A People Counting System Based on Single Frame Image Processing

Abstract

The invention relates to a people counting system based on single-frame image processing. The system includes an offline training module and an online detection module; the offline training module first obtains an image training set of an application scene, and then manually marks positive samples and negative samples from the training set. It is used to train the cascade detector, and use the trained detector to perform detection tasks on the training image set; then reconstruct the head positive and negative sample set according to the detection results, and use it to train the classifier; the online detection module first uses the cascade detector to treat The rough head detection is performed on statistical images to obtain the suspected head area, and then the classifier is used to further identify and confirm the suspected head area. Finally, various prior information in the application scene is used to post-process the detection results to obtain the final detection result. The system can save a lot of human resources, avoid erroneous statistics caused by human factors, and well overcome the counting disadvantages of manual counting in some scenarios.

Description

technical field [0001] The invention belongs to the technical field of image processing and video monitoring, and relates to a people counting system based on single-frame image processing. Background technique [0002] At present, with the continuous development and influence of intelligent technology, image processing and video surveillance systems have been widely used in various aspects, and traditional manual control methods are gradually being replaced by intelligent equipment. Among them, the people counting system, as a counting system for evaluating open or closed environments, plays a very important role in real-life applications. For example, by automatically counting the number of classrooms, the attendance rate of each subject can be inspected, so as to reasonably evaluate the teaching quality, and at the same time, it can help students to quickly choose the appropriate self-study classroom. Through the statistics of the number of people entering and leaving th...

AI Technical Summary

Problems solved by technology

[0003] However, the traditional manual counting method may consume a lot of human resources or cause wrong statistics. Therefore, there is an urgent need for a real-time automatic counting system that can overcome the disadvantages of manual counting.

Images