Automatic and efficient fruit harvesting method

Abstract

According to the automatic and efficient fruit harvesting method, part of leaves falling into the collecting frame are preliminarily removed through the first impurity removing mechanism, and fruits and the remaining leaves in the collecting frame fall onto the first conveying belt from the lower portion of the collecting frame; then the first conveying belt and the leaves falling onto the first conveying belt are shot in real time through an image acquisition device, the leaves in the shot picture are recognized through a control module, and the proportion of the total area of the leaves in the picture to the area of the first conveying belt is calculated; then, the second impurity removing mechanism is driven to be at different gears in real time based on the different proportions obtained through calculation, so that the leaves conveyed to the second conveying belt from the first conveying belt are blown away from the second conveying belt, and it is ensured that fruits on the second conveying belt are separated from the leaves; therefore, the situation that the leaves are conveyed into the collecting box along with the fruits on the second conveying belt is effectively reduced, and then the follow-up conveying process is effectively avoided due to the fact that the leaves need to be removed again after collection is time-consuming and labor-consuming.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent agricultural picking machinery, and particularly relates to an automatic high-effect real picking method. BACKGROUND

[0002] In recent years, with the rapid development of forestry and fruit industry, fruit picking has become a technical evaluation affecting the development of the industry. The existing fruit picking technology mainly relies on manual picking. For manual picking, the picking method is low in efficiency, high in labor intensity and high in labor cost. Some use mechanical picking. The existing picking equipment can be roughly divided into single picking tools, single picking tools and picking integrated machines. The picking efficiency of the picking integrated machine is high and the picking effect is good, which is suitable for large-scale forestry and fruit planting users. The efficiency is significantly improved compared with manual picking. For example, Chinese patent application No. 202311549955.8, classified as A01D46 / 26, published on January 12, 2024, discloses a modular fruit picking device, which comprises a connecting piece detachably mounted on a walking machine; a vibrating machine is detachably connected with the connecting piece, and the vibrating machine is provided with a clamping mechanism; a collecting frame is detachably connected with the connecting piece; a driving machine is installed on the collecting frame and is in transmission connection with a main support rod, and the collecting frame is provided with a vice support rod and a fixed rod, the two ends of the fruit collecting cloth are connected with the fixed rod and the main support rod respectively, and the vice support rod is connected with the fruit collecting cloth; the two fruit collecting cloths can form a bucket-shaped collecting cavity, and the clamping mechanism and the fruit tree trunk to be picked can be limited in the bucket-shaped collecting cavity.

[0003] The above-mentioned document realizes single picking, single picking and integrated picking operation of fruits through the cooperation of detachable connecting piece, vibrating machine, clamping mechanism and collecting frame, thereby realizing flexible operation mode and high-efficiency picking effect, and solving the problems of low picking efficiency and inconvenient maintenance. However, in the process of collecting fruits, part of the leaves will fall into the collecting frame together with the fruits. In the process of collecting, the leaves cannot be removed, and the fruits in the collecting frame need to be picked and removed by manual method again, which is time-consuming, labor-intensive and low in efficiency, and affects the packaging process of the subsequent fruits. SUMMARY

[0004] The purpose of the present application is to provide an automatic high-effect real picking method, which can collect fruits and effectively remove fallen leaves in the process of collecting fruits.

[0005] To achieve the above-mentioned purpose, the present application provides an automatic high-effect real picking method, which collects the picked fruits by a fruit collecting machine. The fruit collecting machine comprises a mechanical arm, a vehicle frame and a collecting frame arranged on the mechanical arm. An image acquisition device is arranged on the mechanical arm arranged at one end of the vehicle frame. A first impurity removal mechanism is arranged in the collecting frame. A first driving mechanism and a collecting assembly are arranged above the collecting frame. The method further comprises the following steps: S1 drives the mechanical arm and the collecting frame to transport to below the fruit to be collected, and then opens the collecting assembly through the first driving mechanism to form a funnel-shaped collecting umbrella, so that the falling fruit falls into the collecting frame along the collecting umbrella; S2 the first driving mechanism drives the collecting frame to swing on the vehicle frame at a preset angle, and preliminarily removes the leaves falling into the collecting frame along with the fruit through the first impurity removal mechanism during and after the swinging, so that part of the leaves in the collecting frame are separated from the collecting frame from one side of the collecting frame; S3 after the fruit falling into the collecting frame falls onto the first conveying belt arranged below the collecting frame, the first conveying belt is photographed in real time through the image acquisition device, the leaves in the photographed picture are identified, and the total area of the leaves on the first conveying belt in the picture is calculated; S4 the area ratio of the leaves is calculated through the total area of the leaves and the area of the first conveying belt, and then based on different area ratios of the leaves, the second impurity removal mechanism arranged on the two sides of the second conveying belt is driven to be in different gears to remove the leaves conveyed onto the second conveying belt again, so that the leaves are separated from the second conveying belt from the two sides of the second conveying belt; S5 the second conveying belt continues to convey the fruit to the collecting box arranged on the vehicle frame.

[0006] The above method, since in the process of collecting fruit, leaves will fall into the collecting frame along with the fruit, and the collecting frame is driven to swing on the vehicle frame through the first driving mechanism, so as to reduce the accumulation of the fruit and the leaves falling into the collecting frame, thereby further facilitating the fruit to fall into the conveying belt through the collecting frame, on the other hand, the leaves and the fruit can be prevented from accumulating, and then part of the leaves falling into the collecting frame can be preliminarily removed through the first impurity removal mechanism, thereby further increasing the effect of removing the leaves by the first impurity removal mechanism, and the fruit in the collecting frame falls from the lower part of the collecting frame to the first conveying belt, since the sizes of the leaves falling along with the fruit are different, part of the leaves will still fall onto the first conveying belt along with the fruit, at this time, the first conveying belt and the leaves falling onto the first conveying belt are photographed in real time through the image acquisition device, the leaves in the photographed picture are identified, and then the total area of the leaves in the picture is calculated to account for the proportion of the area of the first conveying belt, so that the control module drives the second impurity removal mechanism and the first impurity removal mechanism to be in different gears based on the calculated different proportions, thereby further adjusting the ability of the first impurity removal mechanism and the second impurity removal mechanism to remove the leaves from the removing of the leaves in the collecting frame and the removing of the leaves on the second conveying belt, respectively, thereby effectively reducing the leaves along with the fruit conveyed onto the second conveying belt to the collecting box, thereby avoiding the fruit after collection to be removed from the leaves again, which is time-consuming and laborious, and effectively improving the subsequent packaging process.

[0007] Further, the mechanical arm comprises a first fixed arm, a second fixed arm and a third fixed seat, one end of the first fixed arm is connected to the frame, two ends of the second fixed arm are respectively connected to the first fixed arm and the third fixed seat, the image acquisition device is connected to the second fixed arm, the outside of the collecting frame is provided with a fixed frame hinged to the collecting frame, one end of the fixed frame is connected to the third fixed seat, and the third driving mechanism arranged between the collecting frame and the fixed frame drives the collecting frame to rotate around the hinge between the collecting frame and the fixed frame.

[0008] The above arrangement facilitates the adjustment of the angle of the collecting frame by the third driving mechanism, so that the collecting frame is inclined and swung relative to the fixed frame, facilitating the falling of leaves and fruits in the collecting frame.

[0009] Further, the third driving mechanism is provided with two and is respectively located at two sides of the collecting frame, the third driving mechanism comprises a third driving cylinder and a third push rod, the third push rod is connected to the third driving cylinder, one end of the third driving cylinder is hinged to the fixed frame, one end of the third push rod is hinged to the outside of the collecting frame, the bottom of the collecting frame is protruding downward to form a funnel, a funnel mouth is formed on the funnel, the first impurity removal mechanism in the cavity comprises a screen and a first fan, the screen is fixedly arranged in the cavity to divide the cavity into a first cavity and a second cavity, the second cavity communicates with the funnel mouth, the screen is provided with two or more than two screen strips, and the screen strips form interval openings therebetween, the first cavity communicates with the second cavity through the interval openings, and the first fan is fixedly arranged in the first cavity and located on one side close to the third driving mechanism. The side of the collecting frame opposite to the first fan is provided with an opening, and the opening communicates with the first cavity.

[0010] The above arrangement facilitates the rotation of the third push rod around the hinge between the collecting frame and the fixed frame under the action of the third driving cylinder, and in the rotating process, since the air outlet direction of the first fan is the same as the length direction of the screen strips, the leaves falling on the screen strips are guided along the length direction of the screen strips, so that the leaves falling on the screen strips are blown out of the first cavity from the opening by the first fan, and the fruits falling into the first cavity fall into the second cavity through the interval openings of the screen strips, and then are discharged from the funnel mouth of the collecting frame.

[0011] Further, the first driving mechanism comprises a first driving cylinder, a first push rod and a rack, the first driving cylinder is connected to the first push rod, one end of the first driving cylinder is fixedly connected to the upper end of the collecting frame, one end of the rack is fixedly connected to the first push rod, the collecting assembly comprises two or more than two gears and connecting rods, one end of the connecting rod is rotatably connected to the outer side edge of the upper end of the collecting frame, the other end of the connecting rod is inclined upward and outward, the collecting cloth is connected between the connecting rods, the other end of the rack is arranged across the collecting frame and is in meshing connection with the gears on the connecting rods.

[0012] The above setting, through the action of the first driving cylinder, makes the first push rod drive the rack to move, and then makes the rack engage with the gears located on both sides of the rack, thereby driving the connecting rod fixedly connected with the gears to rotate, so that before the fruits are collected, the collecting cloth is opened to form a funnel-shaped collecting umbrella, so as to facilitate the fruits to fall into the collecting frame along the inclined surface of the collecting umbrella, and after the fruits are collected, the rack is moved in reverse direction, so that the collecting cloth can be folded up to avoid occupying space.

[0013] Further, the second conveying belt is arranged on the vehicle frame through the second support, the collecting box is arranged below one end of the second conveying belt, the other end of the second conveying belt is arranged upwardly inclined relative to the vehicle frame, the second impurity removing mechanism fixed on the second support is arranged on both sides of the second conveying belt in a relative staggered manner, the second impurity removing mechanism is a second fan, and the first fan and the second fan are both provided with adjustable different gears.

[0014] The above setting makes the fruits falling from the funnel outlet fall onto the first conveying belt, and then are conveyed to the second conveying belt through the first conveying belt, and in the conveying process of the second conveying belt, the second impurity removing mechanism arranged in a staggered manner blows away the leaves on the second conveying belt twice, thereby effectively avoiding the leaves from being conveyed into the collecting box with the fruits through the second conveying belt, and increasing the workload of subsequent fruit picking.

[0015] Further, the inclination angle of the second conveying belt relative to the vehicle frame ranges from 5° to 10°, and the angle formed between the first conveying belt and the second conveying belt ranges from 145° to 160°.

[0016] The above setting can make the fruits and leaves be well conveyed on the conveying belts by inclining the second conveying belt and the first conveying belt.

[0017] Further, the step S2 further includes starting the first fan and adjusting the first fan to the first gear, and blowing part of the leaves falling into the collecting frame out of the first cavity along the length direction of the separation strip of the separation net through the first fan.

[0018] The above setting can blow part of the leaves away from the collecting frame through the first fan, thereby reducing the leaves entering into the second cavity with the fruits.

[0019] Further, the step S3 includes steps S3.1-S3.3. S3.1 calculates the gray value of the photographed picture, Gray(i,j)=0.299*R(i,j)+0.587*G(i,j)+0.114*B(i,j)(1), Gray(i, j) is the gray value of the picture at position (i, j) in formula (1), R(i, j) is the red component value of the original pixel in the picture, G(i, j) is the green component value of the original pixel in the picture, and B(i, j) is the blue component value of the original pixel in the picture; S3.2, the Gaussian filter kernel is convolved with the gray value to realize Gaussian filtering of the picture, L[i,j] = [1 / (2*π*σ 2 )]*exp[-1 / (2*σ 2 )]*[ (i-k-1) 2 + (j-k -1) 2 )] (2), In formula (2), L[i,j] is the value of the Gaussian kernel at position (i, j), represents the weight, [i,j] is the position in the Gaussian kernel matrix, where i represents the row index and j represents the column index, and the index starts from 0; σ is the standard deviation of the preset Gaussian distribution; exp is the exponential function, k is related to the center position of the Gaussian kernel, and the center of the Gaussian kernel is located at (k, k); S 3.3, based on the difference between green leaves and non-green fruits, calculate the green part feature T in the picture, T(i, j) = 2.2*G(i, j)-1.1*B(i, j)-1.1*R(i, j) (3); Then compare the green feature value T with the preset green feature value K, if the green feature value T is greater than or equal to the preset green feature value K, then L[i,j]=1, judging that the position pixel is a green leaf pixel; if the green feature value T is less than the preset green feature value K, then L[i,j]=0, judging that the position pixel is a non-green leaf pixel; Calculate the total pixel number of all green leaves in the picture, since the number of pixels in the image is proportional to the area and is 1:1, the total area of the leaves on the first conveying belt can be obtained by calculating the total pixel number of all green leaves in the picture, (4), In formula (4), S is the total area of the leaves on the first conveying belt, H and W are the height and width of the picture respectively; is the total pixel number of the green part feature T; Calculate the area Q of the first conveying belt in the picture, Q = c*d (5), In formula (5), c is the width of the first conveying belt, and d is the length of the first conveying belt; Solve formula (4) and (5) to calculate the leaf area ratio Z, Z = S / Q (6).

[0020] The above arrangement facilitates calculating the total area of the leaves on the first conveying belt by calculating the total number of pixels of the features of the green part in the picture, so as to calculate the proportion of the leaf area on the first conveying belt, so as to control the first fan and the second fan to be in different gears through the proportion of the leaf area, and then blow away the leaves in the fruit conveying process.

[0021] Further, the step S4 includes steps S4.1-S4.3, S4.1 If the leaf area proportion Z in step S3.3 is between [0%, 20%], the first fan and the second fan are adjusted to the first gear; S4.2 If the leaf area proportion Z in step S3.3 is between [20%, 50%], the first fan and the second fan are adjusted to the second gear; S4.3 If the leaf area proportion Z in step S3.3 is between [50%, 100%], the first fan and the second fan are adjusted to the third gear.

[0022] The above arrangement adjusts the first fan to different gears through different proportions of the leaf area, which can remove part of the leaves falling into the collection frame, and the remaining leaves will fall from the collection frame to the first conveying belt, and then be conveyed to the second conveying belt. Adjusting the second fan to different gears through different proportions of the leaf area can accurately and effectively blow away the leaves on the second conveying belt from both sides of the second conveying belt. The two second fans arranged in different positions do not affect each other and blow away the leaves in different directions during the blowing process, which can remove the leaves on the second conveying belt twice, effectively reducing the leaves with the fruit conveyed to the collection box through the second conveying belt.

[0023] Further, the step S4 includes steps S4.1-S4.3, and step S4.1 includes that if the leaf area proportion Z in step S3.3 is between [0%, 20%] and the leaf area proportion Z is relatively increased compared with the leaf area proportion in the last detection time, the first fan is adjusted to the next gear of the gear in the last detection time, and if the leaf area proportion Z is relatively decreased compared with the leaf area proportion in the last detection time, the first fan is kept at the gear in the last detection time, and the second fan is adjusted to the first gear. Step S4.2 includes that if the leaf area proportion Z in step S3.3 is between [20%, 50%] and the leaf area proportion Z is relatively increased compared with the leaf area proportion in the last detection time, the first fan is adjusted to the next gear of the gear in the last detection time, and if the leaf area proportion Z is relatively decreased compared with the leaf area proportion in the last detection time, the first fan is kept at the gear in the last detection time, and the second fan is adjusted to the second gear. S4.3 If the leaf area ratio Z in step S3.3 is between [50%, 100%], and the leaf area ratio Z is relatively increased compared with the leaf area ratio in the last detection time, the first fan is adjusted to the next gear of the gear in the last detection time, and if the leaf area ratio Z is relatively decreased compared with the leaf area ratio in the last detection time, the first fan is kept at the gear in the last detection time, and the second fan is adjusted to the third gear.

[0024] The above setting determines whether the first fan increases or keeps the original gear according to whether the leaves on the first conveying belt increase or decrease, so that the first fan can blow away most of the leaves, and the second fan is directly adjusted. BRIEF DESCRIPTION OF DRAWINGS

[0025] Figure 1 It is a schematic diagram of the overall structure of the present application.

[0026] Figure 2 It is a schematic diagram of the structure after removing the collecting cloth in the present application.

[0027] Figure 3 It is Figure 2 the cross-sectional view at B-B in the present application.

[0028] Figure 4 It is Figure 1 the enlarged view at A in the present application. DETAILED DESCRIPTION

[0029] The present application will be further described in detail below in combination with the drawings and specific embodiments.

[0030] As Figures 1-4 shown, an automatic high-effect harvesting method is collected by a fruit collecting machine, in this embodiment, the fruit is picked by a picking arm, and during the picking process, the fruit collecting machine is located below the picking arm, so that after the picking arm picks the fruit, the fruit collecting machine can collect the fallen fruit, the picking arm is fixed on the mechanical arm of the fruit collecting machine, and the harvested fruit in this embodiment takes litchi or red dates as an example, the litchi or red dates are red when picked in the mature period, and the leaves of litchi or red dates are green. The structure of the picking arm is described in detail in the literature with Chinese patent application No. 202323584120.4 in paragraphs

[0032] to

[0041] , which will not be repeated here.

[0031] The fruit collecting machine comprises a mechanical arm, a frame 1 and a collecting box 2 arranged on the mechanical arm, the collecting box 2 is internally provided with a first impurity removing mechanism, the frame 1 is provided with a control module, the mechanical arm comprises a first fixed arm 3, a second fixed arm 4 and a third fixed seat 5, one end of the first fixed arm 3 is fixedly connected to the frame 1, two ends of the second fixed arm 4 are respectively fixedly connected to the first fixed arm 3 and the third fixed seat 5, the other end of the first fixed arm 3 is provided with a connecting table (not marked in the figure), a picking arm is fixedly installed on the connecting table, an image collecting device 6 is connected to the second fixed arm 4, the outer side of the collecting box 2 is provided with a fixed frame 7 hinged to the collecting box 2, one end of the fixed frame 7 is fixedly connected to the third fixed seat 5, a third driving mechanism arranged between the collecting box 2 and the fixed frame 7 drives the collecting box 2 to rotate around the hinge between the collecting box 2 and the fixed frame 7, so as to facilitate the rotation and angle adjustment of the collecting box 2 by the third driving mechanism, so that the collecting box 2 is arranged to be inclined relative to the fixed frame 7, and through the image collecting device 6, the first conveying belt 25 can be photographed in real time to obtain the image of the first conveying belt 25 and the leaves on the first conveying belt 25. The image collecting device 6 is a camera.

[0032] As Figures 2-4As shown, the third driving mechanism is provided with two and is located at both sides of the collecting frame 2, the third driving mechanism comprises a third driving cylinder 8 and a third push rod 9, one end of the third push rod 9 is connected with the third driving cylinder 8, one end of the third driving cylinder 8 is hinged with the fixed frame 7, the other end of the third push rod 9 is hinged with the outside of the collecting frame 2, the bottom of the collecting frame 2 is protruding downward to form a funnel 10, the funnel 10 is formed with a funnel mouth 11, the first impurity removing mechanism located in the cavity comprises a screen 12 and a first fan 15, the screen 12 fixedly arranged in the cavity divides the cavity into a first cavity 17 and a second cavity 18, the second cavity 18 is communicated with the funnel mouth 11, the screen 12 is provided with two or more than two partitions 13, the partitions 13 form a spacing gap 14, the size of the spacing gap 14 is greater than the diameter of the collected fruits, the first cavity 17 is communicated with the second cavity 18 through the spacing gap, so as to facilitate the fruits to enter the second cavity 18 from the spacing gap 14, the first fan 15 is fixedly arranged in the first cavity 17 and is located at the side close to the third driving mechanism, the side of the collecting frame 2 opposite to the first fan 15 is provided with an opening 16, the opening 16 is communicated with the first cavity 17, the size of the opening 16 is smaller than the diameter of the collected fruits, in order to further ensure that the first fan removes the influence of the leaves on the conveying belt, a collecting cover can be arranged on the opening 16, the collecting cover is a collecting channel made of cloth, one end of the collecting cover is fixed on the opening through a fixing piece, the other end of the collecting cover can be put into the leaf collecting basket, so as to prevent the fruits from separating from the collecting frame 2 from the opening 16, so that under the action of the third driving cylinder 8, the third push rod 9 drives the collecting frame 2 to rotate around the hinge between the collecting frame 2 and the fixed frame 7, in the rotating process, since the air outlet direction of the first fan 15 is the same as the length direction of the partitions 13, so that the leaves falling on the screen 12 are guided along the length direction of the partitions 13, so as to blow out the first cavity 17 from the opening 16, while the fruits falling into the first cavity 17 fall into the second cavity 18 through the spacing gap 14 on the screen 12, and then are discharged from the collecting frame 2 through the funnel mouth 11.

[0033] The upper part of the collecting frame 2 is provided with a first driving mechanism and a collecting assembly. The first driving mechanism comprises a first driving cylinder 19, a first push rod 20 and a rack 21. The first driving cylinder 19 is connected with the first push rod 20. One end of the first driving cylinder 19 is fixedly connected to the upper end of the collecting frame 2. One end of the rack 21 is fixedly connected with the first push rod 20. The collecting assembly comprises two or more than two gears 22 and two or more than two connecting rods 23. One end of the connecting rod 23 is rotatably connected to the outer side edge of the upper end of the collecting frame 2. The other end of the connecting rod 23 is obliquely arranged upwards and outwards. The collecting cloth 24 is connected between the connecting rods 23. The other end of the rack 21 is arranged across the collecting frame 2 and is in meshing connection with the gear 22 which is sleeved on the connecting rod 23. The collecting cloth is sleeved on the connecting rod 23. Through the action of the first driving cylinder 19, the first push rod 20 drives the rack 21 to move, and then the rack 21 is in meshing connection with the gears 22 located on both sides of the rack 21, so as to drive the connecting rod 23 fixedly connected with the gear 22 to rotate. The connecting rod 23 fixedly connected with the gear 22 drives the collecting cloth to move. The movement of the collecting cloth drives other connecting rods 23 to open or fold, so that before the fruit is collected, the collecting cloth 24 is opened to form a funnel-shaped collecting umbrella, so as to facilitate the fruit to fall into the collecting frame 2 along the inclined surface of the collecting umbrella. After the fruit is collected, the rack 21 is reversely moved, so that the collecting cloth 24 can be folded, so as to avoid occupying space. In the embodiment, when the rack 21 extends outwards, the gears located on both sides of the rack 21 rotate counterclockwise, so that the connecting rod rotates to the middle part of the collecting frame, and then the folding effect is realized. When the rack 21 is retracted, the gears located on both sides of the rack 21 rotate clockwise, so that the connecting rod rotates to the outer side of the collecting frame, and then the opening effect is realized.

[0034] The second conveying belt 26 is arranged on the vehicle frame 1 through a second support (not marked in the figure), the collecting box 28 is arranged below one end of the second conveying belt 26, the other end of the second conveying belt 26 is arranged upwardly inclined relative to the vehicle frame 1, the inclination angle ranges from 5° to 10°, in this embodiment, the inclination angle a is 10°, the second impurity removal mechanism fixed on the second support is provided with two and arranged oppositely on the two sides of the second conveying belt 26, the second impurity removal mechanism is a second fan 27, the first fan 15 and the second fan 27 are all provided with adjustable different gears, and are all connected with a control module (not marked in the figure) through a Bluetooth module (not marked in the figure), the different gears include a first gear, a second gear and a third gear, one end of the first conveying belt 25 is fixed on the second support, the other end of the first conveying belt 25 is fixed on a first support (not marked in the figure), the first support is fixedly connected on the fixed frame 7, and the first conveying belt 25 is below the funnel mouth 11, the angle formed between the first conveying belt 25 and the second conveying belt 26 ranges from 145° to 160°, so that the fruits discharged from the funnel mouth 11 fall on the first conveying belt 25, are conveyed to the second conveying belt 26 through the first conveying belt 25, and the leaves on the second conveying belt 26 can be blown away twice forward and backward through the oppositely arranged second impurity removal mechanism during the conveying process of the second conveying belt 26, so that the leaves are effectively avoided from being conveyed with the fruits to the collecting box 28 through the second conveying belt 26, and the workload of subsequent picking fruits is increased, in this embodiment, the angle b formed between the first conveying belt 25 and the second conveying belt 26 is 160°.

[0035] Further comprising the following specific steps: S1, the mechanical arm and the collecting frame 2 are transported to below the fruits to be collected through the second driving mechanism arranged on the two sides of the vehicle frame 1, then the collecting assembly is opened through the first driving mechanism to form a funnel-shaped collecting umbrella, so that the fallen fruits fall into the collecting frame 2 along the collecting umbrella. In this embodiment, the structure of the second driving mechanism is specifically described in the

[0043] paragraph in the authorized document with the Chinese patent application No. 2024110316818, which will not be repeated here.

[0036] S2, the first fan 15 is started, and the first fan 15 is adjusted to the first gear, part of the leaves falling into the collecting frame 2 are blown out of the first cavity 17 from the opening 16 along the length direction of the separation strip 13 on the separation net 12 through the first fan 15, so as to preliminarily remove the leaves falling into the collecting frame 2 with the fruits through the first impurity removal mechanism, so that part of the leaves are separated from the collecting frame 2 from one side of the collecting frame 2; at the same time, the third push rod 9 is driven to extend and retract through the third driving cylinder 8, so as to drive the collecting frame 2 to rotate around the hinge between the collecting frame 2 and the fixed frame 7, in this process, the fruits falling into the collecting frame 2 are promoted to fall from the separation net 12 into the second cavity 18 through the rotation of the collecting frame 2, so that the fruits are discharged from below the collecting frame 2 to the first conveying belt 25.

[0037] S3 After the fruit entering the collection box 2 falls onto the first conveyor belt 25 set below the collection box 2, the image acquisition device 6 takes a real-time picture of the first conveyor belt 25, and then the control module identifies the leaves in the captured picture and calculates the total area of ​​the leaves on the first conveyor belt 25 in the picture; specifically including steps S3.1 to S3.3; S3.1 Calculate the grayscale value of the captured image. Gray(i,j)=0.299*R(i,j)+0.587*G(i,j)+0.114*B(i,j)(1), In equation (1), Gray(i,j) is the gray value of the image at position (i,j), R(i,j) is the red component value of the original pixel in the image, G(i,j) is the green component value of the original pixel in the image, and B(i,j) is the blue component value of the original pixel in the image. In digital images, the values ​​of R(i,j), G(i,j), and B(i,j) are all between 0.0 and 1.0. S3.2 implements Gaussian filtering by convolving the Gaussian filter kernel with the grayscale values. (2), In equation (2), L[i,j] is the value of the Gaussian kernel at position (i, j), representing the weight used in convolution operations in image processing. [i,j] is the position in the Gaussian kernel matrix, where i represents the row index and j represents the column index, and the index starts from 0. σ (i.e., sigma) is the standard deviation of the preset Gaussian distribution, which controls the fluctuation of the Gaussian function. The larger σ is, the smoother the Gaussian kernel is, that is, the stronger the image blurring effect. The smaller σ is, the sharper the Gaussian kernel is, that is, the weaker the image blurring effect. exp is the exponential function, that is, the natural constant is e (approximately 2.71828) raised to a power. k is related to the center position of the Gaussian kernel and represents the size of the Gaussian kernel. Since the index starts from 0, in equation (2), the center of the Gaussian kernel is located at (k, k) through two offsets of (k+1). S 3.3 Based on the difference between green leaves and non-green fruits, calculate the green feature value T of each pixel located in the first conveyor belt area of ​​the image. (3); Then, the green feature value T is compared with the preset green feature value K. In this embodiment, the preset green feature value K is the maximum feature value of the non-green leaf pixels. If the green feature value T is greater than or equal to the preset green feature value K, then L[i,j]=1, and the pixel at this position is determined to be a green leaf pixel; if the green feature value T is less than the preset green feature value K, then L[i,j]=0, and the pixel at this position is determined to be a non-green leaf pixel. calculating the total pixel number of all green leaves in the picture, since the pixel number in the picture is proportional to the area and is 1:1, the total area of leaves on the first conveying belt 25 can be obtained by calculating the total pixel number of all green leaves in the picture, (4), In formula (4), S is the total area of leaves on the first conveying belt 25, H and W are the height and width (pixel dimension) of the picture respectively; is the total pixel number of green leaves; calculating the area Q of the first conveying belt 25 in the picture, Q = c * d (5), In formula (5), c is the width of the first conveying belt 25, and d is the length of the first conveying belt 25; Z is calculated by combining formula (4) and (5), Z = S / Q (6).

[0038] S4 calculates the leaf area ratio by summing the total area of leaves and the area of the first conveying belt 25, and then based on different leaf area ratios, first adjusts the first fan 15 to different gears through the control module, preliminarily removes part of the leaves that fall into the collection frame 2, and the remaining part of the leaves falls from the collection frame 2 to the first conveying belt 25, and then is conveyed to the second conveying belt 26, and then the second fan 27 relatively offset arranged on both sides of the second conveying belt 26 is driven to different gears through the control module, so that the second fan 27 blows away the remaining part of the leaves from both sides of the second conveying belt 26 from different directions, and then the leaves on the second conveying belt 26 are processed twice to remove the leaves, so that the leaves are separated from the second conveying belt 26 from both sides of the second conveying belt 26; In an embodiment, it specifically includes steps S4.1-S4.3; S4.1 If the leaf area ratio Z in step S3.3 is between [0%, 20%], then the first fan 15 and the second fan 27 are adjusted to the first gear through the control module; S4.2 If the leaf area ratio Z in step S3.3 is between [20%, 50%], then the first fan 15 and the second fan 27 are adjusted to the second gear through the control module; S4.3 If the leaf area ratio Z in step S3.3 is between [50%, 100%], then the first fan 15 and the second fan 27 are adjusted to the third gear through the control module.

[0039] In another embodiment, the step S4 comprises steps S4.1-S4.3, step S4.1 comprises: if the leaf area ratio Z in step S3.3 is between [0%, 20%], and the leaf area ratio Z is relatively increased compared with the leaf area ratio in the last detection time, the first fan is adjusted to the next gear of the gear in the last detection time, and if the leaf area ratio Z is relatively decreased compared with the leaf area ratio in the last detection time, the first fan is kept at the gear in the last detection time, and the second fan is adjusted to the first gear; Step S4.2 comprises: if the leaf area ratio Z in step S3.3 is between [20%, 50%], and the leaf area ratio Z is relatively increased compared with the leaf area ratio in the last detection time, the first fan is adjusted to the next gear of the gear in the last detection time, and if the leaf area ratio Z is relatively decreased compared with the leaf area ratio in the last detection time, the first fan is kept at the gear in the last detection time, and the second fan is adjusted to the second gear; S4.3 if the leaf area ratio Z in step S3.3 is between [50%, 100%], and the leaf area ratio Z is relatively increased compared with the leaf area ratio in the last detection time, the first fan is adjusted to the next gear of the gear in the last detection time, and if the leaf area ratio Z is relatively decreased compared with the leaf area ratio in the last detection time, the first fan is kept at the gear in the last detection time, and the second fan is adjusted to the third gear.

[0040] According to whether the leaves on the first conveying belt increase or decrease, it is determined whether the first fan is increased or kept at the original gear, so that the first fan can blow away most of the leaves, and the second fan is directly adjusted. S5 The second conveying belt 26 continues to convey the fruits to the collecting box 28 arranged on the vehicle frame 1, and the fruits separated from the leaves are collected.

[0041] The working principle of the present application is as follows: in the process of collecting fruits, part of the leaves falling into the collecting frame 2 are preliminarily removed by the first impurity removing mechanism, and the fruits and the remaining part of the leaves in the collecting frame 2 fall from the lower part of the collecting frame 2 to the first conveying belt 25, then the first conveying belt 25 and the leaves falling onto the first conveying belt 25 are photographed in real time by the image acquisition device 6, the leaves in the photographed picture are recognized by the control module, and then the proportion of the total leaf area in the picture to the area of the first conveying belt 25 is calculated, so that the control module drives the second impurity removing mechanism to be at different gears based on the calculated different proportions, thereby blowing the leaves conveyed from the first conveying belt 25 to the second conveying belt 26 away from the second conveying belt 26, so as to ensure that the fruits on the second conveying belt 26 are separated from the leaves, thereby effectively reducing the leaves with the fruits on the second conveying belt 26 to be conveyed into the collecting box 28, and avoiding the time-consuming and laborious process of removing the leaves from the collected fruits again, and effectively improving the subsequent transportation process.

Claims

1. An automatic high-yield fruit picking and collecting method, wherein picked fruits are collected by a fruit collecting machine, the fruit collecting machine comprising a mechanical arm, a vehicle frame, and a collecting frame arranged on the mechanical arm, the mechanical arm arranged at one end of the vehicle frame is provided with an image collecting device, the collecting frame is provided with a first impurity removing mechanism, and the upper portion of the collecting frame is provided with a first driving mechanism and a collecting assembly. It also includes the following steps: S1 drives the robotic arm and collection box to transport them to the area below the fruit to be collected. Then, the first drive mechanism opens the collection component to form a funnel-shaped collection umbrella, allowing the falling fruit to fall into the collection box along the collection umbrella. S2 The first drive mechanism drives the collection box to swing at a preset angle on the frame. During and after the swing, the first cleaning mechanism removes the leaves that fall into the collection box along with the fruit, so that some leaves in the collection box detach from one side of the collection box. After the fruit entering the collection box falls onto the first conveyor belt set below the collection box, the image acquisition device takes a real-time picture of the first conveyor belt, identifies the leaves in the captured picture, and calculates the total area of ​​the leaves on the first conveyor belt in the picture. S4 calculates the leaf area ratio by combining the total area of ​​the leaves with the area of ​​the first conveyor belt. Then, based on the different leaf area ratios, it drives the second impurity removal mechanism, which is relatively misaligned on both sides of the second conveyor belt, to different gears to remove the leaves conveyed to the second conveyor belt again, so that the leaves are separated from the second conveyor belt from both sides. The second conveyor belt of S5 continues to transport the fruit to the collection box set on the frame.

2. A method in accordance with claim 1, wherein: The robotic arm includes a first fixed arm, a second fixed arm, and a third fixed base. One end of the first fixed arm is connected to the frame, and both ends of the second fixed arm are connected to the first fixed arm and the third fixed base, respectively. An image acquisition device is connected to the second fixed arm. A fixed frame is provided on the outside of the collection frame and is hinged to the collection frame. One end of the fixed frame is connected to the third fixed base. A third drive mechanism located between the collection frame and the fixed frame drives the collection frame to rotate around the hinge point between the collection frame and the fixed frame.

3. An automatic high impact lead generation method as claimed in claim 2, wherein: The third driving mechanism has two parts, located on both sides of the collection frame. The third driving mechanism includes a third driving cylinder and a third push rod. The third push rod is internally connected to the third driving cylinder. One end of the third driving cylinder is hinged to the fixed frame. One end of the third push rod is hinged to the outside of the collection frame. The bottom of the collection frame protrudes downward to form a funnel with a funnel opening. The first impurity removal mechanism located in the cavity includes a partition net and a first fan. The partition net, fixedly installed in the cavity, divides the cavity into a first cavity and a second cavity. The second cavity communicates with the funnel opening. The partition net has two or more partition strips, forming a gap between the partition strips. The first cavity communicates with the second cavity through the gap. The first fan is fixedly installed in the first cavity and located on the side closer to the third driving mechanism. The collection frame has an opening on the side opposite to the first fan, which communicates with the first cavity.

4. An automatic high impact lead generation method as claimed in claim 1, wherein: The first driving mechanism includes a first driving cylinder, a first push rod, and a rack. The first driving cylinder is connected to the first push rod, and one end of the first driving cylinder is fixedly connected to the upper end of the collection frame. One end of the rack is fixedly connected to the first push rod. The collection assembly includes two or more gears and a connecting rod. One end of the connecting rod is rotatably connected to the outer edge of the upper end of the collection frame, and the other end of the connecting rod is inclined upward and outward. A collection cloth is connected between the connecting rods. The other end of the rack is arranged across the collection frame and meshes with the gear fixedly sleeved on the connecting rod.

5. An automatic high impact lead generation method as claimed in claim 1, wherein: The second conveying belt is arranged on the vehicle frame through the second support, the collecting box is arranged below one end of the second conveying belt, the other end of the second conveying belt is arranged upwardly inclined relative to the vehicle frame, the second impurity removing mechanism fixed on the second support is arranged on both sides of the second conveying belt in a relative staggered manner, the second impurity removing mechanism is a second fan, and the first fan and the second fan are both provided with adjustable different gears.

6. An automatic high impact real-time recovery method as claimed in claim 5, wherein: The inclination angle of the second conveying belt relative to the vehicle frame is 5°-10°, and the angle between the first conveying belt and the second conveying belt is 145°-160°.

7. An automatic high impact lead generation method as claimed in claim 1, wherein: In the step S2, the first fan is started, and the first fan is adjusted to the first gear, and the first fan blows the part of the leaves falling into the collecting box out of the first cavity along the length direction of the separation strip of the separation net through the opening.

8. An automatic high impact lead generation method as claimed in claim 1, wherein: The step S3 includes steps S3.1-S3.3, S3.1, calculating the gray value of the obtained picture, Gray(i,j)=0.299*R(i,j)+0.587*G(i,j)+0.114*B(i,j)(1), in formula (1), Gray(i,j) is the gray value of the picture, R(i,j) is the red component value of the original pixel in the picture, G(i,j) is the green component value of the original pixel in the picture, and B(i,j) is the blue component value of the original pixel in the picture; S3.2, performing convolution operation on the gray value through the Gaussian filter kernel to realize Gaussian filter processing on the picture, （2）， in formula (2), L[i,j] is the value of the Gaussian kernel at position (i, j), represents the weight, is used for convolution operation in image processing, [i,j] is the position in the Gaussian kernel matrix, wherein i represents the row index, j represents the column index, and the index starts from 0; σ is the preset standard deviation of the Gaussian distribution; exp is an exponential function, k is related to the center position of the Gaussian kernel, and the center of the Gaussian kernel is located at (k, k); S3.3, calculating the green feature value T of each pixel in the picture located in the first conveying belt region based on the difference between the green leaves and the non-green fruits, T(i,j)=2.2*G(i,j)-1.1*B(i,j)-1.1*R(i,j)(3); Then, the green feature value T is compared with the preset green feature value K, if the green feature value T is greater than or equal to the preset green feature value K, L[i,j]=1, and it is judged that the position pixel is a green leaf pixel; if the green feature value T is less than the preset green feature value K, L[i,j]=0, and it is judged that the position pixel is a non-green leaf pixel; calculating the total pixel quantity of all green leaves in the picture, since the pixel quantity in the image is proportional to the area and is 1:1, the total leaf area on the first conveying belt can be obtained by calculating the total pixel quantity of all green leaves in the picture, （4）， S is the total area of leaves on the first conveyor belt in formula (4), H and W are the height and width of the picture, respectively; is the total number of pixels of the green portion feature T. calculating the area Q of the first conveying belt in the picture, Q=c*d(5), In formula (5), c is the width of the first conveying belt, and d is the length of the first conveying belt; The leaf area ratio Z is calculated by formula (4) and (5) simultaneously, Z=S / Q (6).

9. An automatic high impact lead generation method as claimed in claim 1, wherein: The step S4 comprises steps S4.1-S4.3, S4.1 If the leaf area ratio Z in step S3.3 is between 0% and 20%, the first fan and the second fan are both adjusted to the first gear; S4.2 If the leaf area ratio Z in step S3.3 is between 20% and 50%, the first fan and the second fan are both adjusted to the second gear; S4.3 If the leaf area ratio Z in step S3.3 is between 50% and 100%, the first fan and the second fan are both adjusted to the third gear.

10. The method of claim 1, wherein: The step S4 comprises steps S4.1-S4.3, and step S4.1 comprises that, if the leaf area ratio Z in step S3.3 is between 0% and 20%, and the leaf area ratio Z relatively increases compared with the leaf area ratio in the last detection time, the first fan is adjusted to the next gear of the gear in the last detection time, and if the leaf area ratio Z relatively decreases compared with the leaf area ratio in the last detection time, the first fan is kept at the gear in the last detection time, and the second fan is adjusted to the first gear; Step S4.2 comprises that, if the leaf area ratio Z in step S3.3 is between 20% and 50%, and the leaf area ratio Z relatively increases compared with the leaf area ratio in the last detection time, the first fan is adjusted to the next gear of the gear in the last detection time, and if the leaf area ratio Z relatively decreases compared with the leaf area ratio in the last detection time, the first fan is kept at the gear in the last detection time, and the second fan is adjusted to the second gear; S4.3 If the leaf area ratio Z in step S3.3 is between 50% and 100%, and the leaf area ratio Z relatively increases compared with the leaf area ratio in the last detection time, the first fan is adjusted to the next gear of the gear in the last detection time, and if the leaf area ratio Z relatively decreases compared with the leaf area ratio in the last detection time, the first fan is kept at the gear in the last detection time, and the second fan is adjusted to the third gear.

Images