JPH04364585A - Decision device for surface pattern of fruit - Google Patents

Abstract

PURPOSE:To speedily perform a decision making process by evaluating the whole image according to a specific evaluation value and deciding the pattern on the surface of the fruit. CONSTITUTION:The image signal outputted by a color CCD camera 2 is passed through an A/D converter 41 in a processor 4 by color components and latched in respective image memories 42. An arithmetic part 44 extracts effective lines from data from the image memories 42 by removing lines where unnecessary picture elements more than a specific number such as a part where a background light or light source light is totally reflected are present as to lateral lines and longitudinal lines set in the image. Then the arithmetic part 44 calculates the absolute value of the differential value of each color component value as to the effective lines. The frequency at which the value crosses a predetermined value is detected as a line evaluation value. The mean value of line evaluation values of all the effective lines is calculated and the mean value is used as a stripe evaluation value indicating the extent of stripes of, for example, an apple 6 to discriminate between a striped apple and an equal- color applied.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for automatically determining a specific surface pattern of an object, and particularly to an apparatus suitable for objects having complex surface patterns such as fruits such as apples.

0002

[Prior Art] When sorting apples, etc., in order to improve their appearance when packed in boxes and to adjust the shipping time, some apples have a relatively uniform surface pattern, while others have a striped surface pattern (hereinafter referred to as "striped"). They are sometimes sorted into apples (also called apples). Conventionally, such discrimination has generally been performed by visual inspection by a worker. However, discrimination based on human senses often lacks reliability as it often varies due to individual and environmental differences. A method using photodiodes is known to automatically discriminate between colored objects. However, with this method, the detection range is a spot, so when a fruit such as an apple has a complex surface pattern, it is impossible to determine the entire surface pattern. Therefore, the inventors
He invented a color distribution determining device that automatically determines the surface pattern of fruits (Japanese Patent Application No. 10123/1999). This device is composed of an imaging means and a color distribution determination processing means. The imaging means is for imaging spherical or block-shaped objects with different hues. The color distribution determination processing means divides the image captured by the imaging means into pixels, calculates the hue for each pixel by taking the absolute value or ratio of the hue elements of the red component, green component, and blue component of each pixel. It functions to evaluate the value of the variance of the entire image from the evaluation. By determining the dispersion value from the evaluation value of each hue, it is determined whether the surface pattern of the entire object is uniform or striped.

0003

However, in the case of the device of this prior application, there is a problem in that the processing time is too long due to the large amount of calculation. SUMMARY OF THE INVENTION An object of the present invention is to provide a fruit surface pattern discriminating device that performs processing quickly.

0004

[Means for Solving the Problems] According to the present invention, there is provided an imaging means for capturing an image of a fruit and outputting an image signal, and image information for each pixel obtained from the image signal on a line set in the image. A fruit surface pattern discriminating device comprising a processing means for storing each data and discriminating a specific pattern on the surface of the fruit based on a red component, a green component, and a blue component of these stored data, The means includes a first step of extracting only effective lines containing a large amount of image information of the fruit, differentiating the luminance signal of the effective lines, and comparing the result of the differentiation with a predetermined threshold value. a second step of calculating an evaluation value for each effective line;
There is obtained a fruit surface pattern discriminating device characterized in that the discrimination is performed through a third step of evaluating the entire image based on the evaluation value.

[0005]

DESCRIPTION OF THE PREFERRED EMBODIMENTS A discriminating apparatus according to the present invention will be described in detail below with reference to the drawings. FIG. 1 shows an example in which the object to be determined is a striped apple. In FIG. 1, a belt conveyor 11 constitutes a fruit sorting line. A color CCD camera 2 (hereinafter referred to as camera 2) serving as an imaging means is disposed on the belt conveyor 11. Apple 6 is
It moves on a belt conveyor 11 while being placed on a pallet 1. This pallet 1 and the belt conveyor 11 constitute a conveyance means. In order to keep the illumination conditions for imaging by the camera 2 constant, an illumination device 5 is disposed near the camera 2. The camera 2, illumination device 5, and detector 3 are connected to a processing device 4, which is a processing means.

Next, this embodiment will be explained in detail with reference to the operation of the apparatus. When the pallet 1 enters the imaging range, the detector 3 detects this and activates the processing device 4. The processing device 4 captures an image using the camera 2 and also performs stripe determination. Figure 2
3 is a diagram showing an image information processing system, and FIG. 3 is a diagram showing an example of a captured image. The camera 2 outputs an image signal. Note that this image signal is composed of a red component, a green component, and a blue component for each pixel in which the image is divided into small parts as shown in FIG. The image signal is sent to the A/D converter 41 in the processing device 4 for each color component, converted into digital information, and latched into each image memory 42. Note that the imaging timing by the camera 2 is determined by outputting an imaging timing signal from the gate circuit 43 based on the signal from the detector 3.

[0007] An arithmetic unit 44 composed of a microcomputer or the like calculates lines for each of the horizontal and vertical lines (continuous pixels described above) set in the image, as shown in FIG. 4, from the data from the image memory 42. The evaluation value is calculated, and based on this, the stripe evaluation value of apple 6 is calculated. FIG. 5 is a diagram showing the flow of the processing (calculation of fringe evaluation value) system performed by the processing device 4. Step (2): The apple 6 is imaged by the camera 2. Step 2: The captured images are stored in the image memory 42 in three colors (red, green, and blue). Step ■...Data for the same pixel obtained from each image memory 42 (i.e., red image component value r obtained from the red image, green image component value g obtained from the green image, blue image component value b obtained from the blue image) is input to the calculation unit 44. Step ■...In one of the lines mentioned above, determine whether or not there are unnecessary pixels such as areas where the background or light source is totally reflected.
, g, b). Step ■... Lines where the ratio of unnecessary pixels (number of unnecessary pixels / number of necessary pixels) is more than a predetermined ratio (for example, the line on the background or the line AA shown in FIG. 5 where the number of unnecessary pixels is more than a predetermined number),
A line at the end of apple 6 (for example, line CC) is removed as an unnecessary line. The lines that were not removed are
Extracted as a valid line. Step ■...Calculate the line evaluation value for the effective line. For example, the arithmetic unit 44 operates on a line B as shown in FIG.
The red image component value r (luminance signal) of B is extracted, and the absolute value of the differential value of this value is calculated as shown in FIG. In the same figure, the number of times the threshold value S has been crossed is detected in advance,
This number of times is taken as the line evaluation value. Similarly, line BB
The line evaluation value for the green image component value g and the blue image component value b is calculated. Note that the reason why the differential value of luminance is used in the step of calculating the line evaluation value is to eliminate the influence of the brightness of the surface color of the apple 6. As shown in FIG. 9, even if apples have substantially the same stripes, if the brightness of the surface color differs, the brightness values on the lines will differ. For this reason, if the brightness value is used as it is, the predetermined threshold value S cannot be used, and there will be a problem that the line evaluation values will differ. Step ■...Determine whether or not the evaluation of all valid lines has been completed. If the evaluation has not been completed, proceed to step ■; if it has been completed, proceed to step ■. Step ■...Calculate the average value by dividing the sum of line evaluation values of all valid lines by the number of valid lines. This average value is defined as a stripe evaluation value indicating the degree of stripes on the apple 6. By comparing the fringe evaluation value obtained as described above with a predetermined threshold value, it is possible to determine whether the apple is a striped apple or a uniformly colored apple. The determination result is displayed visually or audibly using a known method as appropriate. Although this embodiment has been described with reference to a striped apple as the object to be determined, it goes without saying that the present invention can also be applied to other objects whose surface has an important striped pattern.

[0008]

Effects of the Invention Since the apparatus according to the present invention calculates an evaluation value for each line of pixels of a captured image, it is possible to perform the discrimination process quickly.

[Brief explanation of the drawing]

FIG. 1 is a schematic configuration diagram of a fruit surface pattern discriminating device according to an embodiment of the present invention.

FIG. 2 is a block diagram showing processing means used in the apparatus shown in FIG. 1;

FIG. 3 is a diagram for explaining evaluation value calculation by the apparatus of the present invention.

FIG. 4 is a diagram for explaining evaluation value calculation by the apparatus of the present invention.

FIG. 5 is a diagram showing the flow of a fringe evaluation value processing system using the apparatus of the present invention.

FIG. 6 is a diagram for explaining evaluation value calculation by the apparatus of the present invention.

FIG. 7 is a diagram for explaining evaluation value calculation by the apparatus of the present invention.

FIG. 8 is a diagram for explaining evaluation value calculation by the apparatus of the present invention.

FIG. 9 is a diagram for explaining evaluation value calculation by the apparatus of the present invention.

[Explanation of symbols]

1 Palette 2 Color CCD camera 3 Detector 4 Processing equipment 41 A/D converter 42 Image memory 43 Gate circuit 44 Arithmetic unit

Claims (1)

[Claims] 1. An imaging means for capturing an image of a fruit and outputting an image signal, and storing image information for each pixel obtained from the image signal for each line set in the image, and storing the stored data. A fruit surface pattern discriminating device comprising processing means for discriminating a specific pattern on the surface of the fruit based on a red component, a green component, and a blue component of the fruit, the processing means comprising: A first step of extracting only effective lines that include many effective lines, and differentiating the luminance signal of the effective lines, and calculating an evaluation value for each effective line based on the result of comparing the differentiation result with a predetermined threshold value. A fruit surface pattern discriminating device characterized in that the discrimination is performed through a second step and a third step of evaluating the entire image based on the evaluation value.