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

Moving target detection algorithm based on depth and color image fusion

A color image, moving target technology, applied in image analysis, complex mathematical operations, image data processing and other directions, can solve problems such as difficulty in obtaining better results, and achieve the effect of improving accuracy and performance

Inactive Publication Date: 2022-08-02
JIANGSU COLLEGE OF INFORMATION TECH
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Among them, moving target detection is one of the basic and important research directions in computer vision image processing. This kind of image processing technology is often used in intelligent monitoring. The existing moving target detection methods only rely on the gray level features of pixels to In the case of sufficient illumination, the detection effect is good, but when the scene is insufficiently illuminated, it is difficult to achieve better results only by relying on gray-scale features

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
  • Moving target detection algorithm based on depth and color image fusion
  • Moving target detection algorithm based on depth and color image fusion
  • Moving target detection algorithm based on depth and color image fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] Embodiment 1: a moving target detection algorithm based on depth and color image fusion, mainly includes two main steps, 1. training phase, 2. detection phase, and the specific steps of the entire algorithm are as follows:

[0047] 1) training phase;

[0048] Step 1.1, obtain color image sequence I 1 and the depth image sequence D 1 , the sequence contains 10 images;

[0049] Step 1.2, use image registration to obtain the aligned color image sequence I 1 and the depth image sequence D 1 ;

[0050] Step 1.3. Using H.Cai, K.Hamm, L.Huang, J.Li and T.Wang, "Rapid RobustPrincipal Component Analysis: CUR Accelerated Inexact Low Rank Estimation," in IEEE Signal Processing Letters, vol.28, pp.116 -120, 2021, doi: 10.1109 / LSP.2020.3044130. The method in the depth image sequence D 1 Process to get candidate foreground sequence Dc 1 ;

[0051] Step 1.4, according to the coordinate information, to the candidate foreground sequence Dc 1 The synthetic gradient matrix G of 1 ...

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 relates to the technical field of moving target detection, in particular to a moving target detection algorithm based on depth and color image fusion, which comprises the following steps: firstly, carrying out model training, then judging and outputting a result according to a trained model, and in the training process, obtaining a color image sequence I1 and a depth image sequence D1; a color image sequence I1 and a depth image sequence D1 which are aligned are obtained through image registration, a candidate foreground sequence Dc1 is obtained through calculation, a synthetic gradient matrix G1 is obtained according to coordinate information, the color image sequence I1 is segmented through super-pixel segmentation, and area segmentation information A1 is obtained; and performing mean value, variance and histogram feature extraction on abnormal values in the candidate foreground sequence Dc1 and the governor of the synthetic gradient matrix G1 to form vectors and perform classification so as to obtain a classification model, and in a detection stage, using a training model to discriminate a result and outputting the result.

Description

technical field [0001] The invention relates to the technical field of moving target detection, in particular to a moving target detection algorithm based on depth and color image fusion. Background technique [0002] Image processing technology is a technology for processing image information by computer, mainly including image digitization, image enhancement and restoration, image data coding, image segmentation and image recognition, etc. Since images are the main source of human acquisition and exchange of information, images The application field of processing must involve all aspects of human life and work, including aerospace and aviation technology, biomedical engineering, communication engineering, industry and engineering, military and public security, culture and art, etc. With the continuous expansion of human activities, image The field of application of processing will also continue to increase. [0003] Among them, moving object detection is one of the basic ...

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
IPC IPC(8): G06T7/00G06T7/20G06T7/32G06F17/16
CPCG06T7/0002G06T7/20G06T7/32G06F17/16G06T2207/10024
Inventor 李阳朱爱玺
Owner JIANGSU COLLEGE OF INFORMATION TECH