Fine-grained vehicle image retrieval method and device based on multi-scale constraints

Abstract

The invention discloses a fine-grained vehicle image retrieval method and device based on multi-scale constraints, and the method comprises the steps: carrying out multi-scale region information labeling of a first training data set, and obtaining a second training data set; processing the second training data set based on an improved bounding box constraint algorithm and a Helen constraint algorithm to obtain a third training data set; training parameter values of target parameters of a fine-grained neural network model by adopting the third training data set to obtain a trained fine-grainedneural network model; and inputting a to-be-identified vehicle image into the trained fine-grained neural network model for identification to obtain a target vehicle image of the same type as the to-be-identified vehicle image. 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 fine-grained vehicle image retrieval based on multi-scale constraints. Background technique [0002] Vehicle image retrieval is a technique for retrieving similar images through input images, which mainly involves image vehicle feature extraction and image vehicle feature similarity analysis. Fine-grained image recognition is to find local area features with subtle differences in the image, and can identify different subclasses in a large class. Applying fine-grained image recognition technology to vehicle image retrieval can extract image fine-grained features and analyze the similarity of image fine-grained features. [0003] At the same time, with the continuous advancement of smart cities, people pay more and more attention to urban road traffic safety. However, the existing vehicle image ret...

AI Technical Summary

Problems solved by technology

[0004] In view of this, the present invention provides a fine-grained vehicle image retrieval method, device, computer equipment and storage medium based on multi-scale constraints, to solve the inaccurate positioning of vehicle images that need to search for information and the inaccurate information extraction in the prior art. issues such as clarity

Images