Multi-region people counting method and device

Abstract

The invention provides a multi-region people counting method and device, and relates to the field of artificial intelligence, and the method comprises the steps: obtaining video stream data of monitoring equipment, wherein the video stream data comprises a video stream timestamp; extracting a detection picture from the video stream data according to a preset picture extraction parameter; carrying out people number identification on the detection picture to generate detection data, wherein the detection data comprises area information, a video stream timestamp and the number of people of each picture; and generating a multi-region people counting result according to the region information, the timestamp information and the people number of each picture. The invention provides multi-region people counting, and adopts multiple monitoring cameras to realize video stream acquisition, thereby achieving the function of multi-camera real-time region people counting.

Description

technical field [0001] The invention relates to data processing technology, in particular to a multi-area people counting method and device. Background technique [0002] With the development of smart security and smart cities, the deployment of surveillance equipment is getting larger and larger, and a large amount of video data has brought more opportunities, and various applications based on surveillance video streams have been continuously expanded. [0003] Traditional functions such as personnel retrieval and population statistics based on video streams require huge manpower and material resources, and are almost all presented offline, lacking timeliness and low efficiency. With the gradual development of artificial intelligence, more and more industries As a result, it has gradually become more intelligent, and more and more machines have replaced human labor, which has demonstrated huge advantages. Regional people counting is a function that requires high effectiven...

AI Technical Summary

Problems solved by technology

However, this method only considers the case of a single camera, and a required range may not be exactly covered by a surveillance camera, so the applicability is poor, and the synchronization of the video is not considered in the single camera problem, which is more common in real life problem, so it will be powerless to meet the needs of more complex scenarios

The minimum number of drones is calculated by counting the overlapping area between the area information and the drones, and according to the built-in parameters of the drones and manually controlling all the drones to the specified position to cover the required statistical range, and finally according to the detection The results and the marked overlapping areas and non-overlapping areas are used to calculate the number of people in each area. Due to the use of drones and too many human manipulation factors, it takes a lot of manpower and material resources to count the number of people.

Images