출하여 과일의 등급을 결정하게 된다. 또한, 과일의 화상신호로부터 장축과 단축의 길이를 측정하여 장축길이와 단축길이의 비율이 설정된 값보다 크게되면 기형과일로 판단하게 되므로 자동으로 신속하고 정확하게 과일의 색상등급 및 기형여부를 판단하여 과일의 품질등급을 결정한다.As described above, when the fruit is placed on the fruit plate of the fruit sorting apparatus, the present invention photographs the fruit placed on the fruit plate using a camera, and the color signal of the fruit is obtained from the image signal of the fruit taken by the camera. Calculate the RGB signal. The a * value of the L * a * b * coordinate system is obtained using NTSC RGB calculated by correcting the RGB signal. The a * value of the L * a * b * coordinate system is compared with the a * value of the set color pixel to calculate the coloring ratio of the fruit to determine the grade of the fruit. In addition, by measuring the length of the long axis and short axis from the image signal of the fruit, if the ratio of the long axis length and short axis length is larger than the set value, it is judged as malformed fruit, so it is automatically and quickly and accurately judges the color grade and malformation of the fruit. Determine the quality grade of
Claims (6)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1019990008846A KR100290764B1 (en) | 1999-03-16 | 1999-03-16 | Method for selecting a apple with a shape and color |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1019990008846A KR100290764B1 (en) | 1999-03-16 | 1999-03-16 | Method for selecting a apple with a shape and color |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| KR20000060509A KR20000060509A (en) | 2000-10-16 |
| KR100290764B1 true KR100290764B1 (en) | 2001-05-15 |
Family
ID=19576735
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| KR1019990008846A KR100290764B1 (en) | 1999-03-16 | 1999-03-16 | Method for selecting a apple with a shape and color |
Country Status (1)
| Country | Link |
|---|---|
| KR (1) | KR100290764B1 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101228200B1 (en) * | 2011-02-09 | 2013-01-30 | 한국도로공사 | Classification apparatus and method of flat or elongated particles in aggregate using image analysis |
| KR20190065604A (en) | 2017-12-04 | 2019-06-12 | 한국식품연구원 | Method for determining quality of object |
| CN113680711A (en) * | 2021-10-27 | 2021-11-23 | 南通瑞隆农产品开发有限公司 | Agricultural product sorting method and system based on color gradation display card |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20020032453A (en) * | 2002-02-19 | 2002-05-03 | 강구태 | color measuring and comparison device |
| CN109365323A (en) * | 2018-12-03 | 2019-02-22 | 广东技术师范学院 | One kind being based on RGB fruit classification method and sorter |
| KR102239387B1 (en) | 2019-04-30 | 2021-04-12 | 대한민국 | Strawberry quality discrimination device and method for operating the same |
| KR20210017338A (en) | 2019-08-08 | 2021-02-17 | 김양희 | The selection service of“agricultural and marine products”based on “artificial intelligence(AI)”using big data and sound waves |
| KR102408045B1 (en) * | 2020-01-22 | 2022-06-14 | 서울대학교 산학협력단 | Method of sensing strain applied to structural color sensor based on camera and system for the same |
| CN114586546A (en) * | 2022-03-14 | 2022-06-07 | 西南大学 | Automatic strawberry picking device based on electronic nose and image recognition and control method thereof |
| CN116307899A (en) * | 2023-03-22 | 2023-06-23 | 荆州华洋供应链管理有限公司 | Food material supply management system and method based on artificial intelligence |
-
1999
- 1999-03-16 KR KR1019990008846A patent/KR100290764B1/en not_active IP Right Cessation
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101228200B1 (en) * | 2011-02-09 | 2013-01-30 | 한국도로공사 | Classification apparatus and method of flat or elongated particles in aggregate using image analysis |
| KR20190065604A (en) | 2017-12-04 | 2019-06-12 | 한국식품연구원 | Method for determining quality of object |
| CN113680711A (en) * | 2021-10-27 | 2021-11-23 | 南通瑞隆农产品开发有限公司 | Agricultural product sorting method and system based on color gradation display card |
Also Published As
| Publication number | Publication date |
|---|---|
| KR20000060509A (en) | 2000-10-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1960439B (en) | Methods and devices for image signal processing | |
| US7477781B1 (en) | Method and apparatus for adaptive pixel correction of multi-color matrix | |
| KR100290764B1 (en) | Method for selecting a apple with a shape and color | |
| CN100474891C (en) | Image processing apparatus and image processing method | |
| US7835573B2 (en) | Method and apparatus for edge adaptive color interpolation | |
| US20080123946A1 (en) | Image processing device | |
| JP2001202508A (en) | Method and device for converting color image and recording medium | |
| CN100473141C (en) | Image angle detection device and scan line interpolation device having the same | |
| KR102286470B1 (en) | LED Digital Signage Defect Detection system through Similarity Analysis of Original Image Information and Displayed Image Information | |
| US6259813B1 (en) | Method and apparatus for image processing using a signal of multiple scale values | |
| JP2010008159A (en) | Visual inspection processing method | |
| CN100445807C (en) | System and method for inspecting LCD panel | |
| CN110891172B (en) | Dead pixel detection and correction device | |
| JPH0520426A (en) | Fruit hue judgement device using neural network | |
| CN105184826A (en) | Method and device for determining image brightness background and display device | |
| CN115546141A (en) | Small sample Mini LED defect detection method and system based on multi-dimensional measurement | |
| US6340988B1 (en) | Method and apparatus for displaying video data for testing a video board | |
| US20050111019A1 (en) | Color image processing method and color image processing system | |
| JP3344731B2 (en) | Chestnut shape identification method | |
| KR100274430B1 (en) | Examination apparatus and method for osd of automatic investigator | |
| JP3985891B2 (en) | Image defect detection apparatus and image defect detection method | |
| US9041815B2 (en) | Digital camera imaging evaluation module | |
| JP4583871B2 (en) | Pixel signal generation device, imaging device, and pixel signal generation method | |
| CN106408559B (en) | Method and terminal for detecting resolution | |
| JPH04285845A (en) | Inspecting apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A201 | Request for examination | ||
| E701 | Decision to grant or registration of patent right | ||
| GRNT | Written decision to grant | ||
| FPAY | Annual fee payment |
Payment date: 20080305 Year of fee payment: 8 |
|
| LAPS | Lapse due to unpaid annual fee |