Title:
VEHICLE AND PEDESTRIAN DETECTION AND RECOGNITION METHOD COMBINING INTER-FRAME DIFFERENCE AND BAYES CLASSIFIER
Document Type and Number:
WIPO Patent Application WO/2020/248515
Kind Code:
A1
Abstract:
The present invention relates to the field of image recognition, and relates to a vehicle and pedestrian detection and recognition method combining an inter-frame difference and a Bayes classifier. According to the present invention, local vehicle and pedestrian information is utilized to overcome defects of conventional inter-frame difference missed detection and incapability of detecting a static object, and complete areas of vehicles and pedestrians moving on a road and incomplete static vehicle and pedestrian areas can be detected. Moreover, compared with a conventional SVM + HOG or neural network method, the operation speed of the method is higher under the condition of the same accuracy, and the costs are lower. According to the method of the present invention, only positive samples are needed, the number of the needed samples is smaller than that of SVM + HOG samples, and compared with a neural network learning process, the method is simpler and high in real-time performance.
Inventors:
WANG GUOJU (CN)
LIU HUILIN (CN)
LIU HUILIN (CN)
Application Number:
PCT/CN2019/119689
Publication Date:
December 17, 2020
Filing Date:
November 20, 2019
Export Citation:
Assignee:
SUZHOU UNION INTELLIGENT TECH CO LTD (CN)
International Classes:
G06K9/00
Foreign References:
| CN110263693A | 2019-09-20 | |||
| CN107122734A | 2017-09-01 | |||
| US20120230553A1 | 2012-09-13 |
Attorney, Agent or Firm:
CENTRAL SOUTH WELL INTELLECTUAL PROPERTY OFFICE (CN)
Download PDF:
Previous Patent: AGRICULTURAL FILM AND PREPARATION METHOD THEREFOR
Next Patent: AIR-COUPLED ULTRASONIC INTERFEROMETRY METHOD
Next Patent: AIR-COUPLED ULTRASONIC INTERFEROMETRY METHOD