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

SVM-AdaBoost algorithm-based pedestrian detection method

A pedestrian detection and algorithm technology, applied in computing, computer components, instruments, etc., can solve problems such as high error rate and time-consuming calculation, and achieve the effect of improving accuracy

Inactive Publication Date: 2017-05-10
JIUQUAN VOCATIONAL & TECHN COLLEGE
View PDF3 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a pedestrian detection method based on the SVM-AdaBoost algorithm for the problems of time-consuming calculation and high error rate in the traditional pedestrian detection method

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
  • SVM-AdaBoost algorithm-based pedestrian detection method
  • SVM-AdaBoost algorithm-based pedestrian detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0018] A pedestrian detection method based on the SVM-AdaBoost algorithm, comprising the following steps:

[0019] (1) Obtain video images of pedestrians on the road ahead through the installed camera.

[0020] The camera is used to collect video images of four situations: single pedestrian, multi-pedestrian, far view, and near view, and obtain pedestrian samples and branch pedestrian samples for connection output.

[0021] (2) Take the video image obtained in step (1) as a test sample, give a specific training sample S={(x1, y1),..., (xn, yn )} with n variables, loop t times, and set the SVM kernel Function parameters: ={ ...}, C={C1,C2...}; where (x1, y1), ..., (xn, yn) are the parameters of the training sample S, , C is the parameter value of the SVM kernel function, is a one-dimensional array, and C is a multidimensional array.

[0022] A variety of test samples in different situations are respectively input into the training samples, and after t cycles, the parame...

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 an SVM-AdaBoost algorithm-based pedestrian detection method which comprises the following steps: a video image of a road on which pedestrians are moving forward is obtained as a testing sample via an installed camera; a specific training sample is provided, circulation for t times is realized, parameters of an SVM kernel function are set, a weight of the training sample is set, an SVM weak classifier is trained via the weight of the training sample, and a sample training error rate is trained via the SVM weak classifier; if the training error rate is greater than 50%, re-training is performed; if the training error rate is smaller than 50%, a weight of the SVM weak classifier is reset, and a weight of a sample set is reset; after a specific accuracy of 0.3-0.5 is reached after t times of circulation, the trained SVM weak classifier is output, positions of the pedestrians can be accurately determined, and detection can be completed. Via the SVM-AdaBoost algorithm-based pedestrian detection method, detection accuracy can be improved, and statistics can be rapidly run on the quantity and the density of the pedestrians.

Description

technical field [0001] The invention relates to the field of computer vision technology and security monitoring, in particular to a pedestrian detection method based on the SVM-AdaBoost algorithm. Background technique [0002] Pedestrian detection is to use a specific method to search for the presence or absence of pedestrians in the image to be detected within a certain period of time. If it exists, it returns the number of detected pedestrians and calculates the pedestrian density within the area. In recent years, with the rapid development of Internet technology, the application of pedestrian detection technology has become more and more extensive. Influenced by various factors, there is no real-time, accurate and unified pedestrian detection method at present. [0003] At present, pedestrian detection methods are mainly divided into four categories: feature-based methods, statistical methods, and template matching methods; the more common pedestrian detection methods ba...

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): G06K9/62
CPCG06F18/2148G06F18/2411
Inventor 张莉康保护
Owner JIUQUAN VOCATIONAL & TECHN COLLEGE