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

Pedestrian target detection method based on cascade optimization

A pedestrian target and detection method technology, which is applied in the fields of instruments, biological neural network models, character and pattern recognition, etc., to achieve the effect of increasing feature response, improving accuracy, and suppressing background noise

Pending Publication Date: 2020-07-03
深圳北航新兴产业技术研究院
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The problem to be solved by the technology of the present invention is: to overcome the existing deficiencies in pedestrian detection, provide a pedestrian target detection method based on cascade optimization, make full use of the advantages of the characteristics of each layer of the network, and eliminate the occlusion and target deformation in pedestrian detection. Difficult problems such as scale and complex background are collectively classified as difficult sample problems with inaccurate network judgments, so that each layer of the network feature pyramid can cascade optimize the samples, and finally achieve more accurate pedestrian detection results

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
  • Pedestrian target detection method based on cascade optimization
  • Pedestrian target detection method based on cascade optimization
  • Pedestrian target detection method based on cascade optimization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] Such as figure 1 Shown, the whole implementation process of the inventive method is as follows:

[0035]A kind of pedestrian detection method based on cascading optimization of the present invention, this method comprises following implementation steps:

[0036] Step 1: Candidate region generation based on unsupervised attention mechanism

[0037] The main framework of the region generation network part is the VGG16 network, such as figure 1 shown. Different from the initial design, this scheme adds an unsupervised attention mechanism design. The feature map F is generated after the image I is input into the network, and the present invention does not use the feature pyramid structure when generating candidate regions. The feature map generated by the VGG16 network is then input into a sub-network to generate its own weight w. The sub-network consists of a convolutional layer and an activation layer. The channel of the last convolutional layer is 1, and then passes ...

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 a pedestrian detection method based on cascade optimization, and belongs to the field of pedestrian detection in computer vision. The method comprises the following steps: firstly, generating different pedestrian candidate target areas by utilizing a candidate area generation network based on an unsupervised attention mechanism, then extracting candidate target areas fromdifferent feature layers by utilizing a designed optimization network, optimizing samples layer by layer, and finally obtaining accurate pedestrian target prediction.

Description

technical field [0001] The invention relates to a pedestrian target detection method based on cascade optimization, which belongs to the field of pedestrian detection in computer vision. Background technique [0002] Pedestrian detection technology is a basic problem in the field of computer vision, and it is widely used in autonomous driving, automatic robots, video surveillance and other fields. [0003] The current main challenges in pedestrian detection lie in occlusion, scale, complex background, object deformation, etc. First of all, there are many vehicles and pedestrians on the road, so the occlusion between pedestrians, the occlusion of pedestrians by buildings, cars, etc. may be more serious, and both intra-class occlusion and inter-class occlusion may bring challenges to the detector. Second, due to the inconsistent distance between the pedestrian and the camera, the scale range of the pedestrian presented in the image may vary greatly, and this scale problem may...

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): G06K9/00G06K9/32G06K9/62G06N3/04
CPCG06V40/10G06V10/25G06N3/045G06F18/2193
Inventor 冷彪郝杰
Owner 深圳北航新兴产业技术研究院
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