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

A People Counting Method Based on Image Analysis and Deep Learning

A deep learning and people counting technology, applied in the field of image target recognition and deep learning, can solve the problems of large amount of calculation and slow speed, and achieve the effect of good versatility, fast speed, and reduced scale space search range

Active Publication Date: 2020-12-01
上海远洲核信软件科技有限公司
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the large amount of calculation and slow speed, it has not been widely used in monitoring scenarios with high real-time requirements.

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
  • A People Counting Method Based on Image Analysis and Deep Learning
  • A People Counting Method Based on Image Analysis and Deep Learning
  • A People Counting Method Based on Image Analysis and Deep Learning

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach

[0036] refer to figure 1 , a specific embodiment of the present invention comprises the following steps:

[0037] A. Perform pyramid model calculation on the input image to generate images with multiple resolutions and sizes;

[0038] B. Carry out window sliding on each layer of the pyramid, calculate the HOG feature value of the window area, and classify through the linear SVM classifier, and judge whether the window is a head and shoulders area;

[0039] C. For each head and shoulder area given in step B, extract the corresponding image, normalize it to the same set size, input it into the deep neural network, and obtain the classification output;

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 method for counting people based on image analysis and deep learning, comprising the following steps: A, calculating the pyramid model on the input image to generate images with multiple resolutions and sizes; B, on each layer of the pyramid Carry out window sliding, calculate the HOG eigenvalue of window area, and classify by linear SVM classifier, judge whether this window is head-and-shoulders area; C, for each head-shoulders area that provides in step B, extract corresponding image, return Convert to the same set size, input it into the deep neural network, and obtain the classification output; D. Perform non-maximum suppression on all head and shoulder windows in the output of step C to merge the overlap of adjacent areas and scales test results. The invention can improve the deficiencies of the prior art, and can achieve higher people counting performance at a faster speed.

Description

technical field [0001] The invention relates to the technical field of image target recognition and deep learning, in particular to a method for counting people based on image analysis and deep learning. Background technique [0002] Using computer vision technology to count the number of people in surveillance images or videos can be widely used in project scenarios such as stampede warning, traffic guidance, shop people flow assessment, and attendance rate statistics. However, existing people counting systems often have large errors in crowded environments. This is because there are usually a lot of occlusions in a crowded environment, resulting in the fact that the features of the parts below the shoulders of the human body can hardly be used reliably and effectively. However, if only feature extraction and positioning are performed on the head and shoulders, since the shape curve of the head and shoulders is relatively simple, traditional hand-designed feature extractio...

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 Patents(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06M11/00
CPCG06M11/00G06V20/53G06N3/045G06F18/2411
Inventor 黄建华俞启尧
Owner 上海远洲核信软件科技有限公司
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