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

Pedestrian Detection Method Based on Local Decorrelation Features

A pedestrian detection and decorrelation technology, applied in biometric identification, computer parts, instruments, etc., can solve problems such as insufficient filter abundance, reduce filter tolerance and other problems, achieve near real-time processing, short training and detection time Effect

Active Publication Date: 2021-09-28
UNIV OF SHANGHAI FOR SCI & TECH
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This will result in a less rich filter and reduce the tolerance of the filter

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 Detection Method Based on Local Decorrelation Features
  • Pedestrian Detection Method Based on Local Decorrelation Features
  • Pedestrian Detection Method Based on Local Decorrelation Features

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0018] In order to make the techniques, creative features, achievements and efficacy of the present invention, and the following examples are specifically described in conjunction with the accompanying drawings.

[0019] figure 1 It is a process diagram of a pedestrian detecting method based on partial entry-related feature in an embodiment of the present invention; and figure 2 It is a schematic diagram of a cut-head shoulder area in an embodiment of the present invention.

[0020] like figure 1 As shown, a pedestrian detection method based on a localized-related feature of the present invention includes the following steps:

[0021] Step 1. All sample images in the sample image are labeled, and all sample images after the label are detected as pedestrian detection data sets.

[0022] Step 2, 10 channel conversion processing for each sample image in the pedestrian detection data set to obtain the training sample data set.

[0023] In step two, 10 channel transform processing con...

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 present invention provides a pedestrian detection method based on local decorrelation features, comprising the following steps: mark all sample images in a sample image set with pedestrian areas, and use all the marked sample images as a pedestrian detection data set; train pedestrian detection Each sample image in the data set is processed by 10 channel transformations; the average value of the 10 transformed channels of all training sample images in the training sample data set is calculated as the average human body model; covariance is performed on the head and shoulders of all training sample images Extraction to achieve decorrelation processing, and the head and shoulders area in the average human body model is extracted by covariance to generate a matrix as a filter; the filter is applied to the training sample data set to obtain the final feature, and the final feature is input based on the decision tree In the AdaBoost classifier of AdaBoost classifier, to train the AdaBoost classifier; the filter and the classifier are used as the final generated detector to perform pedestrian detection on the image to be detected.

Description

Technical field [0001] The present invention relates to the field of computer vision and image processing, and more particularly to a pedestrian detecting method based on partial entry-related features. Background technique [0002] Pedestrian Detection is the use of computer vision techniques to determine whether or not there are pedestrians in the video sequence and give precise positioning. This technology can be applied to artificial intelligence systems, vehicle aided driving systems, intelligent robots, smart video surveillance, human behavior analysis, intelligent transportation, etc. [0003] The LDCF method is one of the most common methods commonly used in pedestrian-detected methods. The LDCF method is in the training phase through the hog transformation (ie, calculating the gradient amplitude between the image pixels based on the image pixel value. To the 360-degree retraction direction) and the 10 transform channel features generated by the LUV image channel (L, 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): G06K9/00G06K9/46G06K9/62
CPCG06V40/10G06V10/50G06F18/285G06F18/211G06F18/214
Inventor 孙一品李航
Owner UNIV OF SHANGHAI FOR SCI & TECH