JPH0265725A - Introducing device in harvester of fruit vegetable - Google Patents

Abstract

PURPOSE:To make possible to perform harvesting work in turn by introducing harvesting hand to another fruit vegetable in case of no harvest for introducing harvesting hand in some set times for harvesting some fruit vegetable forward of the harvester of fruit vegetables. CONSTITUTION:In an automatic harvester of fruit vegetables having a function of introducing a harvesting hand H to target fruit vegetables F by detecting the fruit vegetables F with an imaging means S1 installed in the harvesting hand H in a visual field forward of said imaging means, a function introducing the harvesting hand H to other fruit vegetables in another visual field with stopping harvest of first fruit vegetables F in a case of no harvest for some set times of introducing of the harvesting hand H to the first fruit vegetables F.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention provides a work hand for picking up fruits and vegetables.
A work range divided into a plurality of sections is set in advance, and there is an imaging means for taking an image of fruits and vegetables within one work range, and a position of the fruits and vegetables within the two work ranges based on the imaging information of the imaging means. an image processing means for determining the direction in which the fruits and vegetables are located; a hand guiding means for guiding the working hand in the direction in which the fruits and vegetables are located based on the information of the image processing means; If there are no fruits and vegetables in the work box imaged by the imaging means (1), a working range changing means for changing the imaging field of the imaging means to the next working range is provided. Number f! Regarding the guidance system of the aircraft.

[Conventional technology]

The above-mentioned guidance device of this type of fruit and vegetable harvesting machine is designed to automatically harvest multiple fruits and vegetables in succession, when there are no fruits and vegetables within one working range, and when the work Are there no fruits and vegetables within one work area due to manual harvesting? ; If so, the imaging field of view is automatically changed to the next work range.

However, in the conventional method (1), only when there are no fruits and vegetables within one work N'JP, the imaging field of the imaging means is changed to the next work range '1).

[Problem J to be solved by this invention] For example, if there is another object in front of the fruits and vegetables that obstructs the approach of the work hand, no matter how many times the work hand is guided, the fruit and vegetables cannot be moved. It becomes impossible to harvest. As a result, fruits and vegetables continue to exist within the same work area,
The conventional configuration has the disadvantage that it is not possible to automatically change the imaging field of view to the next work range.

The present invention has been made in view of the above circumstances, and its purpose is to enable automatic change to the next work range when there are fruits and vegetables that cannot be harvested.

[Means to solve the problem]

In the guidance device for a fruit and vegetable harvesting machine according to the present invention, a work range divided into a plurality of sections is set in advance on the front side of a work hand that captures fruits and vegetables, and the fruit and vegetables within one work range are imaged. an imaging means; an image processing means for determining the direction in which the fruits and vegetables within the two working ranges are located based on the imaging information of the imaging means; Based on the information from the image processing means and the hand guiding means for guiding the vegetables in the direction in which the vegetables are located, if there are no fruits and vegetables within the work range imaged by the re-recording imaging half-submerged A working range changing means for changing the imaging field of the imaging means to the next working range is provided, and its characteristic configuration is as follows.

That is, the first characteristic configuration includes a capture detection means for detecting whether or not the work hand captures the fruits and vegetables; guidance number counting means for counting the number of times the hand is guided toward the same fruits and vegetables, and the work range changing means is configured to: If the same fruit and vegetable cannot be captured even after being guided a number of times larger than the set number of times, the imaging field of the imaging means is changed to the next working range.'7) be.

Further, in the second special @ configuration, when a plurality of fruit and vegetable heads are present in the - two work ranges, the work range changing means changes the number of times that the husbands of the plurality of fruits and vegetables are all guided. The present invention is configured to change the imaging field of view of the imaging means to the next working range when the number of times exceeds the set number of times.

Furthermore, a third characteristic configuration is that distance detection means for detecting the distance to the fruits and vegetables approached by the working hand is provided, and the guidance number counting means uses information from the image processing means and the distance detection means. The present invention is configured to count the number of times guidance is given to the same fruits and vegetables based on the information.

[For 1′] In the feature configuration of 1i, the number of times the work hand is guided toward the same fruit and vegetables is counted, and when the number of times the hand is guided toward the same fruit and vegetables that have failed to be harvested is less than the set number of times, that fruit and vegetable is This automatically changes the imaging field of view to the next work area by cutting off the items that cannot be harvested.

However, in the characteristic configuration of the above-mentioned rod 1, if there are multiple fruits and vegetables within one working range, if the number of guiding times for any of the fruits and vegetables that have failed to be harvested is greater than the set number, the harvest will be stopped. Even if there are available fruits and vegetables left, there is a risk that the scope of work will change to the next one.

Therefore, in the second characteristic configuration, when there are multiple fruits and vegetables within one work range, if the number of times of guidance for each of the multiple fruits and vegetables is all greater than the set number of times, , change to the next scope of work.

By the way, if another fruit vegetable appears to overlap behind the fruit vegetables, when the fruit vegetable in the front side is harvested, the fruit vegetable present in the rear side will appear at the same position in the imaging field of view. Therefore, it may be difficult to accurately determine whether or not fruits and vegetables that appear at the same position are the same, using only imaging information.

In addition, even if the harvest of one fruit vegetable fails, when the next fruit vegetable is harvested, the position of other objects such as branches that get in the way of guiding the fruit vegetable that failed to be harvested changes,
Fruits and vegetables that failed to be harvested may still be ready for harvesting.

Secondly, in the third characteristic configuration, a distance detection means is provided to detect the distance to the fruits and vegetables approached by the work hand, and by using image information and distance information together, the same position within the imaging field of view is provided. While distinguishing whether or not the fruit vegetables in the same fruit vegetable group are the same fruit vegetables, the number of induction times of the fruit vegetables that failed to be harvested is accurately counted.

C Effect of the Invention] As a result, even if there are fruits and vegetables that cannot be harvested within one working range, the working range can now be automatically changed to the next working range.

Practical example] Embodiments of the present invention will be described below based on the drawings.

As shown in Fig. 6, a boom (2) is attached to a vehicle body (V) equipped with a pair of left and right running wheels (1) at the front and rear so as to be able to freely turn up and down, and the tip of the boom (2) , the auxiliary boom (3) is attached so as to be swingable in the horizontal direction,
A working manipulator (4) equipped with a working hand (H) for fruit harvesting is attached to the tip of the auxiliary boom (3), thereby forming a fruit harvesting machine.

In the figure, (5) is the hydraulic cylinder for lifting the boom, (6)
(7) is a hydraulic cylinder for swinging the boom, and (7) is an electric motor for swinging the auxiliary boom.

The working manipulator (4) consists of a multi-joint telescopic arm (8) and a working hand (H) attached to the tip of the telescopic arm (8).

The telescoping arm (8) is swung around the vertical axis (Y) by an electric motor (9a), and the electric motor (
9c) is configured to be expanded and contracted.

To explain the working hand (1()), as shown in FIG. 1 in the suction opening (10a) of 10)
Capture part case (12) that covers the fruit (F)
) and the fruit (
F) is provided with an emotional cutting device (13) that cuts the handle part while holding it in the vertical direction. The vacuum band <10) is made of a flexible member such as rubber that maintains elasticity so as not to damage the fruit (F).

The trapping part case (12) has the vacuum band (
10) is retracted to the rear side so as not to interfere with the approach until the fruit (F) is adsorbed, and after the vacuum band (10) has adsorbed the fruit (F), the adsorbed fruit ( In order to protrude to the front side of the hand so as to cover F), it is externally supported so as to be movable forward and backward relative to the ventilation pipe (11) of the vacuum pad (10). The ventilation pipe (11) is connected to the suction pump (17) (see Fig. 1) mounted on the vehicle body (V) through a bellows-type hose (14) (see Fig. 6). It's been done.

Further, the trapping part case (12) is configured to be connected to the ventilation pipe (11).
The upper case (12b) is fixed to the support member (12a) which is fitted onto the outside of the support member (12a), and the lower case (12b) is pivotally connected to the support part H (12a) so as to be swingable about the vertical axis.
c) The lower case (1) is formed by being divided into two parts.
2c) in the vertical direction with respect to the ventilation pipe (11), the fruit (F) with the stem cut off held in the trapping part case (12) can be discharged to the outside. It's Nishide.

The emotional cutting device (13) includes a clipper-shaped blade (13a) for cutting the handle of the fruit (F), and a blade (1) for cutting the handle of the fruit (F).
3a) and a handle support member (13b) that presses and supports the handle of the fruit (F) from below.

However, although not described in detail, the work handk 11) includes a harvesting operation such as an actuator for moving in and out of the trapping case (12) and an actuator for driving the emotional center device (13). various actuators for the vacuum pad (10), a negative pressure sensor (S3) for detecting that the fruit (F) has been adsorbed to the vacuum pad (10) (see Figure 1), and the output of the Ura-tatsuja case (12). Various sensors such as a sensor for detecting the retreated position and a sensor for detecting the driving state of the emotional disconnection device (13) will be provided.

In the figure, (S1) is a color image sensor as an imaging means for capturing an image of the fruit (F) located in the direction of the working hand (H), and (S2) is the direction of the working hand (H). H) is an infrared light reflective proximity sensor that detects when the fruit approaches the target fruit within a set distance,
Both sensors (S1) and (S2) are provided inside the vacuum pad (10) so as not to interfere with approach and fruit harvesting. Further, (15) is an illumination device d for illuminating the front side of the working hand (14) with a set light amount in synchronization with the imaging process of the image sensor (S1), which illuminates both the sensors (S1). , (
S2>, in order not to interfere with approach and fruit harvesting, the T: side location near the tip of the ventilation pipe (11) is placed inside the trapping case (12). It is attached to.

Then, by retiring the capturing case (12), the suction port (10a) of the vacuum pad (10) becomes 11.
j. With the work hand (H) protruding further forward than the catching part case (12), point the working hand (H) in the direction in which the fruit to be harvested is located, and while illuminating with the illumination device ff1 (15), The image sensor (S1) performs image processing, and based on the captured image information, extracts image information corresponding to the fruit from other objects such as backgrounds and leaves, and based on the extracted image information, This is to determine the approach direction of the working hand (H).

However, the image sensor (S1)
A plurality of work ranges are set corresponding to the size of the imaging field of view. Then, with the working hand (H) positioned at a set approach start position in one of the plurality of feat ranges, an image is taken in the direction in which the working hand (H) is directed. It is.

To explain, as shown in FIGS. 6 and 7, the vertical position and rotation position of the boom (2) and the rotation position of the auxiliary boom (3) are changed for each set angle. By doing this, the imaging field of view corresponding to each work range is set so that it is adjacent to the adjacent work range. , the working hand (1
4) move linearly toward the determined approach direction, each of the electric motors (9a), (9b)
, (9c) and operate the sensor (S) in the vicinity.
2) As the hand senses the fruit, the neem band (10) is actuated to attract the fruit to be harvested, and the catching case (12) is protruded to the front side of the hand. After the fruit is taken into the trap case (12), the fruit is cut and harvested.

However, if there are multiple fruits within the imaging field of the image sensor (S1), the imaging of the image sensor (S1) will be adjusted in such a way that the one located closer to the front will be approached first. The approach order is automatically determined based on image brightness information.

The proximity sensor (S2) is configured to be able to detect the distance at which it senses approach to the fruit in two stages: far and near. In other words, although I will not go into details, the received light signal, which is the received light intensity of the reflected infrared light projected toward the front side of the hand, is lower than the set threshold corresponding to the first set distance (set to 30 cm). If the vacuum pad (1
0) becomes larger than the set threshold value corresponding to the second set distance (set to 5 am) at which the target fruit is located immediately in front of the fruit to be harvested. It is designed to be output.

The sensing signal of the first set distance is transmitted to the suction pump (17).
) starts the suction operation of the vacuum band (10) by The sensing signal of the second set distance will be used as activation information to make the vacuum pad <10
The arm (8) moves efficiently at high speed until the arm (8) approaches the front side of the fruit, and after it approaches the front side of the fruit, it can accurately adsorb the fruit and stop.
This is used as control information for switching the expansion speed between high and low.

Then, after starting the suction operation of the vacuum pad (10), the extension of the arm (8) is stopped as the negative pressure sensor (S3) turns on and detects fruit adsorption. , which will start the harvesting operation.

However, as will be detailed later, if the harvest fails,
The method automatically approaches the next fruit, counts the number of approaches to all the fruits within the imaging field of view at the approach start position, and makes sure that the number of approaches to each fruit that fails to be harvested is all greater than the set number of times. If this happens, try to move on to the next area of work on your own.

First, a control configuration for detecting the direction in which the fruit is located, that is, the approach direction, based on the image information captured by the image sensor (S1) and guiding the working hand (H) will be explained.

As shown in Fig. 1, the hydraulic cylinder for raising and lowering the boom (
5) Control each of the boom lifting hydraulic cylinder (6) and the auxiliary boom swinging motor (7) to direct the imaging field of view of the image sensor (S1) in a direction corresponding to a predetermined work range. The image sensor (
S1) performs image processing on the TFI image information to detect the direction in which the fruit to be harvested (F) is located, and based on the detected information, directs the working hand (l()) toward the fruit to be harvested. A microcomputer-based control device (1) for controlling each of the arm drive electric motors (9a), (9b), and (9c) so as to guide the arm.
6) is provided.

That is, an image processing means (100) that uses the control device (16) to determine the direction in which a fruit within one working range is located based on the imaging information of the image sensor (S1) as an imaging means; The image processing means (100)
hand guiding means (101,
), +i? Based on the information of j. Image processing failure (100), if there is no fruit within the work/range imaged by the image sensor (S1), and all the number of harvest failures for each fruit within the work range. a work range changing means (for changing the imaging field of view of the image sensor (S1) to the next work range when the number of times becomes more than a set number of people);
102>, a capture detection means (103) for detecting whether the fruit (for the work)/end (H) has captured a fruit, information on the image processing means (100) and the capture detection means (103);
guidance number counting means (
104), and distance detection means (10) for detecting the distance to the fruit approached by the working hand (H).
5) will be configured.

Next, the configuration of each of the means will be described in detail while explaining the operation of the side control device (16) based on the flowchart shown in FIG.

First, each of the hydraulic cylinder for lifting and lowering the boom (5), the hydraulic cylinder for swinging (6), and the hydraulic cylinder for moving the auxiliary boom (7) is operated so that the working hand (H) is in one position. After setting the approach start position in the work range, image processing is performed on the work range, and
The imaging information is image-processed to obtain positional information of the fruit within the imaging field of view.

To explain the image processing, the fruit (F)
A specific color area (F1) corresponding to the color of If there are a plurality of setting areas (F2), the approach order is determined based on the size of the setting area (F2) so that the largest one is approached first. (See Figure 4).

To explain the process for extracting the specific color region (F1), the three primary color signals (R), (G), among the image signals output from the image sensor (S1),
(f]) to binarize the signal obtained by subtracting the red signal (R) containing the color component of the fruit (F) or other color signals (G or B) based on the set darkness value. Accordingly, the image captured by the image sensor (S1) (Fig. 4 (a)
) from the specific color area (F1) (see FIG. 4 (oL$
The aim is to obtain the following:

However, if multiple setting areas (F2) are extracted within the specific color area (F1), it is determined that multiple fruits (F) are visible overlapping each other. Then, the position of the center of gravity (F1) of each of the plurality of setting areas (F2) is determined.

In addition, the brightness of image information is inversely proportional to the distance, and the closer the fruit (F) is, the brighter it appears.
The brightness is set within the specific color area (F1) based on the brightness signal corresponding to the brightness information of the image information output from the set area (F1).
When F2) is extracted, the closer the object is located, the larger the dog. Therefore, based on the size of the set area (F2), the fruit (F) located on the near side, which is easier to harvest, is approached first. Then, a process of performing image processing on the imaging information of the image sensor (S1) to obtain information on each region (Fly, (F2>) and center of gravity (F2) is performed based on the imaging information of the image sensor (S1). - Image processing method 1 (1
00).

The number (N) of specific color regions (F1) extracted by this image processing means (100) is determined as the number of recognized fruits to be harvested, and if the number (N) is zero, one The boom (2) and the auxiliary boom (3)
The approach start position 1 of the work hand (1()) is changed to the next work range by moving the work hand (1()).However, as described later; Even if the number of harvest failures for all unharvested fruits is greater than the set number of times, the approach start position of the work hand (1()) is changed to the next work range. That's how it is.

In other words, when the obtained number (N) of fruits (F) is zero, the process of changing the work range to the next work range is performed when the fruit (F) is within the work range imaged by the image sensor (S1).
) does not exist, the image sensor (S1)
This corresponds to a working range changing means (102) for changing the imaging field of view to the next working range.

If the obtained number (N) of fruits (F) is not zero, as described above, the approach order is determined based on the size of each setting area (F2), and the working hand (l() is The direction in which the fruit (F) is located corresponding to the determined approach order, that is, the center of gravity (F3) of the set area (F2)
), the direction of the working hand (It) is adjusted so that the arm (8) is extended.

After starting the extension of the arm <8), as described below,
In order to identify which fruit (F) in one work range is being approached, position information on the image of the center of gravity (F3) of the set area (F2) that has been approached and the proximity sensor (S2) are used. Based on the information on the extension distance of the arm (8) until the ON operation, fruit identification processing is performed and the harvesting operation is performed.

To explain, from the proximity sensor (S2) to the first
As the sensing signal for the set distance is output, the suction operation of the vacuum pad (10) is started, and as the sensing signal for the second set distance is output, the arm (8) is activated. The elongation speed is switched from high speed to low speed, and it is determined whether or not the fruit (F) has been sucked based on the detection information of the negative pressure sensor (S3). Stop the extension of the arm (8),
Harvest will be activated.

In other words, the image processing means ( This corresponds to hand guiding means (101) that guides the working hand (11) toward the direction in which the fruit (F) is located based on the information of (100).

After starting the harvesting operation, the negative pressure sensor (S
Based on the detection information of various sensors attached to the work hand (H) such as 3), it is determined whether or not the harvest has been completed.

In other words, the process of determining whether or not the harvest has been completed based on the detection information of various sensors attached to the work hand (l()) such as the negative pressure sensor (S3) is performed by the work hand (11). This corresponds to capture detection means (103) that detects whether or not the fruit (F) has been captured.

However, even if the arm (8) is extended from the approach start position until it reaches a preset distance, the proximity sensor (S2) is not turned on, so that it is determined that the harvest has failed.

If the harvest is successful, the work hand () 1) is returned to the approach start position and the harvested fruit (F) is discharged, and then the same work area is imaged again and image processed; Based on the information, it is determined whether the number (N>) of fruits (F) located within the work area is zero, and if it is not zero, it is determined that unharvested fruits (F) remain. Based on this judgment, the approach order will be re-determined and harvested as described above.

If the harvest fails, count the number of harvest failures (K) for that fruit (F), and then determine whether or not the number of approaches corresponding to the number (N) of the fruit (F) has been approached. , if the number (N) of fruits (F) has not been reached, the next fruit (F) is selected based on the determined approach order.
F) will be approached.

In other words, the process of identifying the fruits (F) that failed to be harvested and counting the number of failed harvests (K) is performed using information from the image processing means (100) and the capture detection means (103).
This corresponds to a guidance number counting means (104>) that counts the number of times the work hand (1) is guided toward the same fruits and vegetables based on the information.

When approaching all of the fruits (F) that have undergone J2ala, harvesting is performed based on the number (N) of fruits (F) obtained, their position information, and the information on the number of harvest failures (K). It is determined whether the number of guiding times for each failed fruit (F) has all reached the set number of times (set to 3 times), and if the set number of times (N) has not been reached, the imaging process is performed again. This will force them to re-approach the next unharvested fruit.

However, if the number of approaches to each fruit (F) that failed to be harvested has reached the set number of times (N),
After completing the harvesting work in one work range, the work range will be changed to the next work range.

Although not described in detail, when the harvesting work for the entire work range is completed, the harvesting work for one fruit tree is completed and the harvesting work for the next fruit tree is started.

Next, the fruit identification process will be described in detail based on the flowchart shown in FIG.

First, it is determined whether or not the number of fruits to be identified (04) is zero, and if it is not zero, the number of fruits to be identified (M) is
'', the coordinates (X, Y) of the center of gravity (F1) on the image and the arm (8) until the proximity sensor (S2) is turned on at the second set distance.
The length of the extended distance (σ) and the approached fruit (F)
The information for identifying the approach order (J) is stored in correspondence with the determined approach order (J).

However, if harvesting fails in a state where the proximity sensor (S2) does not turn on even if the arm (8) is extended the set distance, the value of the set distance is extended to the arm (8).
This will be stored as the extension distance (σ) of .

The value of the extension distance (σ) is detected by an encoder (not shown) for detecting driving fireflies attached to the electric motors (9b), (9c) for extending and retracting the arm. It is determined based on the amount of drive from the approach start position.

That is, the electric motor (9b) for extending and retracting the arm.

The process of determining the distance (σ) by which the arm (8) is extended based on the drive amount of the work hand (1(9C))
) corresponds to the distance detection means (105) that detects the distance to the fruit (F) approached.

Then, if the number of fruits to be identified is not zero,
The above-mentioned center of gravity (
The coordinates (X, Y) on the image of F3), the extension distance n(σ) of the arm (8), and the stored coordinates (
It is determined which of the above-mentioned stored extension distance (σ) corresponds to.

To explain, the coordinates (X, Y) on the image of the center of gravity (F1) corresponding to the fruit (F) that we approached this time
) and the extension distance (σ) of the arm (8) are all based on the memorized coordinates (X, Y) and the memorized extension distance (
If the fruit (F) that was approached this time is within the set range for each of the
) is identified as the fruit (F) corresponding to the fruit (F). However, if the position of the fruit (F) approached this time does not correspond to any of the positions of the fruit (F) corresponding to the determined approach order (J), for example, because another fruit (F) has been harvested, , it is judged that the original position has changed, or that the fruit (F) located behind it has been newly recognized because one fruit (F) has been harvested, and the fruit (F) that was approached this time is In order to process it as a fruit (F), the number of fruits to be identified <M) is added.

[Another example]

In the above embodiment, as the position information of the fruit (F), the fruit (
The center of gravity (F3) of the setting area (F2) where the brightness in the specific color area (F1) corresponding to the color of F) is equal to or higher than the setting value
In this example, the area where the brightness is maximum within the specific color area (F1) and the center of gravity of the specific color area (F1) itself can be used as position information for one fruit (F). It can also be used as a second-class.

In other words, if multiple fruits (F) overlap and are extracted as one specific color area (F1), the fruit (F) located at the rear side will be compared to the fruit (F) located at the front side. Since it can only be harvested after harvesting, even if multiple fruits (F) overlap, the specific color area (F1) containing those multiple fruits (F) can be combined into one fruit ( There is no problem even if it is processed as an area corresponding to F).

Further, in the above embodiment, the present invention is applied to a working machine for harvesting fruits, but the present invention can also be applied to a machine for harvesting vegetables such as tomatoes, and the specific configuration of each part may vary. Can be changed.

Incidentally, although reference numerals are written in the claims section for convenient comparison with the drawings, the present invention is not limited to the structure shown in the accompanying drawings.

[Brief explanation of the drawing]

The drawings show an embodiment of the guidance device for a fruit and vegetable harvesting machine according to the present invention, in which FIG. 1 is a control diagram, FIG. 2 is a flowchart of control operation, FIG. ), (0)
5 is an explanatory diagram of image processing, FIG. 5 is a sectional view showing a schematic configuration of a working hand, FIG. 6 is an overall side view of the working machine, and FIG. 7 is an explanatory diagram of a working range. (l4)... Working hand, (S1)...
...imaging means, (100)...image processing means,
(101)... Hand guiding means, (102).
...Work range changing means, (103)...
Acquisition detection means, (104) . . . guidance number counting means, (105) . . . distance detection means.

Claims (1)

[Claims] 1. On the front side of the working hand (H) that captures fruits and vegetables,
A work range divided into a plurality of sections is set in advance, and an imaging means (S_1) is used to take images of fruits and vegetables within one work range, and based on the imaging information of the imaging means (S_1), a work range is divided into a plurality of parts. an image processing means (100) for determining the direction in which fruits and vegetables are located;
hand guiding means (10) for guiding the working hand (H) in the direction where the fruits and vegetables are located based on the information of
1), and based on the information of the image processing means (100), if there are no fruits or vegetables within the work range to be imaged by the imaging means (S_1), the imaging field of view of the imaging means (S_1) is changed to the following: Work range changing means (1) to change the work range to
02) A guidance device for a fruit and vegetable harvesting machine, comprising: a capture detection means (103) for detecting whether or not the work hand (H) has captured the fruit and vegetables; The information of the means (100) and the capture detection means (103)
) for counting the number of times the working hand (H) has been guided toward the same fruits and vegetables based on the information of the working range changing means (104).
2) is based on the information of the capture detection means (103) and the information of the number of induction detection means (104), and the same fruit and vegetables could not be captured even if they were guided a number of times greater than the set number of times. In this case, the guidance device of a fruit and vegetable harvester is configured to change the imaging field of view of the imaging means (S_1) to the next working range. 2. A guidance device for a fruit and vegetable harvesting machine according to claim 1,
The work range changing means (102) is configured to, when a plurality of fruits and vegetables exist within the one work range, the number of guiding times for each of the plurality of fruits and vegetables is all greater than the set number of times. A guidance device for a fruit and vegetable harvesting machine, which is configured to change the imaging field of view of the imaging means (S_1) to the next working range in the case of the invention. 3. The guidance device for a fruit and vegetable harvesting machine according to claim 1 or 2, further comprising distance detection means (105) for detecting a distance to the fruits and vegetables approached by the working hand (H), The frequency counting means (104) is configured to count the number of times the same fruit and vegetables are guided based on the information of the image processing means (100) and the information of the distance detection means (105). The aircraft's guidance system.