Unlock instant, AI-driven research and patent intelligence for your innovation.

Target image capturing method and system for inspection robot

A technology for patrolling robots and target images, applied in the field of robotics, can solve problems such as low efficiency, high artificial complexity of geometric analysis methods, and poor performance of data-driven reasoning methods, so as to reduce workload, reduce inspection time, and improve inspection efficiency. The effect of inspection efficiency

Inactive Publication Date: 2021-07-23
JIAXING UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Through the above analysis, the existing problems and defects of the existing technology are: in the traditional methods to solve the problem of autonomous robot grasping, the geometric analysis method has high artificial complexity, the data-driven reasoning method performs poorly in complex scenes, and the automatic positioning and grasping point, the efficiency is low

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
  • Target image capturing method and system for inspection robot
  • Target image capturing method and system for inspection robot
  • Target image capturing method and system for inspection robot

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0083] The inspection robot target image capture method provided by the embodiment of the present invention is as follows: figure 1 As shown, as a preferred embodiment, such as image 3 As shown, the grabbing position detection module provided by the embodiment of the present invention uses a position detection program to detect the grabbing position of the target object of the inspection robot, including:

[0084] S201, collecting target RGB images and target Depth images of the target object under different viewing angles, wherein the target RGB images correspond to pixels in the target Depth image one by one;

[0085] S202. Input each target RGB image into a target object segmentation network for calculation, and obtain the RGB pixel area of ​​the target object in the target RGB image and the Depth pixel area of ​​the target object in the target Depth image;

[0086] S203. Input the RGB pixel area of ​​the target object to the grasping position generation network to obtain...

Embodiment 2

[0088] The inspection robot target image capture method provided by the embodiment of the present invention is as follows: figure 1 As shown, as a preferred embodiment, the grabbing position optimization module provided by the embodiment of the present invention uses the position optimization program to optimize the grabbing position of the detected inspection robot target object to obtain the optimal grabbing position, including:

[0089] Input the Depth pixel area of ​​the target object and the grasping position to the grasping position quality assessment network, and calculate the score of the optimal grasping position; select the grasping position corresponding to the highest score as the global optimum of the inspection robot Grab location.

Embodiment 3

[0091] The inspection robot target image capture method provided by the embodiment of the present invention is as follows: figure 1 As shown, as a preferred embodiment, such as Figure 4 As shown, the target image acquisition module provided by the embodiment of the present invention utilizes the image acquisition device carried by the inspection robot to capture the target reference image and the inspection image of the target object to be captured, and determine the target image, including:

[0092] S301, the inspection robot walks to a predetermined target position, and takes a target reference image of the target object;

[0093] S302, when the inspection robot moves for inspection, continuously shoot inspection images;

[0094] S303, the inspection robot compares and analyzes the inspection image and the target reference image, and if the inspection image overlaps with the target reference image, the inspection robot takes the inspection image as the target image.

[0095...

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 belongs to the technical field of robots, and discloses a target image capturing method and system for an inspection robot. The target image capturing system of the inspection robot comprises a capturing position detection module, a capturing position optimization module, a central control module, a target image acquisition module, an image feature extraction module, a network model construction module, a target image capturing module, a cloud storage module and an update display module. Through the capturing position detection module and the capturing position optimization module, the robot can automatically grab a target object at the optimal capturing position; and the convolutional neural network is trained by adopting the synthetic feature data, so that the over-fitting of specific data set distribution can be avoided, and the invention has the advantages of high reliability, strong robustness and good real-time performance. Meanwhile, by adopting the invention to inspect, the inspection time can be shortened, the inspection efficiency can be improved, the workload of operation and maintenance personnel can be reduced, the system complexity can be reduced, and the image recognition positioning precision requirement of the inspection robot can be met.

Description

technical field [0001] The invention belongs to the technical field of robots, and in particular relates to a method and system for grabbing target images of inspection robots. Background technique [0002] At present, in the field of intelligent robots, when an inspection robot performs an inspection task, it runs to the inspection target position according to the inspection route, stops and shoots the target object through its own shooting unit, and then continues to move to the next inspection target position , During the movement, the robot automatically recognizes the target without pictures taken before, reads and backs up the meter value or switch position information. For the robot autonomous grasping problem, traditional solutions mainly include geometric analysis methods and data-driven reasoning methods. However, the geometric analysis method has high artificial complexity, and the data-driven reasoning method performs poorly in complex scenes, and its efficiency...

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): G06K9/00G06K9/46G06K9/62G06N3/04
CPCG06N3/04G06V20/10G06V10/44G06N3/045G06F18/214
Inventor 朱耀东朱珠楼平
Owner JIAXING UNIV