Vehicle model identification method and device

Abstract

The embodiments of the invention provide a vehicle model identification method and device. The method includes the following steps: acquiring a to-be-processed image and a preset training model, wherein the preset training model includes a classifying sub-network model and a positioning sub-network model, and the to-be-processed image contains a target vehicle; training the to-be-processed image through the classifying sub-network model to get a global feature set; training the to-be-processed image through the positioning sub-network model to get a local feature set; fusing the features in the global feature set and the local feature unit to get a target feature set; and determining a target model corresponding to the target vehicle according to the target feature set. According to the embodiments of the invention, the local detailed features and global features in the to-be-processed image can be obtained, the global features and the local features are fused to ensure that the important information is brought into full play, and the efficiency of vehicle model identification can be improved.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method and device for identifying a vehicle type. Background technique [0002] With the continuous increase of the number of urban vehicles, the resulting traffic and environmental problems are also increasing. For the above problems, intelligent transportation system has become the key research object of urban development. In the intelligent transportation system, the fine-grained identification of vehicle types has been proved to be a key technology, which has unique advantages in dealing with the following events: improving toll efficiency, adjudicating traffic responsibilities, and tracking hit-and-run traffic issues , therefore, has received the attention of various research units. [0003] At present, vehicle type recognition is a research hotspot in the field of computer vision. It has been widely used in traffic safety, bayonet management, traffic statistics...

AI Technical Summary

Problems solved by technology

However, the existing car model recognition methods mainly use manually designed features, which are low-level features and have great limitations in dealing with fine-grained classification problems.

Images