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

Method, system and device for recursive error detection and verification based on hog feature pedestrian detector

A pedestrian detection and verification method technology, applied in biometric identification, instrumentation, computing, etc., can solve the problem of false detection of pedestrian targets

Active Publication Date: 2021-03-23
DALIAN NATIONALITIES UNIVERSITY
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problem of false detection of pedestrian targets in vehicle-mounted monocular camera pedestrian detection, the present invention provides a pedestrian target recursive verification method for HOG 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
  • Method, system and device for recursive error detection and verification based on hog feature pedestrian detector
  • Method, system and device for recursive error detection and verification based on hog feature pedestrian detector
  • Method, system and device for recursive error detection and verification based on hog feature pedestrian detector

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0072] Outdoor single environmental personnel calibration box error inspection

[0073] In this embodiment, the outdoor single environment actually peers to detect the complexity of the image feature information to be complex by the output, and the present invention is applied to figure 2 As the input pedestrian detection system, the image to be detected, image 3 In order not to start recursive operation, there is an output result of the wrong pedestrian target box. Figure 4 Successfully remove the output results of the wrong pedestrian target box for recursive verification.

Embodiment 2

[0075] Indoor single case pedestrian calibration box error check

[0076] In this embodiment, the interior single condition is complicated to the output of the person's target block error detection, which is applied to the output. Figure 5 As the input pedestrian detection system, the image to be detected, Figure 6 In order not to start recursive operation, there is an output result of the wrong pedestrian target box. Figure 7 Successfully remove the output results of the wrong pedestrian target box for recursive verification.

Embodiment 3

[0078] Outdoor multi-person environmental personnel calibration box error inspection

[0079] In this embodiment, the outdoor multiplayer environment actually peers to detect the image characteristic information complex by the output, and the present invention is applied to Figure 8 As the input pedestrian detection system, the image to be detected, Figure 9 In order not to start recursive operation, there is an output result of the wrong pedestrian target box. Figure 10 Successfully remove the output results of the wrong pedestrian target box for recursive verification.

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 recursive error detection method, system and device based on HOG feature pedestrian detector belong to the technical field of vehicle-mounted assisted driving. The ratio of the sum of the pixel area and the pixel area, and the ratio falls into different value ranges, including at least one range, which means that the target data cannot be determined. If the ratio falls within this range, the pedestrian target corresponding to the coordinates of the pedestrian detection frame will be The image is used as the pedestrian target image to be verified after steps S2 to S4 are executed cyclically, until the ratio no longer falls within the range indicating that the target data cannot be determined. The effect is to calculate through recursive structure, so that the system can eliminate detection errors caused by factors such as the angle of the shooting lens and the uneven lighting of the shooting scene in practical applications.

Description

Technical field [0001] The present invention belongs to the field of in-vehicle auxiliary driving techniques, and in particular, the present invention relates to a pedestrian target verification method based on a recursive structure. Background technique [0002] With the arrival of the unmanned car era, pedestrian testing related treatment technology has also become one of the hot research directions. Pedestrian detectors based on HOG feature have high computational precision and lower complexity to become a hot tool in machine learning. This type of detector is numerous, and the structure is easy to understand, but because the image to be detected will cause a series of factors such as different scene differences, uneven light lines, intensive obstruction, etc., which can easily make the characteristics of the personnel. There is a misuse, and the accuracy of the detection system is reduced. The pedestrian detection method based on the HOG feature is performed by filtering the ...

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): G06K9/00
CPCG06V40/10
Inventor 杨大伟毛琳陈思宇张静
Owner DALIAN NATIONALITIES UNIVERSITY
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