Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Far infrared pedestrian detection method based on two-stage cascade segmentation

A pedestrian detection, far-infrared technology, applied in the research field of computer vision and intelligent transportation, can solve the problems of insufficient information such as color and texture, uneven heat, and the similarity of different regions that cannot be merged.

Active Publication Date: 2020-01-17
SOUTH CHINA AGRI UNIV
View PDF9 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

For infrared images, the information such as color and texture is insufficient, and the imaging of different parts of the same human body may show uneven heat due to factors such as clothing thickness and environment, so it is not possible to use grayscale features to distinguish different regions. Do similarity merge
[0005] Therefore, in summary, although some achievements have been made in infrared image pedestrian detection, it still cannot meet the requirements in practical applications. In order to be more suitable for practical applications, it is urgent to improve the detection accuracy and real-time detection make improvements

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
  • Far infrared pedestrian detection method based on two-stage cascade segmentation
  • Far infrared pedestrian detection method based on two-stage cascade segmentation
  • Far infrared pedestrian detection method based on two-stage cascade segmentation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0117] A far-infrared pedestrian detection method based on two-stage cascade segmentation, such as figure 1 shown, including the following steps:

[0118] The first step is to obtain the infrared image, perform two median filtering and one Laplace differential processing on the infrared image, that is, first perform a median filtering process on the infrared image, suppress the image background, and then perform Laplace differential processing Processing, strengthen the outline and edge of the target, and finally perform a median filter on the image to remove some noise strengthened after Laplace differential processing, and finally obtain the processed infrared image; median filter refers to a nonlinear smoothing The technology uses a sliding window to sort the pixel values ​​in the window, and replaces the pixel value of the center point of the window with the median value of the pixel value of the field, so that the pixel value of the field is closer to the real value, and ...

Embodiment approach

[0144] Its implementation method is as follows:

[0145] 1) sort the weights of the undirected graph in ascending order;

[0146] 2), S 0 is the initial segmentation state, that is, each vertex is regarded as a segmentation area;

[0147] 3), process one edge each time, repeat the operation of 4);

[0148] 4), according to the last S m-1 , select an edge e(v i ,v j ). if v i and v j Not in the same segmented area, take the weight of this edge w(e(v i ,v j )) Compared with the minimum intra-class difference IntD of the two segmentation regions where the two vertices are located, if w(e(v i ,v j ))m = S m-1 ;

[0149] 5) Finally, the desired segmented area is obtained.

[0150] Double Threshold Segmentation

[0151] The image is segmented using a global threshold and a local threshold. First, the image is initially segmented with a global threshold, and then the target area of ​​the initial segmentation is segmented twice with a local threshold.

[0152] Based on ...

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 far-infrared pedestrian detection method based on two-stage cascade segmentation. According to the method, the contour of an image can be enhanced by adopting median filtering and Laplace integral preprocessing; according to the rapid image segmentation method based on the graph model, regions with similar brightness in the image can be merged, so that each local region where an object possibly exists in the image can be listed; the combination of the double-threshold method and the rapid image segmentation method based on the graph model enables the segmentation result to be more accurate. Candidate regions based on priori knowledge hierarchy merging are listed according to characteristics of human body imaging. According to the method, the human body regions aremerged preferentially, the accuracy of the regions of interest is improved, the robustness and real-time performance of the system are effectively improved by adopting the aggregation channel characteristics and the high and low frequency infrared images, and the classification result is more accurate by using the two-stage cascade classifier.

Description

technical field [0001] The invention relates to the research fields of computer vision and intelligent transportation, in particular to a far-infrared pedestrian detection method based on two-stage cascade segmentation. Background technique [0002] With the advancement of science and technology, intelligent life is more and more yearned for by the public, and intelligent transportation is also widely concerned. Regarding pedestrian safety, the existing technologies include pedestrian protection devices and pedestrian protection bumpers (Chinese patent authorization announcement number: CN107021058A, authorization announcement date: August 08, 2017), etc., all belong to passive pedestrian protection methods, which play a certain protective role after traffic accidents occur, and cannot prevent accidents in advance. At present, one of the solutions to improve night driving safety is the night vision system. The night vision system is used to assist the driver to drive at nig...

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): G06T7/11G06T7/136G06T7/187G06K9/62
CPCG06T7/11G06T7/136G06T7/187G06T2207/10048G06T2207/20032G06T2207/30196G06T2207/20081G06F18/2411Y02T10/40
Inventor 王国华李露郑沐杭郑永森刘财兴
Owner SOUTH CHINA AGRI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com