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

Occlusion target recognition and positioning method based on binocular vision

A technology of target recognition and localization method, which is applied in the field of occlusion target recognition and localization based on binocular vision, which can solve the problems of poor recognition and localization effect of occluded objects, and achieve the goal of improving the recognition rate, accurate positioning and good robustness. Effect

Pending Publication Date: 2021-04-16
HANGZHOU DIANZI UNIV
View PDF1 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Aiming at the problem of poor recognition and positioning of occluded objects in the prior art, the present invention provides a binocular vision-based occlusion target recognition and positioning method, based on feature point detection and matching technology and binocular vision triangulation Features, optimize the recognition and positioning effect

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
  • Occlusion target recognition and positioning method based on binocular vision
  • Occlusion target recognition and positioning method based on binocular vision
  • Occlusion target recognition and positioning method based on binocular vision

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0025] A method for identifying and locating an occluded target based on binocular vision, comprising the following steps:

[0026] S1: Perform feature point detection on the image to be detected and the template image, and determine the main direction of the feature points.

[0027] Before the feature point detection, it also includes grayscale processing of the input image to be detected and the template image. And the size of the image to be detected is w×h, and the size of the template image of the target is w 1 × h 1 , where w 1 ≤ w, h ≤ h.

[0028] The process of feature point detection includes: using the SURF algorithm to define feature points using the Hessian matrix. The Hessian matrix is ​​a Hessian matrix, which is a second-order derivative matrix that can describe the local curvature of the function. The specific form is as follows:

[0029]

[0030] Among them, L xx , L xy , L yy is the second-order derivative of the Gaussian-filtered image g(σ) in all ...

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 an occlusion target recognition and positioning method based on binocular vision, and the method comprises the following steps: obtaining a to-be-detected image and a template image, carrying out the feature point detection, and determining the main direction of a feature point; constructing a feature descriptor, screening and matching the feature points of the template image and the feature points of the to-be-detected image for multiple times according to the feature descriptor, and performing template restoration on the position of the target object in the to-be-detected image; calling a corresponding expected positioning model from a preset template library, carrying out preset positioning point matching, and selecting an expected positioning point; and performing binocular positioning on the obtained expected positioning point by using a triangulation method to obtain a three-dimensional coordinate. According to the method, the non-shielded area of the recognized object is positioned through the steps, then binocular positioning is carried out through a triangulation method to obtain the three-dimensional coordinates of the object, the method has good robustness in a complex environment, and the shielded object can be accurately positioned.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a method for identifying and locating an occluded target based on binocular vision. Background technique [0002] Object recognition and localization technology is a prerequisite for indoor service robots to navigate and grasp. In an unstructured environment, there is often a problem of mutual occlusion between objects. When an indoor service robot is looking for a target object, the object to be recognized is very likely to be blocked by other objects, which will have a great impact on the recognition, positioning and grasping of the robot. Obstacles, occlusion recognition technology came into being. The occlusion recognition technology based on feature point detection and matching is widely used as a common recognition method. For example, the invention of publication number CN111310624A discloses a occlusion recognition method, device, computer equipment and storage medium. I...

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): G06K9/00G06K9/32G06K9/46G06K9/62G06T7/73
Inventor 张波涛王煜升谢佳龙吴秋轩何中杰
Owner HANGZHOU DIANZI UNIV