Dynamic feature point removing method and device in warehouse semi-structured environment

A semi-structured, dynamic feature technology, applied in neural architecture, image data processing, instruments, etc., can solve problems such as inability to accurately screen out system feature points, and achieve the effect of easy implementation, high precision, and simple processing methods

Pending Publication Date: 2021-11-19
BEIJING WUZI UNIVERSITY
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The technical problem solved by the embodiment of the present invention is that in the semi-structured environment of the warehouse, it is impossible to accurately screen out the problem of feature points that have a serious impact on the system

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
  • Dynamic feature point removing method and device in warehouse semi-structured environment
  • Dynamic feature point removing method and device in warehouse semi-structured environment
  • Dynamic feature point removing method and device in warehouse semi-structured environment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0086] This application proposes a method for removing dynamic feature points in a semi-structured warehouse environment based on a pyramid scene analysis network (a method for removing dynamic feature points in a semi-structured warehouse environment), which can be combined with a pyramid scene analysis network for feature detection and extraction. Optimization, eliminating dynamic feature points in the process of feature detection and extraction, has better accuracy and robustness in the warehouse semi-structured environment, and the main steps in the implementation process are as follows: image 3 , the specific detailed flow chart is as Figure 4 , Oriented FAST key point extraction diagram as shown in Figure 5 , the pyramid scene analysis network structure diagram is as follows Image 6 . By performing image semantic segmentation and feature detection and extraction on the sequence images collected by the camera, the dynamic target area obtained by image semantic segme...

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 embodiment of the invention provides a dynamic feature point removing method and device in a warehouse semi-structured environment. The method comprises the following steps: acquiring an image sequence; performing feature detection and extraction on the image sequence to obtain feature points of the image sequence; performing image semantic segmentation on the image sequence to obtain an area where dynamic targets are located; removing the feature points located in the area where the dynamic targets are located. The precision of the method in a warehouse semi-structured environment is higher than that in the prior art; a pyramid scene analysis network is adopted to deal with a warehouse semi-structured environment, and the processing method is simple and accurate; a camera is adopted as a sensor, the cost is low, and the patent is easy to implement.

Description

technical field [0001] The invention relates to the technical field of simultaneous positioning and mapping, in particular to a method and device for removing dynamic feature points in a semi-structured environment of a warehouse. Background technique [0002] At the same time, positioning and mapping technology has become a hot spot in the field of robot research. It means that a robot equipped with sensors can establish a model of the environment during the movement process and estimate its own movement without prior environmental information. If the main sensor is The camera is called visual simultaneous positioning and mapping. The visual simultaneous positioning and mapping based on the feature point method uses the sequential images collected by the camera for map construction and self-pose estimation, all relying on the image feature points obtained from the sequential images. [0003] The placement of warehouse shelves is usually regular, which is called a structured...

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): G06K9/34G06K9/46G06T7/66G06N3/04
CPCG06T7/66G06N3/045
Inventor 苏庆华傅景超李俊韬张娣娣陈雨吴昊
Owner BEIJING WUZI UNIVERSITY
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