JPH0133225B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a fruit and vegetable sorting device that automatically determines the class of fruits and vegetables.

[Prior Art] Conventionally, gravimetric, mechanical, and light beam type devices are known as this type of fruit and vegetable sorting devices.

[Problem that the invention seeks to solve]

The gravimetric method measures the weight of fruits and vegetables regardless of their appearance, such as shape or color, and is therefore a convenient fruit sorting method when conducting transactions based on weight. However, even if the weight is the same, the size is not necessarily the same, and items that have an unsightly shape or appearance may be mixed in, making it difficult to carry out the packaging process smoothly or reducing the transaction price. There were problems such as a decline in the performance. In addition, mechanical methods sort fruits and vegetables based on the spacing between rollers that convey them, and are suitable for rough sorting, but they only sort fruits and vegetables based on a single standard, the maximum diameter. It is difficult to say that it is a sufficient fruit sorting device because it only performs the following tasks, and the results do not necessarily match the actual appearance. Another drawback is that the quality deteriorates due to "damage" that occurs when the material passes through the rollers.

On the other hand, the light beam type is almost the same as the mechanical type except that it uses light, but it has the advantage that it does not suffer from "damage" due to penetration, unlike the mechanical type. However, there still remains the drawback that the screening results do not match the actual appearance. Furthermore, among these light beam methods, there is a method called the surface line method that measures the width and height of the target fruit or vegetable, but there is a problem in that the measurement accuracy is difficult.

Therefore, an object of the present invention is to provide a fruit sorting device capable of high-speed and highly accurate discrimination.

[Means for solving problems]

first and second imaging means that are arranged at a predetermined interval from each other and take individual images of the fruits and vegetables to be measured; and a binary value that binarizes the imaging signal from the first imaging means at a predetermined threshold level. a storage means for storing the binarized imaging signal; a first determination means for determining the class based on the shape and size of fruits and vegetables based on the stored data; and a second imaging means. A signal extracting means for extracting individual signal components of a predetermined wavelength from an imaging signal, and calculating brightness, hue, and saturation of fruits and vegetables from the value of each extracted signal component and their mixing ratio, and grade determination based on the color of fruits and vegetables. A second determination means for performing the determination is provided.

[Effect]

The fruits and vegetables to be measured are imaged by a video sensor and a color video sensor, and based on the image signals obtained by the video sensors, the class is determined based on the size of the fruits and vegetables and the position of the fruit core. Based on the obtained signals, the fruits and vegetables are graded based on brightness, hue, and saturation, and the fruits and vegetables are sorted by integrating these judgment results.

〔Example〕

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a block diagram showing a color determining device, FIG. 3 is a block diagram showing a shape determining device, and FIG. 4 is a block diagram showing an embodiment of the present invention. FIG.

In Figure 1, FR is the fruit or vegetable to be measured, BC is the bucket conveyor, SE ₁ is the color video sensor,
SE ₂ is an ITV (industrial television) camera, FS is a photo sensor, DE ₁ is a color judgment device, DE ₂ is a shape judgment device, SY is a general judgment device, and SR is a shift register.

The fruits and vegetables FR to be measured are placed on a bucket conveyor BC and transported, for example, in the direction of the arrow in the figure at a predetermined speed. Although only one bucket conveyor is shown in the figure, two bucket conveyors may be provided to measure two pieces at a time. Fruits and vegetables transported FR
First, the coloring state of the surface is detected by the color sensor SE ₁ , and its grade (Excellent A, Excellent B, Good C) is determined by the color determining device DE ₁ . Next, when the fruits and vegetables FR come into the field of view of the ITV camera SE ₂ as the bucket conveyor BC advances, a shutter (not shown) opens to photograph the fruits and vegetables, and the shape determination device
DE ₂ performs predetermined arithmetic processing and judgment to classify and grade fruits and vegetables (extra large LLL, extra large LL, large L, medium M, small S). In addition, the grading performed here is to judge excellent A, excellent B, or good C depending on the degree of deviation, such as when the position of the fruit core is deviated from the center of the fruit or vegetable. The comprehensive judgment device SY combines these judgment results and stores the results in the shift register SR. The judgment result stored in the shift register SR is
The signal is sent to a sorting device (not shown) at a predetermined timing based on the signal from the photo sensor FS. The photo sensor FS outputs a judgment timing signal t to the judgment devices SE ₁ and SE ₂ in accordance with the photographing timing of the camera, and also sorts the fruits and vegetables FR that have arrived at a predetermined position on the bucket conveyor BC at a predetermined timing. Outputs a signal to discharge through the device. The data regarding fruits and vegetables sorted in this way can be displayed by producer and sorting as needed, and can be printed out as individual data and daily report data.

Next, the color determination device will be explained with reference to FIG.

The reflected light from fruits and vegetables is separated into wavelength components of predetermined colors (red, green, and blue) and extracted by a filter, converted into electrical signals by photoelectric conversion elements, and then sent to amplifiers 1 ₁ to _{1 3} . It is applied to the arithmetic/judgment device 2 via the signal. The calculation/judgment device 2 calculates the lightness (color strength or brightness) by performing the following calculations from the signal values of the three components and their mixing ratio.
Li, hue (tone) La, and saturation (degree of vividness) Lb are calculated.

Li=√・100(%) La=175(1.02X−Y)/√ Lb=70(Y−0.847Z)/√ Here, X, Y, and Z are the red, green, and blue signal component values, respectively. It is.

These calculation results are compared with a predetermined value preset in the setting device 4, and the grade (excellent A, excellent B, good C) is determined. Matrix circuit 3
selects a predetermined output terminal based on this determination result, and supplies the determined output to the comprehensive determination device shown in FIG.

Note that a commercially available color sensor can be used as such a color sensor.

Next, the shape determination device will be explained with reference to FIG.

The image signal obtained by shooting with the ITV camera is
A preprocessing and binarization circuit 11 performs noise removal, binarization by comparison with a predetermined level signal, etc., and stores the resultant signal in a pixel memory 12. Based on the data stored in the pixel memory 12, the calculation/judgment circuit 13 calculates the size of the target fruit or vegetable, such as the maximum cross-sectional area, circumference, and maximum values in two axis directions, of the target fruit or vegetable converted into a binary image (pixel). Find the characteristic values to represent. At this time, for example, since the area of one pixel is known in advance, the maximum cross-sectional area can be calculated by counting the number of significant pixels in the binarized signal, that is, the number of pixels indicating fruits and vegetables. You can ask for it. On the other hand, the perimeter can be calculated by counting the number of pixels that form the boundary between the fruit and vegetable pattern and the background, since the unit length of each pixel is known in advance, or by measuring a straight line parallel to the horizontal or vertical line of the screen. It can be easily calculated by counting the number of pieces in the single part and approximating the other parts with diagonal lines, that is, by multiplying the counted number by √2, and then adding these pieces. In addition, each maximum value (maximum diameter) in two orthogonal axes directions can be determined, for example, from the difference between the maximum coordinate position and the minimum coordinate position of each axis.

The method for determining such area and perimeter length is publicly known, but if necessary, for example, Japanese Patent Application Laid-Open No.
Please refer to Publication No.-25634.

On the other hand, the quality of the shape of fruits and vegetables, that is, the quality of appearance, such as whether or not they are distorted or symmetrical, is determined as follows.

Now, suppose that a binary image obtained by imaging a fruit or vegetable such as a pear or an apple from the upper part including the fruit core is shown as shown in FIG. First, find the center of gravity W from the point where the areas of the four parts divided into four parts (left, right, top, and bottom) are equal to each other, find the center coordinates of the fruit core CF, and calculate the center of gravity of this center coordinate. Distance from W (also called deviation)
Find ΔW. Note that the method for determining the center of gravity is not limited to the above-mentioned method, but may be any known method disclosed in, for example, Japanese Patent Publication No. 51-27973, Japanese Patent Application Laid-Open No. 51-120754, and the like. Note that the same applies to the method of determining the center coordinates of the fruit core. The smaller the deviation ΔW from the center of gravity of the fruit core obtained in this way is, the better the product, so this deviation is compared with a predetermined set value to determine the grade.

〔Effect of the invention〕

According to this invention, the classification of fruits and vegetables, etc. is automatically performed at high speed and with high accuracy, so that the unfairness caused by manual labor is corrected, and
This makes packaging work easier, and because the colors and shapes of the packed products are consistent, transaction prices can be increased.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a block diagram showing a color determining device, FIG. 3 is a block diagram showing a shape determining device, and FIG. 4 is a block diagram showing an embodiment of the present invention. FIG. 3 is a reference diagram for explaining a shape determination method according to the present invention. Symbol explanation, SE ₁ ……Color sensor, SE ₂ ……
ITV camera, FR...Fruits and vegetables, CF...fruit core, BC
...bucket conveyor, FS...photo switch,
DE ₁ ...Color judgment device, DE ₂ ...Shape judgment device,
SY... Comprehensive judgment device, SR... Shift register,
W...center of gravity, ΔW...deviation, 1...amplifier, 2,
13... Arithmetic/judgment device, 3... Matrix,
4, 14...setting device, 11...preprocessing and binarization circuit, 12...pixel memory.

Claims (1)

[Claims] 1. First and second imaging means arranged at a predetermined interval from each other to individually image the fruits and vegetables to be measured, and binarizing the imaging signal from the first imaging means at a predetermined threshold level. a binarization means for storing the binarized image signal, a storage means for storing the binarized image signal, and a feature quantity representing the size of the fruit or vegetable, a center coordinate of the fruit core, and a center of gravity, respectively, based on the stored data. At the same time, a first determination is made to determine the deviation between the center of gravity and the central coordinates of the fruit core, and to compare the deviation and the feature amount with predetermined set values, respectively, to determine the class based on the shape and size of the fruits and vegetables. means for extracting individual signal components of a predetermined wavelength from the imaging signal of the second imaging means; and second determining means for determining the grade based on the color of fruits and vegetables by calculating the degree and comparing the calculated result with a predetermined set value, and both the first and second determining means A fruit and vegetable sorting device characterized by sorting fruits and vegetables by integrating judgment results.