Pedestrian counting method based on multiple features

Abstract

The invention discloses a pedestrian counting method based on multiple features. A background lattice model and a capture prospect lattice are set, so the position of the light spot of the background lattice model and the position of the light spot of the capture prospect lattice are compared, a prospect light spot offset model is set, then a light spot offset local maximum value point area in the prospect light spot offset model is extracted to serve as a head area of a pedestrian and is clustered to a head block mass, a motion trail is traced finally, and the number of people is determined and recorded based on motion features. The pedestrian counting method is based on light stream computing, renewal of a pedestrian individual track is achieved by means of intra-frame block mass bilateral matching, and statistics of the number of people is accurate and reliable. The pedestrian counting method is further based on sparse depth data, and improves the execution speed of an algorithm compared with a method based on a dense depth. At last, relative depth acquisition equipment adopted in the pedestrian counting method is low in cost, so the pedestrian counting method has a greater advantage in cost.

Description

technical field [0001] The invention relates to the technical field of video passenger flow statistics, in particular to a multi-feature-based pedestrian counting method. Background technique [0002] The passenger flow information provided by video passenger flow statistics technology can provide data support for personnel scheduling, resource allocation, business strategy, security monitoring and other aspects of the industry. Early passenger flow statistics can only use traditional contact methods such as manual statistics, bill statistics, mechanical means, etc. that will cause interference to pedestrians. Later, electronic triggers or photoelectric sensors were used. These devices are relatively simple to implement, but the effect is not good in situations such as high crowd density and serious pedestrian occlusion. The video-based pedestrian counting method can count people in real time, accurately and without interference, and has become a research hotspot of many sc...

AI Technical Summary

Problems solved by technology

Although in recent years, video pedestrian counting methods have made great progress in terms of real-time and accuracy, there are still many challenges in this technology: (1) Since ordinary cameras mainly capture scene color images, foreground detection is vulnerable The impact of environmental conditions such as illumination changes, severe shadows, and complex and changeable scenes

(2) In the scene of dense pedestrians or abnormal pedestrian behavior, it is difficult to effectively segment or estimate the number of pedestrian clusters, which seriously affects the accuracy of the algorithm

(3) Some pedestrian detection and tracking algorithms put forward requirements for the resolution of video images, but in practical applications, due to the consideration of equipment cost, the resolution of video images is generally low

[0014] At present, s

Images