阅读说明：本技术 一种基于机器视觉的荸荠捡拾机 (Machine is picked up to water chestnut based on machine vision ) 是由 周勇 刘浩蓬 张国忠 吴泽栋 王鑫尧 刘沿 王宏波 黄琳 于 2021-08-06 设计创作，主要内容包括：本发明公开了一种基于机器视觉的荸荠捡拾机,包括行走系统、图像识别系统、捡拾系统、荸荠输送系统、收集仓和机架；所述行走系统为履带式底盘结构,用于行走及转向；所述图像识别系统置于机架上,用于识别荸荠并进行图像处理和数据传输；所述捡拾系统用于完成目标荸荠的抓取并传输给荸荠输送系统；所述荸荠输送系统用于将捡拾系统传输过来的荸荠输送到收集仓中；所述收集仓置于机架后端,用于储存已捡拾的荸荠。本发明可以实现对机械挖掘过后铺在地面上的荸荠进行识别并捡拾,结构简单,操作方便,提高了生产效率。(The invention discloses a water chestnut pickup machine based on machine vision, which comprises a walking system, an image recognition system, a pickup system, a water chestnut conveying system, a collection bin and a rack, wherein the image recognition system is used for recognizing water chestnuts; the walking system is of a crawler-type chassis structure and is used for walking and steering; the image recognition system is arranged on the rack and used for recognizing the water chestnuts and performing image processing and data transmission; the picking system is used for finishing grabbing of the target water chestnuts and transmitting the water chestnuts to the water chestnut conveying system; the water chestnut conveying system is used for conveying the water chestnuts conveyed by the picking system to the collection bin; the collecting bin is arranged at the rear end of the rack and used for storing the picked water chestnuts. The water chestnut picking machine can recognize and pick up water chestnuts paved on the ground after mechanical excavation, is simple in structure and convenient to operate, and improves production efficiency.)

1. The utility model provides a machine is picked up to water chestnut based on machine vision which characterized in that: comprises a walking system, an image recognition system, a picking system, a water chestnut conveying system, a collection bin (7) and a rack; the walking system is of a crawler-type chassis structure and is used for walking and steering; the image recognition system is arranged on the rack and used for recognizing the water chestnuts and performing image processing and data transmission; the picking system is used for finishing grabbing of the target water chestnuts and transmitting the water chestnuts to the water chestnut conveying system; the water chestnut conveying system is used for conveying the water chestnuts conveyed by the picking system to the collection bin (7); the collection bin (7) is arranged at the rear end of the rack and used for storing the picked water chestnuts.

2. The machine vision-based water chestnut pickup machine according to claim 1, characterized in that: the walking system comprises a power supply box (6), a stepping motor (16), crawler wheels (12), a crawler belt (13), a speed reducer (15) and a coupler (14), wherein the power supply box (6) is fixed on a front platform (4) of the rack, the stepping motor (16) is fixed at the bottom end of a crawler belt chassis, the stepping motor (16) is connected with a power input shaft of the speed reducer (15), a power output shaft of the speed reducer (15) is connected with the crawler wheels (12) through the coupler (14), and the crawler wheels (12) drive the crawler belt (13) to achieve the functions of advancing, retreating and steering.

3. The machine vision-based water chestnut pickup machine according to claim 2, characterized in that: the steering utilizes a differential principle, namely, the right crawler wheel forwards transmits when turning left, and the left crawler wheel reversely rotates; when the crawler belt wheel rotates rightwards, the crawler belt wheel on the left side is positively transmitted, and the crawler belt wheel on the right side rotates reversely.

4. The machine vision-based water chestnut pickup machine according to claim 2, characterized in that: the power box (6) is internally provided with a lithium battery.

5. The machine vision-based water chestnut pickup machine according to claim 1, characterized in that: the image recognition system comprises a recognition camera (18) arranged on a rack (17) at the front end of the rack and a fixed PC processing end (5) arranged at the upper part of a front platform (4) of the rack for image processing; the recognition camera (18) collects image information and sends the image information to the fixed PC processing end (5), the fixed PC processing end (5) carries out image processing to obtain the coordinates of the target water chestnut, and the real-time coordinate information is continuously transmitted to the embedded mainboard of the lower computer through the data acquisition card and is used for controlling the motion of the picking system.

6. The machine vision-based water chestnut pickup machine according to claim 1, characterized in that: the picking system comprises a parallel mechanical arm (2) and a flexible mechanical claw (1) which are arranged at the lower part of a front platform (4) of the frame, the parallel mechanical arm (2) is connected below the front platform (4) of the frame through a bolt, the flexible mechanical claw (1) is arranged at the tail end of the parallel mechanical arm (2), and the parallel mechanical arm and the flexible mechanical claw are connected through a bolt.

7. The machine vision-based water chestnut pickup machine according to claim 6, characterized in that: the motion control system runs the manipulator inverse solution program, the coordinate information of the target water chestnut is solved, the joint angles of the parallel manipulator (2) are obtained, and the solved angle information is sent to a manipulator steering engine in a PWM wave mode, so that the grabbing action of the target water chestnut is completed and the water chestnut is placed into the water chestnut conveying system.

8. The machine vision-based water chestnut pickup machine according to claim 1, characterized in that: water chestnut conveying system is including fixing driving motor (8) on the frame right side, front sprocket (9) that link to each other through the shaft coupling with driving motor (8), conveyer belt (3) of taking the recess, rear sprocket (11) and driving chain (10) of being connected with conveyer belt (3), rear sprocket (11) are connected with front sprocket (9) through driving chain (10), and the power that driving motor (8) produced passes through the shaft coupling and transmits for front sprocket (9), and front sprocket (9) drive rear sprocket (11) through driving chain (10), and rear sprocket (11) drive conveyer belt (3) make its operation.

9. The machine vision-based water chestnut pickup machine according to claim 1, characterized in that: the collection bin (7) is a rectangular box body and is arranged at the rear end of the crawler chassis.

10. The machine for picking up water chestnuts based on machine vision as claimed in any one of claims 1 to 9, wherein: the water chestnut pickup machine is used for identifying and picking up the water chestnuts which are mechanically excavated and then flatly paved on the ground surface; the water chestnut picking device further comprises a counter, wherein the counter is used for counting the number of picked water chestnuts, the alarm gives an alarm and stops working after the set number is reached, and the water chestnut picking device can be used for picking other crops.

Technical Field

The invention belongs to the technical field of agricultural machinery, relates to crop harvesting machinery, and particularly relates to a water chestnut pickup machine based on machine vision.

Background

The water chestnut is a herbaceous plant growing in shallow water, is usually eaten by underground corms, can be directly eaten as fruits, and is slightly sweet in taste, crisp and juicy; meanwhile, the food can be cooked or made into sweetmeat. The shape of the water chestnut is very similar to that of a water chestnut, so that the water chestnut is named as the water chestnut. The water chestnut has purple and black peel and white flesh, is cleaned and peeled off, is bitten by a bite, and feels that juice overflows all around, is sweet and refreshing.

The water chestnut harvesting period is generally determined to be about 30 days after stems and leaves above the ground are completely withered and yellow, and the optimal harvesting period is generally in the period of winter and spring every year. At this time, the peel of water chestnut is dark reddish brown, the expansion of underground corms is stopped, and the corms have low water content but much starch content. The sugar content of the water chestnuts is at the highest value, and the taste is sweet. If the water chestnuts are harvested in advance, the color of the water chestnuts is not completely red, and the water chestnuts are thin in peel, tender in meat quality but not sweet in taste, so that the water chestnuts are not suitable for storage. The specific harvesting method is that water in the field is dried within l 0-l 5 days before water chestnuts are to be harvested, and water chestnuts are harvested by forking after a mud layer is dried. Because the water chestnuts mainly grow in a soil layer 9-20 cm below the ground surface, soil on the upper layer of about 8-9 cm is scraped off by a fork during collection, then the soil layer below the water chestnuts is scraped off, and the corms are carefully kneaded out by hands after the water chestnuts are seen.

The main problems and drawbacks of the prior art include:

the mechanical digging of water chestnuts is a key ring in the production process of the water chestnuts, and at the present stage, the mechanical harvesting of the water chestnuts is divided into two types: 1. a water chestnut collector which sprays jet flow to wash out soil in a paddy field environment; 2. a water chestnut digger aiming at an anhydrous field environment. Wherein, the latter is the main development direction of the mechanical harvest of water chestnuts. However, the work which can be completed by the water chestnut digging machine can only reach the link of mechanically digging and paving the water chestnuts on the ground.

In order to better adapt to the current production situation and realize the whole-process mechanization of water chestnut harvesting, the invention of the water chestnut picking machine based on machine vision is particularly important. Moreover, the picking up of mechanically excavated crops such as potatoes, taros, sweet potatoes and the like can be realized by modifying corresponding programs, so that the multifunctional picking machine has multiple functions.

Disclosure of Invention

Aiming at the problems and the defects in the prior art, the invention provides the water chestnut pickup machine based on machine vision, which can realize the recognition and pickup of water chestnuts which are mechanically excavated and laid on the ground, has a simple structure and is convenient to operate, and the production efficiency is improved.

Therefore, the invention adopts the following technical scheme:

a machine vision-based water chestnut pickup machine comprises a walking system, an image recognition system, a pickup system, a water chestnut conveying system, a collection bin and a rack; the walking system is of a crawler-type chassis structure and is used for walking and steering; the image recognition system is arranged on the rack and used for recognizing the water chestnuts and performing image processing and data transmission; the picking system is used for finishing grabbing of the target water chestnuts and transmitting the water chestnuts to the water chestnut conveying system; the water chestnut conveying system is used for conveying the water chestnuts conveyed by the picking system to the collection bin; the collecting bin is arranged at the rear end of the rack and used for storing the picked water chestnuts.

Preferably, the traveling system comprises a power box, a stepping motor, crawler wheels, a crawler belt, a speed reducer and a coupling, the power box is fixed on the front platform of the frame, the stepping motor is fixed at the bottom end of a crawler belt chassis and connected with a power input shaft of the speed reducer, a power output shaft of the speed reducer is connected with the crawler wheels through the coupling, and the crawler wheels drive the crawler belt to achieve the functions of advancing, retreating and steering.

Preferably, the steering utilizes a differential principle, namely, the right crawler wheel positively transmits when turning left, and the left crawler wheel reversely rotates; when the crawler belt wheel rotates rightwards, the crawler belt wheel on the left side is positively transmitted, and the crawler belt wheel on the right side rotates reversely.

Preferably, a lithium battery is arranged in the power supply box.

Preferably, the image recognition system comprises a recognition camera arranged on a rack at the front end of the rack, and a fixed PC processing end arranged on the upper part of a front platform of the rack for image processing; the identification camera collects image information and sends the image information to the fixed PC processing end, the fixed PC processing end carries out image processing to obtain a target water chestnut coordinate, and the real-time coordinate information is continuously transmitted to the lower computer embedded main board through the data acquisition card and is used for controlling the motion of the picking system.

Preferably, the picking system comprises a parallel mechanical arm and a flexible mechanical claw which are arranged at the lower part of the front platform of the frame, the parallel mechanical arm is connected below the front platform of the frame through a bolt, the flexible mechanical claw is arranged at the tail end of the parallel mechanical arm, and the flexible mechanical claw and the parallel mechanical arm are connected through a bolt.

Preferably, the motion control system runs a manipulator inverse solution program, the coordinate information of the target water chestnut is solved, the angles of all joints of the manipulator are acquired and connected in parallel, and the solved angle information is sent to a manipulator steering engine in a PWM wave mode, so that the grabbing action of the target water chestnut is completed and the target water chestnut is placed into the water chestnut conveying system.

Preferably, the water chestnut conveying system comprises a driving motor fixed on the right side of the rack, a front chain wheel connected with the driving motor through a coupling, a conveying belt with a groove, a rear chain wheel connected with the conveying belt and a transmission chain, wherein the rear chain wheel is connected with the front chain wheel through the transmission chain, power generated by the driving motor is transmitted to the front chain wheel through the coupling, the front chain wheel drives the rear chain wheel through the transmission chain, and the rear chain wheel drives the conveying belt to run.

Preferably, the collection bin is a rectangular box body and is arranged at the rear end of the crawler chassis.

Preferably, the water chestnut pickup machine is used for identifying and picking up the water chestnuts which are mechanically excavated and then flatly paved on the ground; the water chestnut picking device further comprises a counter, wherein the counter is used for counting the number of picked water chestnuts, the alarm gives an alarm and stops working after the set number is reached, and the water chestnut picking device can be used for picking other crops.

Compared with the prior art, the invention has the beneficial effects that:

(1) the robot can replace manpower to identify and pick up the water chestnuts which are mechanically excavated and then flatly paved on the ground surface, the crawler-type chassis can adapt to most of working environments, the trafficability is good, the image identification system can accurately identify and position the water chestnuts on the ground surface, the manipulator is controlled to accurately pick up the target water chestnuts through the processing calculation of the fixed PC processing end, the water chestnut conveying system can convey the water chestnuts into the collection bin and count the water chestnuts in real time, the robot can automatically give an alarm and stop working when the bin is full, and the robot can pick up other crops, thereby realizing multiple purposes.

(2) The invention can greatly save manpower, improve production efficiency and improve the mechanization degree of water chestnut harvesting.

Drawings

Fig. 1 is a schematic structural diagram of a water chestnut picking machine based on machine vision provided by the invention.

Fig. 2 is a bottom view of the water chestnut picking machine based on machine vision provided by the invention.

Fig. 3 is a right side view of the water chestnut picking machine based on machine vision provided by the invention.

Description of reference numerals: 1. a flexible gripper; 2. parallel mechanical arms; 3. a conveyor belt; 4. a frame front platform; 5. fixing a PC processing end; 6. a power supply box; 7. a collection bin; 8. a drive motor; 9. a front sprocket; 10. a drive chain; 11. a rear sprocket; 12. a crawler wheel; 13. a crawler belt; 14. a coupling; 15. a speed reducer; 16. a stepping motor; 17. a camera frame; 18. the camera is identified.

Detailed Description

The present invention will be described in detail with reference to the accompanying drawings and specific embodiments, which are provided for illustration only and are not to be construed as limiting the invention.

A water chestnut pickup machine based on machine vision comprises a walking system, an image recognition system, a pickup system, a water chestnut conveying system, a collection bin 7 and a rack. The walking system is a crawler-type chassis, and the image recognition system comprises a recognition camera 18 arranged on a rack 17 at the front end of the rack and a fixed PC processing end 5 arranged at the upper part of a front platform 4 of the rack for image processing; the picking system comprises a parallel mechanical arm 2 and a flexible mechanical claw 1 which are arranged at the lower part of a front platform 4 of the frame, the parallel mechanical arm 2 is connected below the front platform 4 of the frame through a bolt, the flexible mechanical claw 1 is arranged at the tail end of the mechanical arm, and the flexible mechanical claw 1 and the flexible mechanical claw are connected through a bolt; the water chestnut conveying system comprises a driving motor 8 fixed on the right side of the rack, front chain wheels 9 connected with the driving motor 8 through a coupling, and a conveying belt 3 with a groove matched with the front chain wheels, wherein the two front chain wheels are fixed on the front side of the rack and can rotate along with the conveying belt; the collecting bin 7 is arranged at the rear end of the rack and mainly plays a role in bearing collected water chestnuts.

Specifically, the walking system comprises a power supply box 6, a stepping motor 16, crawler wheels 12, a crawler belt 13, a speed reducer 15 and a coupling 14, wherein the power supply box 6 is fixed on a front platform 4 of the frame and is internally provided with a lithium battery, the stepping motor 16 is fixed at the bottom end of a crawler belt chassis and is connected with a power input shaft of the speed reducer 15, a power output shaft of the speed reducer 15 is connected with the crawler wheels 12 through the coupling, and the crawler wheels 12 drive the crawler belt 13 to realize the functions of advancing, retreating and steering; the steering utilizes a differential principle, namely, the right crawler wheel forwards transmits when turning left, and the left crawler wheel reversely rotates; when the crawler belt turns right, the left crawler wheel forwards transmits, and the right crawler wheel reversely rotates.

Specifically, the image recognition system comprises a recognition camera 18 and a fixed PC processing end 5, wherein the recognition camera 18 is installed on a camera frame 17 at the front end of the frame, the fixed PC processing end 5 is located on the upper plane of a frame front platform 4, the recognition camera 18 collects image information and sends the image information to the fixed PC processing end 5, the fixed PC processing end 5 performs image processing to obtain a target water chestnut coordinate, and real-time coordinate information is continuously transmitted to a lower computer embedded mainboard through a data acquisition card, so that the motion control system can conveniently control the motion of the mechanical arm.

Specifically, the picking system comprises parallel mechanical arms 2 and a flexible mechanical arm 1, wherein the parallel mechanical arms 2 are connected below a front platform 4 of the frame through bolts, the tail ends of the mechanical arms are provided with flexible mechanical claws 1, and the flexible mechanical claws are connected through bolts. The motion control system runs the manipulator inverse solution program, the coordinate information of the target water chestnut is solved, the angle of each joint of the manipulator is obtained, the solved angle information is sent to a manipulator steering engine in a PWM wave mode, and therefore the grabbing action of the target water chestnut is completed and the target water chestnut is placed into a groove of the conveying belt 3.

Specifically, water chestnut conveying system includes driving motor 8, two preceding sprocket 9 that pass through the shaft coupling with driving motor and link to each other, conveyer belt 3, two back sprocket 11 in the recess, wherein, driving motor 8 passes through the bolt fastening on the frame right side, and driving motor 8's power is provided by the lithium cell in the power supply box 6, and back sprocket 11 of conveyer belt 3 passes through driving chain 10 and is connected with preceding sprocket 9, and the power that driving motor 8 produced passes through the shaft coupling and transmits for preceding sprocket 9, and preceding sprocket 9 passes through driving chain 10 and drives back sprocket 11, the back chain, 11 drive conveyer belt 3 makes its operation.

Specifically, the collection bin 7 is a rectangular box body, is arranged at the rear end of the crawler chassis and is used for storing picked water chestnuts.

Examples

As shown in fig. 1-3, a machine vision-based water chestnut picking machine comprises a walking system, a picking system, an image recognition system, a collection bin 7, a water chestnut conveying system and a rack.

The walking system comprises: the power supply box 6 is fixed on the front platform 4 of the frame, a lithium battery is arranged in the power supply box, the stepping motor 16 is fixed at the bottom end of a crawler chassis and is connected with a power input shaft of the speed reducer 15, a power output shaft of the speed reducer 15 is connected with the crawler wheel 12 through the coupling 14, the crawler wheel 12 drives the crawler 13 to realize advancing, retreating and steering functions, and the steering utilizes a differential principle, namely, the right crawler wheel forwards transmits when turning left, and the left crawler wheel reversely rotates; when the crawler belt turns right, the left crawler wheel forwards transmits, and the right crawler wheel reversely rotates.

The image recognition system includes: the identification camera 18 and the fixed PC processing end 5 are arranged, wherein the identification camera 18 is installed on a camera frame 17 at the front end of the frame, the fixed PC processing end is located on the upper plane of the frame front platform 4, the identification camera 18 collects image information and sends the image information to the fixed PC processing end, the PC end carries out image processing to obtain a target water chestnut coordinate, and real-time coordinate information is continuously transmitted to the lower computer embedded mainboard through the data collection card, so that the motion control system can conveniently control the motion of the parallel mechanical arm 2.

The pickup system includes: parallel mechanical arm 2, flexible manipulator 1, parallel mechanical arm 2 passes through bolted connection in frame front platform 4 below, and flexible gripper 1 is equipped with at the arm end, and both pass through bolted connection. The motion control system runs the manipulator inverse solution program, the coordinate information of the target water chestnut is solved, the angle of each joint of the manipulator is obtained, the solved angle information is sent to a manipulator steering engine in a PWM wave mode, and therefore the grabbing action of the target water chestnut is completed and the target water chestnut is placed into a groove of the conveying belt 3.

The water chestnut conveying system comprises: driving motor 8, the preceding sprocket 9 that links to each other through the shaft coupling with driving motor 8, take notched conveyer belt 3, the rear sprocket 11 of being connected with conveyer belt 3, wherein, driving motor 8 passes through the bolt fastening on the frame right side, driving motor 8's power is provided by the lithium cell in the power supply box 6, conveyer belt 3's rear sprocket 11 passes through driving chain 10 and is connected with preceding sprocket 9, the power that driving motor 8 produced passes through the shaft coupling and transmits for preceding sprocket 9, preceding sprocket 9 passes through driving chain 10 and drives rear sprocket 11, rear sprocket 11 drives conveyer belt 3 makes its operation, two preceding sprockets are arranged the frame front end in and are rotated along with the conveyer belt together.

The collection bin 7 is a rectangular box body and is arranged at the rear end of the crawler chassis and used for storing picked water chestnuts.

According to the crawler-type water chestnut picker, ground surface water chestnuts excavated by a machine can be identified and picked up, the power box 6 provides power for the two stepping motors 16 positioned at the bottom of the crawler chassis, the power provided by the stepping motors 16 is transmitted to the crawler wheels 12 through the speed reducer 15, the crawler wheels 12 drive the crawler 13 to move forward, turn, retreat and the like, and the differential principle is utilized for turning. The matched image recognition system is powered by a power supply box 6, a recognition camera 18 can rotate left and right to detect the position of a water chestnut target in real time, images shot by the recognition camera 18 are transmitted to a fixed PC processing terminal 5 in real time, the fixed PC processing terminal 5 performs image processing to obtain ball coordinates, real-time coordinate information is continuously transmitted to an embedded mainboard of a lower computer through a data acquisition card, after the lower computer obtains coordinate signals, a motion control system runs a manipulator inverse solution program to calculate the water chestnut coordinate information to obtain angles of joints of a mechanical arm, the calculated angle information is transmitted to a steering engine in a PWM wave mode, and grabbing and putting back of balls are completed. The counter counts the number of picked water chestnuts, the alarm gives an alarm and stops working after a certain number of picked water chestnuts are reached, and the unloading process is manually realized.

The invention is not the best known technology.

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and scope of the present invention are intended to be covered thereby.