Vehicle fine-grained identification method and device

Abstract

The invention discloses a vehicle fine-grained recognition method and device based on combination of a multiple attention mechanism and regional feature constraints, and the method comprises the steps: constructing a multiple attention convolutional neural network model, and carrying out pre-training of the multiple attention convolutional neural network model; performing multi-scale region information labeling on the training data set to obtain a first training data set; processing the first training data set based on an improved bounding box constraint algorithm and a Helen constraint algorithm to obtain a second training data set; and training parameter values of target parameters of a multi-attention convolutional neural network model by adopting the second training data set to obtainthe trained multi-attention convolutional neural network model. Compared with the prior art, the workload of pre-labeling the image category or selecting the box to label the object position is reduced, the cost is saved, and the efficiency is improved.

Description

technical field [0001] The present invention relates to the technical field of image retrieval, in particular to a method, device, terminal device and computer-readable medium for vehicle fine-grained recognition based on the combination of multiple attention mechanisms and regional feature constraints. Background technique [0002] Vehicle fine-grained recognition is an important research direction in the field of computer vision. Vehicle identification of the same model is difficult for traditional methods, because the differences between similar vehicles are often very small. The difference may only lie in the annual inspection mark on it, or some small decorations in the car. With the rise of deep learning in recent years, many researchers have also tried to apply deep learning to the field of target detection and recognition based on this. Among them, fine-grained image analysis is a popular research topic in the field of computer vision for such problems. Its goal is...

AI Technical Summary

Problems solved by technology

[0004] In view of this, the present invention provides a vehicle fine-grained recognition method, device, computer equipment and storage medium based on the combination of multiple attention mechanisms and regional feature constraints to solve the problem of vehicle image positioning that requires search information in the prior art. Inaccurate, unclear information extraction and other issues

Images