Unmanned aerial vehicle visual angle vehicle identification tracking method based on reinforcement learning

Abstract

The invention discloses an unmanned aerial vehicle visual angle vehicle identification tracking method based on reinforcement learning. Based on unmanned aerial vehicle visual angle scene understanding, monitoring and tracking, efficient and self-adaptive panoramic video management is established, and through a transfer learning target tracking method of reinforcement learning, the unmanned aerialvehicle can perform self-adaptive fast moving vehicle tracking under the non-supervision condition. The cross-view and cross-azimuth space-ground cooperative tracking system is realized through ground camera data, cooperative processing and re-identification information and algorithms, so that traffic analysis does not pay attention to repeated video annotation work any more, the labor force of manual monitoring is liberated, and automatic analysis and monitoring application can be quickly, efficiently and accurately carried out according to an initialization target vehicle provided by software in advance.

Description

technical field [0001] The invention relates to the field of computer vision image and video understanding, and relates to a vehicle recognition and tracking method based on reinforcement learning from an unmanned aerial vehicle perspective. Background technique [0002] UAV-based automatic tracking technology enables users to break through space, time and other objective constraints, and conduct panoramic monitoring and tracking activities from a bird's-eye view, which can greatly improve the performance of security monitoring and high-speed object tracking, and provide high-speed vehicle monitoring. Fast and accurate tracking algorithm. [0003] The existing vehicle recognition and tracking algorithm includes the following steps: manually label the tracking data set from the perspective of the drone, obtain the training label, and perform data enhancement; pre-train the neural network on the general data set from the ground perspective, so that the subsequent Further trai...

AI Technical Summary

Problems solved by technology

[0006] In view of the difficulty in collecting data sets of fast-moving motor vehicles in the prior art, the scale is too small to be fixed, the target resolution of the drone's perspective is low, and high-altitude obstructions affect the appearance of tracking objects, etc., the present invention proposes a A vehicle recognition and tracking method based on reinforcement learning from the UAV perspective, including:

Images