CN110741790A - Multi-claw transplanting-sorting processing method for plug seedlings based on depth camera - Google Patents

Multi-claw transplanting-sorting processing method for plug seedlings based on depth camera Download PDF

Info

Publication number
CN110741790A
CN110741790A CN201910854298.5A CN201910854298A CN110741790A CN 110741790 A CN110741790 A CN 110741790A CN 201910854298 A CN201910854298 A CN 201910854298A CN 110741790 A CN110741790 A CN 110741790A
Authority
CN
China
Prior art keywords
seedling
depth
claw
plug
point cloud
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201910854298.5A
Other languages
Chinese (zh)
Other versions
CN110741790B (en
Inventor
刘继展
赵升燚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jiangsu University
Original Assignee
Jiangsu University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jiangsu University filed Critical Jiangsu University
Priority to CN201910854298.5A priority Critical patent/CN110741790B/en
Publication of CN110741790A publication Critical patent/CN110741790A/en
Application granted granted Critical
Publication of CN110741790B publication Critical patent/CN110741790B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
    • A01C—PLANTING; SOWING; FERTILISING
    • A01C11/00—Transplanting machines
    • A01C11/02—Transplanting machines for seedlings
    • A01C11/025—Transplanting machines using seedling trays; Devices for removing the seedlings from the trays
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
    • B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
    • B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/10—Terrestrial scenes

Abstract

The invention discloses a depth camera-based plug seedling multi-claw transplanting-sorting processing method, which relates to the field of agricultural robots. Aiming at automatic transplanting-sorting operation in industrialized seedling raising, a vertical visual angle and a horizontal visual angle are effectively compressed through a depth camera, single seedling-lump depth images in row conveying of plug seedlings are sequentially obtained frame by frame, single seedling-lump depth images are extracted from a multi-claw multi-seedling-lump continuous depth data stream, then three-dimensional depth templates are used for quick matching processing, quick elimination of pointer interference information and efficient detection of seedling lump damage are realized, and the continuous non-pause detection requirement of multi-claw seedlings-lump in tray changing and transplanting is effectively met.

Description

Multi-claw transplanting-sorting processing method for plug seedlings based on depth camera
Technical Field
The invention relates to the field of agricultural robots, in particular to a plug seedling multi-claw transplanting-sorting processing method based on a depth camera.
Background
The condition that seedlings in seedling raising hole trays grow unevenly is often faced in hole tray seedling tray changing and transplanting, seedling shortage and unhealthy seedlings generally occur, meanwhile, damage of seedling plants and seedling lumps can be caused in transplanting, seedling shortage of seedling raising hole trays can lead to 'blank grabbing' and 'missing planting', seedling shortage, unhealthy seedlings and damage in transplanting lead to emergence of seedlings unevenly after hole tray seedling transplanting, poor growth conditions, and then yield is influenced and waste is caused. The existing seedling shortage and unhealthy seedling detection based on machine vision is mainly carried out statically in a target plug tray, is only suitable for extremely small seedlings with branches and leaves without boundary crossing, has extremely poor detection effect on the commonly occurring blade boundary crossing, and has extremely limited application range and operation effect. Meanwhile, the information of key 'un-taken' and 'damage' operation cannot be fed back, so that the system is complex and the efficiency and the quality of the transplanting operation are seriously influenced.
Disclosure of Invention
The invention provides an image processing method for multi-claw transplanting-sorting of plug seedlings based on a depth camera, which effectively feeds back damage operation information in plug seedling multi-claw transplanting operation, efficiently eliminates pointers of damaged plug seedlings and interference, and improves the efficiency and quality of transplanting operation.
In order to solve the technical problems, the invention adopts the specific technical scheme that:
a multi-claw transplanting-sorting processing method for plug seedlings based on a depth camera is characterized in that when a multi-claw transplanting mechanism sorts multi-claw and multi-seedling one by one, a depth camera is triggered to scan field-of-view images, and damaged seedling lumps and pointers are quickly detected and removed through the processed field-of-view depth images.
Further, the triggering of the depth camera comprises first triggering and subsequent triggering, wherein the first triggering is realized by the field scanning triggering device, and the subsequent triggering is realized by the controller triggering the depth camera at intervals.
Furthermore, the field scanning trigger device comprises an induction sheet and a photoelectric sensor, the induction sheet is arranged at the inner end of a sliding block at the upper part of the module I, the photoelectric sensor is arranged on the side surface of the module I, and the distance from the photoelectric sensor to the outer side of the module I is L'; the module I is located the conveying face upper end.
Further, the process of acquiring the depth image of the field of view is as follows: determining an effective vertical viewing angle of the depth camera from the plug seedlings as
Figure BDA0002197860060000011
Effective horizontal viewing angle of
Figure BDA0002197860060000012
Depth camera field sweep with effective view angleDuring tracing, three-dimensional depth point cloud data sets of different frames are obtainedThe three-dimensional depth point cloud data set A comprises a pointer of a plug seedling state clamped by a single seedling claw in transplantation, a single seedling plant and single seedling lump information;
wherein in the vertical viewing angle
Figure BDA0002197860060000022
And
Figure BDA0002197860060000023
comprises the following steps:
Figure BDA0002197860060000024
wherein Z is the seedling lump fixed height of the plug seedlings of the same batch, HmaxThe maximum height of the plug seedlings in the same batch is obtained, delta H is the set plug seedling height allowance, and L is the closest horizontal distance between the depth camera and the plug seedlings to be measured;
wherein ε in the horizontal viewing angle is:
Figure BDA0002197860060000025
wherein W is the maximum span of the plug seedling leaves in the same batch, and delta W is the set span margin of the plug seedling leaves.
Further, the processing procedure of the depth image of the field of view is as follows: the method comprises the following steps of (1) utilizing an aperture disk seedling integrated automatic sorting, transplanting and reseeding system, clamping single seedling lumps without seedling plants by utilizing a single seedling taking claw according to the same operation parameters, and acquiring a three-dimensional depth point cloud data set B with the same view field by a camera; according to vertical angle of view
Figure BDA0002197860060000026
Segmenting the acquired three-dimensional depth point cloud dataset B into
Figure BDA00021978600600000225
And
Figure BDA00021978600600000226
wherein B is1Constructing a seedling lump three-dimensional depth point cloud collection template, B2Form a template of a three-dimensional depth point cloud containing pointers, B1In
Figure BDA0002197860060000029
B2In
Figure BDA00021978600600000210
In the multi-claw transplanting operation, the acquired three-dimensional depth point cloud data sets of different frames
Figure BDA00021978600600000227
Is divided into
Figure BDA00021978600600000212
Three-dimensional depth point cloud data setAnd
Figure BDA00021978600600000214
three-dimensional depth point cloud data set
Figure BDA00021978600600000229
Furthermore, the process of rapidly detecting and removing the damaged seedling lumps comprises the following steps: three-dimensional depth point cloud data set A1Template B of three-dimensional depth point cloud data set1Matching and comparing all pixel points in the same field of view, and when the pixel points are the same
Figure BDA00021978600600000216
A on a pixel1Depth value
Figure BDA00021978600600000233
And B1Depth valueThe coincidence ratio N is more than or equal to N0The seedling lumps are in a harmless health state; otherwise, the seedlings are considered to be damaged lump seedlings, and the seedlings are discarded; wherein N is0Is a preset anastomosis ratio threshold value.
Furthermore, the rapid detection and elimination process of the pointer is as follows: three-dimensional depth point cloud data set A2Template B of three-dimensional depth point cloud data set2Matching and comparing all pixel points in the same field of view, and when the pixel points are the same
Figure BDA00021978600600000219
A on a pixel2Depth valueAnd B2Depth valueError ratio of ≤ Δ N1Then A is2Each pixel point
Figure BDA00021978600600000222
Upper corresponding depth value
Figure BDA00021978600600000223
All are regarded as pointers and are removed quickly; wherein Δ N1Is a preset error ratio threshold value.
The invention has the beneficial effects that: aiming at automatic transplanting-sorting operation in industrialized seedling raising, the vertical and horizontal visual angles are effectively compressed by a depth camera, single seedling-lump depth images are sequentially obtained frame by frame when plug seedlings are conveyed in rows, the single seedling-lump depth images are extracted from a multi-claw multi-seedling-lump continuous depth data stream, and then a three-dimensional depth template is utilized for quick matching processing, so that quick removal of pointer interference information and efficient detection of seedling lump damage are realized, and the continuous non-pause detection requirement of multi-claw seedlings-lumps in tray changing and transplanting is effectively met.
Drawings
FIG. 1 is an overall structure of a transplanter in the present invention;
FIG. 2 is a schematic view of a field scan trigger apparatus according to the present invention;
FIG. 3 is a schematic view of the depth camera vertical field compression in accordance with the present invention;
FIG. 4 is a top view schematic of the depth camera horizontal field of view compression in the present invention;
FIG. 5 is a schematic diagram of a depth camera compressing a field of view to acquire an object in accordance with the present invention;
in the figure, 1 is a detection mechanism, 2 is a rack, 3 is a conveying surface, 4 is a module I, 5 is a multi-claw transplanting mechanism, 6 is a source hole tray, 7 is a module II, 8 is a target hole tray, 9 is a single-claw transplanting mechanism, 10 is a sliding block, 11 is a sensing piece, 12 is a photoelectric sensor, 13 is an end effector, 14 is a pointer, 15 is a seedling plant, 16 is a seedling lump, and S is a seedling lump1Seedling taking area, S2Transport region, S3Transplanting area, S4The detection area.
Detailed Description
The technical solution of the present invention is further described in detail with reference to the accompanying drawings and specific embodiments.
The invention relates to a depth camera-based image processing method for multi-claw transplanting-sorting of plug seedlings, which utilizes an integrated automatic sorting, transplanting and replanting system for plug seedlings, wherein the key structure design and the whole motion flow planning of the system are recorded in the patent of the integrated automatic sorting, transplanting and replanting system for plug seedlings and the implementation method (the publication number is CN 107371878A).
Aiming at the plug seedling integrated automatic sorting, transplanting and replanting system, the plug seedling multi-claw transplanting-sorting processing method based on the depth camera comprises a multi-claw multi-seedling non-stop one-by-one scanning view field triggering method and a rapid eliminating method for removing damaged seedling lumps and pointer interference information in seedling-lump sorting operation.
The field-of-view triggering method for multi-claw multi-seedling non-stop one-by-one scanning comprises a triggering device installation and scanning triggering method and a field-of-view compression method. As shown in fig. 1 and 2, the operation module of the plug seedling integrated automatic sorting, transplanting and replanting system mainly comprises a depth camera 1, a rack 2, a conveying surface 3, a module I4, a multi-claw transplanting mechanism 5 and a single-claw replanting mechanism 9. The end effectors 13 are equidistantly arranged on the multi-claw transplanting mechanism 5 at intervals S, the number of the end effectors 13 is 5 in the embodiment, the depth camera 1 is arranged at the position H' above the rack 2 on the inner side of the conveying surface 3, the closest horizontal distance between the depth camera 1 and the plug seedling to be measured is L, and the view field of the depth camera 1 is over against the plug seedling to be measured; the photoelectric sensor 12 is arranged on the side face of the linear module I4, the distance from the outer side of the linear module I4 is L', and the induction sheet 11 is arranged at the inner end of the sliding block 10 at the upper part of the linear module I4; in this example, H 'is 10cm, L is 25cm, L' is 75cm, and S is 12 cm.
As shown in fig. 2, the multi-claw transplanting mechanism 5 is arranged in the seedling taking area S1The seedling taking operation is completed, and the seedling is conveyed in a conveying area S through a linear module I42Is conveyed towards the target plug 8 at a fixed speed V and passes through the detection area S4The time induction sheet 11 touches the photoelectric sensor 12, and at this time, the 1 st visual field scanning operation of the depth camera 1 is triggered, and then the 4 visual field scanning operations are controlled by the timer, and the controller triggers the depth camera 1 to scan once at intervals of time t. As shown in fig. 4, each time of field scanning, the horizontal view angle center line of the depth camera 1 is over against the center line of the seedling lump 16, and the measured plug seedling keeps a centered state in each frame of depth image; after 5 times of scanning is finished, the depth camera 1 is set to be in a dormant state and waits for the next round of triggering; wherein the time t relation satisfies:
Figure BDA0002197860060000041
wherein S is the center distance between the adjacent end effectors 13, and V is the transfer area S of the multi-claw transplanting mechanism 52In the present embodiment, V is 1 m/s.
As shown in fig. 3 and 4, the horizontal visual angles of the depth camera 1 are symmetrical along the vertical central line of the seedling lump 16, the horizontal closest distance L from the center of the visual field of the depth camera 1 to the measured plug seedling is a fixed value when the visual field scanning is triggered, and the horizontal visual angle of the depth camera 1 is compressed into
Figure BDA0002197860060000049
The vertical viewing angle is epsilon, and the effective field of view of the depth camera 1 is determined by the plug seedlings, i.e. at the vertical viewing angle
Figure BDA00021978600600000410
And horizontal viewing angleScanning is carried out; wherein in the vertical viewing angle
Figure BDA0002197860060000043
And
Figure BDA0002197860060000044
the approximation is:
Figure BDA0002197860060000045
in the formula (2), Z is the fixed height of the seedling lump of the plug seedlings of the same batch, HmaxThe maximum height of the plug seedlings in the same batch is shown, Δ H is the set height margin of the plug seedlings, in this embodiment, L is 25cm, Z is 4cm, and H ismax＝12cm，ΔH＝1cm。
Wherein ε in the horizontal viewing angle is:
in the formula, WmaxThe maximum spreading of the leaves of the plug seedlings in the same batch is obtained, and delta W is the set spreading margin of the leaves of the plug seedlings, wherein in the embodiment, W is 3.5cm, and delta W is 0.5 cm; and the blade span is 2W<S, when the field of view is scanned, the adjacent blades do not overlap.
Since all the installation positions are fixed, the multi-claw transplanting mechanism 5 is arranged in the conveying area S2The film is conveyed at a constant speed V, and the interval of each field scanning is the same; when field scanning is triggered, the depth camera 1 takes a set vertical view angle
Figure BDA00021978600600000411
And horizontal viewing angle
Figure BDA0002197860060000047
Field of view scanning over a range, three dimensional depth of field acquiredDegree point cloud data set
Figure BDA0002197860060000048
The method comprises the information of a pointer 14 for picking up the plug seedling state by a single seedling picking claw in the transplanting, a single seedling plant 15 and a single seedling lump 16.
In order to realize the measurement and detection of the seedling plant 15 and the seedling lump 16, the pointer 14 is rejected in a three-dimensional depth point cloud data set A of different frames obtained by a field triggering and compressing method of multi-claw multi-seedling non-stop one-by-one scanning, as shown in FIG. 5, because the pointer 14 is partially shielded by blades and the difference between the number, the size and the pose of the blades of each plug seedling is large, the point cloud data of the pointer 14 is discontinuous and the pixel coordinates and the depth values of corresponding pixels are different.
Therefore, the plug seedling integrated automatic sorting, transplanting and replanting system is utilized, single seedling lumps 16 without seedling plants 15 are clamped by single seedling-taking claws according to the same operation parameters, and the three-dimensional depth point cloud data set B with the same compressed view field is obtained by the multi-claw multi-seedling non-stop one-by-one scanning view field triggering and compressing method. According to vertical angle of view
Figure BDA0002197860060000051
Segmenting the acquired three-dimensional depth point cloud data set B into
Figure BDA0002197860060000052
And
Figure BDA0002197860060000053
two moieties, wherein B1Form a seedling lump (16) three-dimensional depth point cloud set template, B2And forming a three-dimensional depth point cloud collection template containing the pointer 14.
In the multi-claw transplanting operation, the point cloud data sets with different frame depths obtained by the field triggering and compressing method of multi-claw multi-seedling non-stop one-by-one scanning
Figure BDA0002197860060000055
Is divided into
Figure BDA0002197860060000056
Three-dimensional depth point cloud data set
Figure BDA0002197860060000058
Figure BDA0002197860060000057
And
Figure BDA0002197860060000059
three-dimensional depth point cloud data set
Figure BDA00021978600600000510
Three-dimensional depth point cloud data set A1Template B of three-dimensional depth point cloud data set1Matching and comparing all pixel points in the same field of view, and when the pixel points are the same
Figure BDA00021978600600000511
A on a pixel1Depth value
Figure BDA00021978600600000512
And B1Depth value
Figure BDA00021978600600000513
The coincidence ratio N is more than or equal to N0The seedling lump 16 is in a harmless health state; otherwise, the seedlings are considered to be damaged lump seedlings, and the seedlings are discarded; wherein N is0Is a preset anastomosis ratio threshold value. Three-dimensional depth point cloud data set A2Template B of three-dimensional depth point cloud data set2Matching and comparing all pixel points in the same field of view, and when the pixel points are the same
Figure BDA00021978600600000518
A on a pixel2Depth valueAnd B2Depth value
Figure BDA00021978600600000515
Anastomosis ratio of≤ΔN1Then A is2Each pixel point
Figure BDA00021978600600000516
Upper corresponding depth value
Figure BDA00021978600600000517
Are all considered as pointers 14 and are quickly culled; wherein Δ N1Is a preset error ratio threshold value. In this example N0＝98％，ΔN1＝2％。
Due to the point cloud data set A in three-dimensional depth2The depth value of the area of the part of the pointer 14 covered by the leafAnd a template B2Depth value of middle corresponding area
Figure BDA0002197860060000054
There are significant differences that are effectively preserved. Completion of pair A by the above-described method2And eliminating the interference of the information of the pointer 14, and further performing multi-feature analysis of height, expansion and stem fracture in the information of the remaining seedling plants 15.
The above embodiments are only used for illustrating the design idea and features of the present invention, and the purpose of the present invention is to enable those skilled in the art to understand the content of the present invention and implement the present invention accordingly, and the protection scope of the present invention is not limited to the above embodiments. Therefore, all equivalent changes and modifications made in accordance with the principles and concepts disclosed herein are intended to be included within the scope of the present invention.

Claims (8)

1. A multi-claw transplanting-sorting processing method for plug seedlings based on a depth camera is characterized by comprising the following steps: when the multi-claw transplanting mechanism (5) sorts multi-claw and multi-seedling one by one, the depth camera (1) is triggered to scan the view field image, and the damaged seedling lump and the pointer are quickly detected and removed through the processed view field depth image.
2. The depth camera-based multi-claw transplanting-sorting processing method for plug seedlings according to claim 1, characterized in that: the triggering of the depth camera (1) comprises first triggering and subsequent triggering, wherein the first triggering is realized by a field scanning triggering device, and the subsequent triggering is realized by the controller triggering the depth camera at intervals.
3. The depth camera-based multi-claw transplanting-sorting processing method for plug seedlings according to claim 2, characterized in that: the visual field scanning trigger device comprises an induction sheet (11) and a photoelectric sensor (12), wherein the induction sheet (11) is arranged at the inner end of a sliding block (10) on the upper part of a module I (4), the photoelectric sensor (12) is arranged on the side surface of the module I (4), and the distance from the outer side of the module I (4) is L'.
4. The depth camera-based multi-claw transplanting-sorting processing method for plug seedlings according to claim 3, characterized in that: the module I (4) is located at the upper end of the conveying surface (3).
5. The depth camera-based multi-claw transplanting-sorting processing method for plug seedlings according to claim 1, characterized in that: the process of acquiring the field depth image comprises the following steps: determining an effective vertical viewing angle of the depth camera (1) from the plug seedlings as
Figure FDA0002197860050000014
Figure FDA0002197860050000015
Effective horizontal viewing angle of
Figure FDA0002197860050000013
When the depth camera (1) performs field scanning with an effective view angle, three-dimensional depth point cloud data sets of different frames are acquired
Figure FDA0002197860050000016
The three-dimensional depth point cloud data set A comprises a pointer (14) of a plug seedling state clamped by a single seedling claw in transplantation, a single seedling plant (15) and single seedling lump (16) information;
wherein in the vertical viewing angle
Figure FDA0002197860050000017
And
Figure FDA0002197860050000018
comprises the following steps:
Figure FDA0002197860050000011
wherein Z is the seedling lump fixed height of the plug seedlings of the same batch, HmaxThe maximum height of the plug seedlings in the same batch is defined, delta H is the set plug seedling height allowance, and L is the closest horizontal distance between the depth camera (1) and the plug seedlings to be detected;
wherein ε in the horizontal viewing angle is:
Figure FDA0002197860050000012
wherein W is the maximum span of the plug seedling leaves in the same batch, and delta W is the set span margin of the plug seedling leaves.
6. The depth camera-based multi-claw transplanting-sorting processing method for plug seedlings according to claim 5, characterized in that: the processing process of the field depth image comprises the following steps: the method comprises the following steps of (1) clamping single seedling lumps (16) without seedling plants (15) by using a plug seedling integrated automatic sorting, transplanting and replanting system with the same operation parameters and using a single seedling taking claw, and acquiring a three-dimensional depth point cloud data set B with the same view field by using a camera; according to vertical angle of view
Figure FDA0002197860050000026
Segmenting the acquired three-dimensional depth point cloud dataset B into
Figure FDA0002197860050000027
Andwherein B is1Form a seedling lump (16) three-dimensional depth point cloud set template, B2Form a three-dimensional depth point cloud set template containing a pointer (14), B1InB2In
In the multi-claw transplanting operation, the acquired three-dimensional depth point cloud data sets of different frames
Figure FDA0002197860050000024
Is divided intoThree-dimensional depth point cloud data set
Figure FDA0002197860050000025
And
Figure FDA00021978600500000210
three-dimensional depth point cloud data set
Figure FDA0002197860050000029
7. The depth camera-based multi-claw transplanting-sorting processing method for plug seedlings according to claim 6, characterized in that: the rapid detection and removal process of the damaged seedling lump is as follows: three-dimensional depth point cloud data set A1Template B of three-dimensional depth point cloud data set1Matching and comparing all pixel points in the same field of view, and when the pixel points are the same
Figure FDA00021978600500000211
A on a pixel1Depth valueAnd B1Depth valueThe coincidence ratio N is more than or equal to N0The seedling lump (16) is in a state of no damage to health; otherwise, the seedlings are considered to be damaged lump seedlings, and the seedlings are discarded; wherein N is0Is a preset anastomosis ratio threshold value.
8. The depth camera-based multi-claw transplanting-sorting processing method for plug seedlings according to claim 6, characterized in that: the rapid detection and elimination process of the pointer comprises the following steps: three-dimensional depth point cloud data set A2Template B of three-dimensional depth point cloud data set2Matching and comparing all pixel points in the same field of view, and when the pixel points are the same
Figure FDA00021978600500000214
A on a pixel2Depth valueAnd B2Depth value
Figure FDA00021978600500000218
Error ratio of ≤ Δ N1Then A is2Each pixel pointUpper corresponding depth value
Figure FDA00021978600500000216
All are regarded as the pointer (14) and are rejected fast; wherein Δ N1Is a preset error ratio threshold value.
CN201910854298.5A 2019-09-10 2019-09-10 Multi-claw transplanting-sorting processing method for plug seedlings based on depth camera Active CN110741790B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910854298.5A CN110741790B (en) 2019-09-10 2019-09-10 Multi-claw transplanting-sorting processing method for plug seedlings based on depth camera

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910854298.5A CN110741790B (en) 2019-09-10 2019-09-10 Multi-claw transplanting-sorting processing method for plug seedlings based on depth camera

Publications (2)

Publication Number Publication Date
CN110741790A true CN110741790A (en) 2020-02-04
CN110741790B CN110741790B (en) 2021-08-03

Family

ID=69276191

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910854298.5A Active CN110741790B (en) 2019-09-10 2019-09-10 Multi-claw transplanting-sorting processing method for plug seedlings based on depth camera

Country Status (1)

Country Link
CN (1) CN110741790B (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114273252A (en) * 2021-11-26 2022-04-05 山东安信种苗股份有限公司 Intelligent vegetable seedling grading method
CN116797601A (en) * 2023-08-24 2023-09-22 西南林业大学 Image recognition-based Huashansong growth dynamic monitoring method and system

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0712569A1 (en) * 1994-11-17 1996-05-22 YANMAR AGRICULTURAL EQUIPMENT Co., Ltd. Transplanter
CN103636333A (en) * 2013-12-18 2014-03-19 中国农业大学 Intelligent seed and seedling sorting and transplanting machine
CN104322187A (en) * 2014-10-23 2015-02-04 江苏大学 Plug seedling image acquiring device and method
CN105783819A (en) * 2016-03-10 2016-07-20 江苏大学 RGB-D-based automatic transplanting seedling growing condition-operation effect composite detection method
CN107371878A (en) * 2017-09-18 2017-11-24 江苏大学 Cultivation system and implementation method are mended in a kind of Plug seedling integrated form automatic sorting transplanting

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0712569A1 (en) * 1994-11-17 1996-05-22 YANMAR AGRICULTURAL EQUIPMENT Co., Ltd. Transplanter
CN103636333A (en) * 2013-12-18 2014-03-19 中国农业大学 Intelligent seed and seedling sorting and transplanting machine
CN104322187A (en) * 2014-10-23 2015-02-04 江苏大学 Plug seedling image acquiring device and method
CN105783819A (en) * 2016-03-10 2016-07-20 江苏大学 RGB-D-based automatic transplanting seedling growing condition-operation effect composite detection method
CN107371878A (en) * 2017-09-18 2017-11-24 江苏大学 Cultivation system and implementation method are mended in a kind of Plug seedling integrated form automatic sorting transplanting

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114273252A (en) * 2021-11-26 2022-04-05 山东安信种苗股份有限公司 Intelligent vegetable seedling grading method
CN116797601A (en) * 2023-08-24 2023-09-22 西南林业大学 Image recognition-based Huashansong growth dynamic monitoring method and system
CN116797601B (en) * 2023-08-24 2023-11-07 西南林业大学 Image recognition-based Huashansong growth dynamic monitoring method and system

Also Published As

Publication number Publication date
CN110741790B (en) 2021-08-03

Similar Documents

Publication Publication Date Title
CN102954762B (en) Method and system for measuring external characters of grafted seedlings based on machine vision
CN110741790B (en) Multi-claw transplanting-sorting processing method for plug seedlings based on depth camera
CN111462058B (en) Method for rapidly detecting effective rice ears
CN109380146B (en) Automatic measurement device and method for live pigs
CN108064560A (en) The automatic picker system of fruit and method based on Kinect depth of field cameras
CN112990103A (en) String mining secondary positioning method based on machine vision
Li et al. Image detection and verification of visual navigation route during cotton field management period
Huang et al. An automatic machine vision-guided grasping system for Phalaenopsis tissue culture plantlets
CN101940096B (en) System for sorting corn seed haploid
Changyi et al. Apple detection from apple tree image based on BP neural network and Hough transform
Yoshida et al. A tomato recognition method for harvesting with robots using point clouds
CN111727457A (en) Cotton crop row detection method and device based on computer vision and storage medium
CN108037123B (en) Precision detection method for sowing performance parameters of hybrid rice bowl body tray
Jin et al. Edge recognition and reduced transplantation loss of leafy vegetable seedlings with Intel RealsSense D415 depth camera
CN107464232B (en) Image detection method for planting quality of unmanned rice transplanter
CN114694047A (en) Corn sowing quality evaluation method and device
CN112464759A (en) Pot seedling disease monitoring-seedling picking system and method of semi-automatic transplanter
CN109509173A (en) A kind of method of counting and counting device of bacterial clump
Li et al. Design and experiment of intelligent sorting and transplanting system for healthy vegetable seedlings
CN108734054B (en) Non-shielding citrus fruit image identification method
CN107993243B (en) Wheat tillering number automatic detection method based on RGB image
CN112338898B (en) Image processing method and device of object sorting system and object sorting system
CN111369497A (en) Walking type tree fruit continuous counting method and device
CN206864487U (en) A kind of solar battery sheet SPEED VISION positioning and correction system
CN111886982B (en) Detection method of dry land planting operation quality real-time detection system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant