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

Pedestrian detection method based on GMM background difference and combined features

A joint feature and pedestrian detection technology, applied in biometric feature recognition, image analysis, image data processing, etc., can solve the problems of low detection accuracy and slow video sequence detection speed, so as to improve pertinence, increase accuracy, reduce The effect of small detection times

Inactive Publication Date: 2017-08-29
KUNMING UNIV OF SCI & TECH
View PDF6 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The invention provides a pedestrian detection method based on GMM background difference and joint features. This method aims at the problem that the traditional machine learning + feature descriptor training classifier has a slow detection speed for video sequences, and the detection accuracy is low. A pedestrian detection method that combines the motion information of video image objects. First, the mixed Gaussian model is used to model the background of the video image. Detect the area, and then use the classifier trained by the joint feature to detect the area to be detected, and finally get the block diagram of the pedestrian

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 GMM background difference and combined features
  • Pedestrian detection method based on GMM background difference and combined features
  • Pedestrian detection method based on GMM background difference and combined features

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0026] Embodiment 1: as Figure 1-2 As shown, a pedestrian detection method based on GMM background difference and joint features,

[0027] First use the mixed Gaussian model to model the background of the video image, and then perform a differential operation with the current frame of the video after obtaining the background image to obtain the position of the moving foreground object, determine the area to be detected in the image, and then use the classifier trained by the joint feature to be detected The area is detected, and finally the block diagram of the pedestrian is obtained.

[0028] The concrete steps of described method are as follows:

[0029] Step1, collecting video sequence images;

[0030] Step2, apply the mixed Gaussian model method to the sequence image collected in step Step 1 to carry out background modeling, obtain the background image;

[0031] Step3. Perform difference calculation between the current frame image and the background frame image to obta...

Embodiment 2

[0039] Embodiment 2: as Figure 1-2 As shown, a pedestrian detection method based on GMM background difference and joint features, such as Figure 1-2 Shown: The pedestrian detection method of GMM background modeling difference and joint features, the algorithm is mainly divided into two parts, one part is to extract the motion features of moving objects in the video as the area to be detected, and the other part is to use joint features for the area to be detected The trained classifier detects whether a moving object is a pedestrian or not.

[0040] Such as Figure 1-2 Shown: The pedestrian detection method based on the GMM background modeling difference and joint features uses the background difference model to extract pedestrian motion features: use the GMM method to model the background in real time to obtain a background image, which can be updated at intervals , to attenuate the effects of light changes and small disturbances in the image.

[0041] Such as Figure 1...

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 GMM background difference and combined features, and belongs to the technical field of video monitoring and recognition. The method comprises the steps: modeling the background of a video image through a hybrid Gaussian model, obtaining a background image, carrying out the difference calculation with a current frame of a video, obtaining the position of a moving foreground object, determining a to-be-detected region of the image, carrying out the detection of the to-be-detected region through a classifier trained through the combined features, and finally obtaining a block diagram of a pedestrian. The method can effectively shorten the search region of a sliding window of the classifier and reduces the number of search times of the classifier. The method quickly obtains the specific position of the pedestrian under the condition of determining the moving object, and also has a better detection effect for a group of pedestrians. The experiment proves that the method, compared with a conventional method, can achieve a better effect.

Description

technical field [0001] The invention relates to a pedestrian detection method based on GMM background difference and joint features, belonging to the technical field of video monitoring and recognition. Background technique [0002] Pedestrians, as one of the most frequent objects in traffic and daily video scenes, are crucial to the research of automatic video recognition technology. Therefore, pedestrian detection technology has become a research hotspot in recent years, and it has developed rapidly in the fields of intelligent transportation and intelligent security. However, although some achievements have been made in pedestrian detection technology after more than ten years of development, there is still no pedestrian detection system that can be used in any environment and background. [0003] In recent years, the pedestrian detection method based on machine learning and feature design has become the mainstream research direction in the field of pedestrian detection. ...

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): G06K9/00G06K9/32G06K9/62G06T5/30G06T7/70
CPCG06T5/30G06T7/70G06V40/10G06V20/42G06V10/25G06V2201/07G06F18/2411G06F18/214
Inventor 吴建德郑锐周唯
Owner KUNMING UNIV OF SCI & TECH
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