Visual identification and picking sequence planning method for oranges on trees and simulation system thereof

A visual recognition, citrus technology, applied in character and pattern recognition, image data processing, instruments, etc., can solve the problem of not considering the size of the fruit, not designing the verification of the robot picking test method, etc., to save the overall picking time and shorten the movement distance. , the effect of high accuracy

Active Publication Date: 2020-02-28
SOUTH CHINA AGRI UNIV
View PDF3 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Scholars at home and abroad have carried out related research on multi-objective picking order planning for picking robots, Zion B. (Harvest-order planning for a multiarm robotic harvester[J].Computers and Electronics in Agriculture,2013,103(2):75–81.) A method for target recognition and multi-machine task assignment of melon picking manipulators is proposed, and the optimal harvesting order of manipulators is determined according to the combination of robot speed, number of arms, manipulator lateral acceleration and fruit processing time; but this method is mainly aimed at the cooperation of multiple manipulators The allocation of picking tasks in real time, there is no research on picking task planning for a single manipulator
Teng Juyuan (Research on motion and task planning of six-degree-of-freedom picking manipulator [D]. Chongqing: Chongqing University of Technology, 2017.) Researched on the task planning strategy of picking multiple fruits continuously by the picking manipulator, and proposed a method based on "minimum joint angle weighting" The task planning method realizes the energy-saving effect of multi-tasking picking manipulator; but this method does not consider factors such as the size of the fruit, and does not design an actual robot picking experiment for method verification

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Visual identification and picking sequence planning method for oranges on trees and simulation system thereof
  • Visual identification and picking sequence planning method for oranges on trees and simulation system thereof
  • Visual identification and picking sequence planning method for oranges on trees and simulation system thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0037] Visual recognition and picking sequence planning methods for citrus on trees, such as figure 2 As shown, the specific steps are as follows:

[0038] (1) Citrus recognition SVM model training: used to identify the area blocks with citrus. Firstly, the citrus image is cut into multiple fixed-size blocks, and then the block containing citrus is extracted as a positive sample, and the rest are negative samples as a training set. By analyzing the color space and texture information of citrus, extracting representative color features and texture features for training, the citrus block classification model module is obtained.

[0039] (2) SVM model training for extracting citrus pixels: it is used to judge the pixels belonging to citrus in the citrus area block. Select several regional blocks with citrus from the above positive samples, mark the citrus regional blocks, mark the citrus in the regional block as white, and mark the others as black to obtain the label image; co...

Embodiment 2

[0043] The simulation system for the visual recognition and picking sequence planning of citrus on a tree adopts the above-mentioned method for visual recognition and picking sequence planning of citrus on a tree, and specifically includes the following steps: performing three-dimensional modeling according to the actual experimental scene, and then according to the vector The three-dimensional coordinates of the citrus area located by the robot fruit identification subsystem, determine the position information of the citrus model corresponding to the simulation system, combine the size of the citrus model and the picking rules of the simulated manipulator, and determine the three-dimensional coordinates of each picking point in the simulation system; then plan the subsystem according to the picking sequence The obtained citrus picking sequence numbers are used to sort the picking points, such as figure 1 As shown, by sequentially calculating the coordinate difference between t...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a visual identification and picking sequence planning method for oranges on trees and a simulation system thereof, and the method comprises the steps: carrying out the image segmentation of the oranges on the trees through an image depth segmentation subsystem, and obtaining a depth segmentation image; a vector machine fruit recognition subsystem is adopted to recognize thedepth segmentation image, different description information is obtained, a citrus region information matrix is formed, and a citrus region which is correctly recognized in the image is obtained; andthen a picking sequence planning subsystem is adopted to carry out picking sequence planning on the fruit targets in the citrus region to obtain the sequential picking sequence numbers of the citrus on the tree so as to guide the picking manipulator to orderly pick the citrus fruits on the tree. According to the method and the simulation system thereof, the requirements for accuracy, orderliness and real-time performance of multi-target picking of fruits on trees can be met, the movement distance of the picking manipulator is effectively shortened, and the overall picking time is saved.

Description

technical field [0001] The invention relates to the field of fruit picking, in particular to a visual recognition and picking sequence planning method for citrus on a tree and a simulation system thereof. Background technique [0002] When the picking robot picks the fruit, it must first identify the location of the fruit. In the case of multiple fruits on the tree, it also needs to plan the order of picking. The traditional visual segmentation method can complete the recognition task under limited conditions, but its stability is low, and it cannot overcome the problems of citrus recognition on the tree, such as being greatly affected by external environmental factors, having many interference targets, and high real-time requirements. In the case of multi-objective picking, reasonable picking sequence planning can greatly improve the operating efficiency of the picking robot and save energy consumption. Scholars at home and abroad have carried out related research on multi...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/34G06K9/46G06T17/00G06T19/00
CPCG06T17/00G06T19/003G06V20/20G06V10/267G06V10/56Y02P90/30
Inventor 熊俊涛钟灼郑镇辉李中行卜榕彬陈淑绵
Owner SOUTH CHINA AGRI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap