JP3489210B2 - Fruit and vegetable detection device of fruit and vegetable harvester - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a fruit and vegetable detection device for a fruit and vegetable harvester. 2. Description of the Related Art Conventionally, fruit and vegetable harvesting machines, for example, fruit and vegetables are cucumbers. While traveling along the cucumber, the shape of the cucumber is photographed by a camera, and the detected image signal obtained by the photographing is binarized to generate a binarized image. The ratio to the width is calculated, and it is determined whether the cucumber is a cucumber or a leaf other than the cucumber, and only the cucumber is harvested. [0003] In the cucumber harvesting operation described above, if the cucumber is photographed with the cucumber and the leaf overlapping when photographing the cucumber, the binarized image becomes an image in which the cucumber and the leaf are integrated. As a result, the vertical / horizontal ratio changes, and the cucumber may be erroneously detected as not being a cucumber. However, this is to be solved and the cucumber is accurately detected to harvest. SUMMARY OF THE INVENTION The present invention provides a visual means for photographing a predetermined range including fruits and vegetables with a camera 2 and binarizing a detection image signal obtained thereby to generate a binarized image. 6, the vertical center line (a) of each pixel group on the binarized image is detected, and the number of pixels in the horizontal direction at predetermined intervals in the vertical direction of the center line (a) is detected. And a detection control device 20 for detecting whether or not it is a fruit or vegetable based on the number of detected pixels is provided as a configuration of a fruit and vegetable detection device of a fruit and vegetable harvesting machine. [0005] In the fruit and vegetable harvester, when detecting and harvesting cucumbers, the shape of the cucumbers is photographed by the camera 2 while self-propelled along the cultivated plantations. The detected image signal obtained by the photographing is binarized by the visual means 6 to generate a binarized image, and a vertical center line (a) of each pixel group on the binarized image is detected. At predetermined intervals in the vertical direction of the center line (a), the number of pixels in the horizontal direction is detected by the detection control device 20, and it is detected that the number of detected pixels is substantially the same during a predetermined length in the vertical direction. Then, this is detected as a cucumber and the cucumber is harvested. If the number of detected pixels is within a predetermined length in the vertical direction, if it is detected that the skirt is spreading or the skirt is narrowed, this is detected as a leaf or stem other than a cucumber. Is done. [0006] According to the above, discrimination from a leaf having a continuous center line (a) which is likely to be erroneously determined as cucumber,
Cucumbers can be accurately detected by being determined based on the change in the number of pixels in the horizontal direction of the center line (a). An embodiment of the present invention will be described below with reference to the drawings. The example of the figure is a diagram showing a fruit and vegetable harvesting machine 1. For example, the shape of a cucumber is photographed by a camera 2, the photographed image is binarized, and the upper and lower parts of each pixel (b) group on the binarized image are taken. A fruit and vegetable detection device that detects a cucumber 7 by detecting the center line (a) in the direction and detecting the number of horizontal pixels (b) at predetermined intervals in the vertical direction of the center line (a). I do. [0008] The fruit and vegetable harvesting machine 1 has a manipulator 4 for harvesting cucumbers 7 of fruits and vegetables on an electric traveling unit 3.
The multi-plate CCD camera 2 for photographing the shape of the cucumber to be harvested and a visual means 6 comprising a position sensor 5 for detecting the position of the cucumber are provided. In a cultivation field using the fruit and vegetable harvester 1, the cucumber 7 is supported on a slanting support 9 to grow the tree 8. The manipulator 4 includes a facing tree body 8.
Frame 10 inclined so as to be substantially parallel to the inclination of
And a base 12 mounted on the inclined frame 10 so as to be able to move up and down along a guide rail 11, a main body 13 provided on the base 12 so as to be horizontally rotatable, and a joint provided on the manipulator 4. The configuration comprises an arm 14 for forming a mold, and a hand device 15 for holding and cutting the cucumbers 7 provided at the distal end of the arm 14. The inclined frame 10 is mounted on the traveling section 3 using a hinge 16, and the back side is supported by a support link 17. The lower end of the support link 17 is provided with a long hole 1 of a support plate 29.
8 is attached to an arbitrary position, and the inclination angle of the inclined frame 10 can be arbitrarily adjusted by adjusting the attachment position. By operating each part of the manipulator 4, the cucumber 7 to be harvested with the hand device 15 is sandwiched, and the stem portion of the cucumber 7 is cut. The visual means 6, as shown in FIG.
CD camera 2, position sensor 5 that scans the field of view of multi-plate CCD camera 2 horizontally and vertically to measure the distance of the object to cucumber 7, and multi-plate CCD camera 2
And an illuminance sensor 19 for detecting illuminance when capturing an image captured with light of a predetermined wavelength. The detection control device 20 is provided in the visual means 6 as shown in FIGS. 1 and 2, and various detections detected by the multi-plate CCD camera 2, the position sensor 5 and the illuminance sensor 19 are input to the CPU 22 via the input circuit 21. The CPU 22 analyzes and processes various commands through the output circuit 23 to operate the manipulator 4 and the like. The detection control device 20 captures an image with the multi-plate CCD camera 2 and inputs a detection image signal obtained by this to the CPU 22 to generate a binary image, thereby generating a binary image. The vertical center line (a) of each pixel (b) group on the converted image is detected from the number of horizontal pixels (b) and the like. FIG. 3 is an original image, and FIG. 4 is a binarized center line image. The number of pixels (b) in the horizontal direction is detected at predetermined intervals in the vertical direction, for example, at intervals of 2 to 3 pixels (b), and the number of pixels (b) in the horizontal direction is detected. ) When the number is calculated and detected to be substantially the same during a predetermined length (L) in the vertical direction, this is detected as a cucumber 7, and the cucumber 7 is hand-held by this detection. 15 and is cut off. FIG. 5 is a binarized center line (a) image of the cucumber 7. If the number of detected pixels (b) is calculated and detected to be a tendency to spread or a tendency to narrow the skirt during a predetermined length in the vertical direction, this is not a cucumber 7 but a cucumber 7. , The leaves and stems other than the cucumbers 7 are detected, and the leaves and stems are not pinched and cut by the hand device 15 by this detection. The detection control of the cucumbers 7 by the detection control device 20 is performed as follows. The cucumber 7 is harvested (step 101), an image captured by the multi-plate CCD camera 2 is input (step 102), preprocessed, and coarse recognition of the cucumber 7 is performed. For example, coarse recognition is performed using two wavelengths of 550 nm and 850 nm, R (red), G (green), and B (blue) methods (step 10).
3). The detected image signal is binarized (step 104),
Noise (c) in the binarized image is removed (step 10).
5) The hole (d) in the binarized image is filled (step 106), the center line (a) is pre-processed, and the center line (a) is roughly sorted. For example, from the aspect ratio of the pixel (b), coarse selection is performed based on the Feret diameter ratio, area, and the like (step 107), and the center line (a) is calculated (step 10).
8), noise (c) in the detected image is removed (step 109), and the center line (a) is combined (step 1).
10). Pixel (b) in the horizontal (horizontal) direction of the center line (b)
The (length) number is detected (step 111), and it is detected whether the number of pixels (b) of the detection result is widened (step 11).
2) If NO is detected, it is detected whether the number of pixels (b) is narrowed (step 113), and if NO is detected, it is determined that the cucumber is 7. (Step 114) It is detected whether all the binarized images have been processed (Step 11).
5) If NO is detected, the process returns to step 111, and YE
When S is detected, RET is performed (step 116). In steps 112 and 113, YE
When S is detected, it is determined that it is a leaf or stem other than the cucumbers 7 (step 117), and the process proceeds to step 115. Hereinafter, the operation of the above embodiment will be described. In the fruit and vegetable harvesting machine 1, for example, fruits and vegetables are cucumber 7, and when the cucumber 7 is detected and harvested, the fruit and vegetable harvesting machine 1 is self-propelled along the cultivated plantations. The shape is photographed by the multi-plate CCD camera 2 of the visual means 6 with light of a predetermined wavelength. The captured image is input to the CPU 22 and binarized to generate a binarized image. The vertical center line (a) of each pixel (b) group on the binarized image is generated. ) Are detected, the number of pixels (b) in the horizontal direction (length) is detected at predetermined intervals in the vertical direction of the detected center line (a), and the number of pixels (b) in the detection result is the predetermined length in the vertical direction. Between (L)
When it is detected that they are almost the same, the cucumber 7 is detected by this detection, and the cucumber 7 is cut off while being held by the hand device 15. If the number of pixels (b) as a result of this detection is detected to be widening or narrowing during the vertical length (L), the detection results in
Leafs and stems other than cucumbers 7 are detected, and the hand device 15 does not pinch or cut. 7 to 10
FIG. 7 is a view showing another embodiment. FIG. 7 shows the distribution of the thickness of the cucumbers 7 in which the cucumbers 7 are sequentially cross-sectioned at predetermined intervals in a horizontal direction from the upper peduncle portion to the flower mark portion through the central portion. FIG.
FIG. 8 is a display diagram showing the thickness variation of the cucumber 7 in which the diameter of the detected image of the cucumber 7 is first-order differentiated. FIG. 9 is a second-order derivative of the diameter of the detected image of the cucumber 7. FIG. 10 is a display diagram showing the thickness variation of the cucumber 7, from FIGS. 8 and 9, it is configured to determine the fruit part and flower trace of the cucumber 7 from the positive / negative and absolute values of the differential coefficient. . The detection control of the cucumber 7 by the detection control device 20 is performed as follows. The cucumber 7 is harvested (step 201),
An image captured by the multi-plate CCD camera 2 is input (step 202), the cucumbers 7 are identified (step 203), and the number of cucumbers 7 in the image is measured and detected as n (step 202). 204), n measured numbers are reset (step 205), and it is detected whether n is n + 1 (step 206), the displacement of the cucumber 7 is calculated (step 207), and the calculated data is saved (step 205). 208), it is detected whether or not all the cucumbers 7 have been measured (step 209), and if NO is detected, the process returns to step 206. When YES is detected, n pieces of the measured number are reset (step 210), and it is detected whether n pieces are n + 1 (step 211), and it is detected whether or not there is data on the stalk of the cucumber 7 (step 212). If detected, a flag is set (step 213), and if NO is detected in step 212, the process proceeds to step 213. It is detected whether there is flower trace data (step 214),
When YES is detected, the flag is set (step 21).
5) If NO is detected in step 214, step 2
Continue to 15. The length of the cucumber 7 is estimated (step 216), the data of this length estimation is saved (step 217), and it is detected whether all the binarized images have been processed (step 218). Returning to step 211, if YES is detected, RET is performed (step 219). As described above, the change in the diameter of the cucumber 7 is first-order and second-order differentiated, and the positive and negative and absolute values of the coefficient are used to determine the fruit pattern,
Since the flower scar portion is determined, the extraction portion of the cucumber 7 can be accurately determined, and the cucumber 7 can be reliably harvested. FIG. 11 and FIG.
FIG. 11 is a view showing another embodiment, and FIG. 11 is a view showing a change in the length of the cucumber 7 when the cucumber 7 is elongated at the maximum. As shown in FIG. , Which is equal to the extension amount during other times. From the above, the harvestable time zone of the fruit and vegetable harvesting machine 1 is set and stored in the detection control device 20 so as to be a time zone excluding between PM 3:00 and PM 10:00. A harvest time setting switch for setting a time zone may be provided, and the harvest time may be set by operating the switch. Even if a harvesting operation is started by operating an operation lever (not shown) for starting and operating the fruit and vegetable harvesting machine 1, PM3: 0 in the setting memory is retained.
Between 0 and 10:00 PM, the fruit and vegetable harvester 1 is not started and does not run. The start of the harvesting operation of the cucumbers 7 by the detection control device 20 is controlled as follows. Cucumber 7 is harvested (Step 3)
01), whether the harvest time setting switch is ON or not is detected (step 302), and if NO is detected, RET is performed (step 303). If YES is detected, it is detected whether the start time has been input (step 304), and if NO, the process returns to step 304. If “YES” is detected, it is detected whether the time is within the prohibited time zone (step 305), and if “NO” is detected, the process returns to step 305. If "YES" is detected, the data is saved (step 306), and it is detected whether the end time has been input (step 307). If "NO" is detected, the process returns to step 307. If “YES” is detected, it is detected whether the time is within the prohibited time zone (step 308), and if “NO” is detected, the process returns to step 308. If YES is detected, the data is saved (step 309) and RET is performed (step 310). According to the above, the harvest time is set in a time zone in which the elongation rate of the cucumbers 7 is high, or by setting the harvest time, a decrease in yield can be prevented, and the harvest operation can be completed immediately before the adjustment operation. It is also possible to maintain the freshness easily.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 shows an embodiment of the present invention. FIG. 1 is an overall front view of a fruit and vegetable harvesting machine in a fruit and vegetable cultivation field. FIG. 2 is a block diagram. FIG. 3 is an original image. FIG. 4 is a center line image. 6 is a flow chart. FIG. 7 is a diagram showing another embodiment, and a cucumber thickness distribution diagram. FIG. 8 is a diagram showing another embodiment, and is a cucumber thickness variation diagram (first derivative). FIG. 10 is a diagram showing an embodiment of the present invention, and is a diagram showing a variation in the thickness of cucumber (secondary derivative). FIG. 10 is a diagram showing another embodiment, and FIG. 11 is a flowchart showing another embodiment. Cucumber length change diagram [FIG. 12] Diagram showing another embodiment, and a flowchart [Description of symbols] 2 Camera 6 Visual means 20 Detection control device (A) Center line (B) Pixel

Claims (1)

(57) [Claims 1] A predetermined range including fruits and vegetables is photographed by the camera 2, and a detection image signal obtained by the photographing is binarized.
In a fruit and vegetable harvesting machine provided with a visual means 6 for creating a binarized image, a vertical centerline (a) of each pixel group on the binarized image is detected and a predetermined vertical line of the centerline (a) is detected. A fruit and vegetable detecting device of a fruit and vegetable harvesting machine, comprising a detection control device 20 for detecting the number of pixels in the horizontal direction at intervals and detecting whether or not the fruit is a vegetable based on the number of detected pixels.