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

A technology of reinforcement learning and vehicle recognition, applied in neural learning methods, character and pattern recognition, computer components, etc., can solve problems that affect the appearance of tracking objects, low resolution and small scale of UAV viewing angle targets, and achieve Fast, efficient and accurate automatic analysis and monitoring applications, accurate and efficient tracking of results, and the effect of liberating labor

Active Publication Date: 2020-03-10
QINGDAO RES INST OF BEIHANG UNIV +1
View PDF7 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

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-alt

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Unmanned aerial vehicle visual angle vehicle identification tracking method based on reinforcement learning
  • Unmanned aerial vehicle visual angle vehicle identification tracking method based on reinforcement learning
  • Unmanned aerial vehicle visual angle vehicle identification tracking method based on reinforcement learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] Focusing on the typical demonstration application requirements of automatic tracking of vehicles from the perspective of drones in the fields of transportation and city construction, the key technology and application system research and development based on reinforcement learning from the perspective of drones proposed by this invention establishes a high-speed track that can support unsupervised The automatic identification and tracking of vehicles can effectively break through the bottleneck problems of high technical threshold, complicated use, and long product generation time in the fields of intelligent transportation and automatic driving, etc.

[0031] Aiming at the lack of datasets for deep learning networks, an automatic generation method of UAV datasets based on reinforcement learning is proposed. Cooperate with the vehicle perspective of the unmanned driving dataset and the unlabeled video collected by the drone to convert the perspective to adapt to the netw...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

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...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06K9/00G06N3/04G06N3/08
CPCG06N3/084G06V20/42G06V20/54G06N3/045Y02T10/40
Inventor 李帅宋文凤于洋石翔
Owner QINGDAO RES INST OF BEIHANG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap