Image-based store identification method, device, and electronic equipment

Abstract

The application discloses an image-based store identification method, which belongs to the field of computer technology and is used to solve the problem of low accuracy of image-based store identification. The image-based store recognition method disclosed in the embodiment of the present application includes: acquiring the global image features of the target plaque image, and acquiring the sub-regional salient features of the target plaque image; determining The image features of the target plaque image; determining the store matching the target plaque image according to the image features of the target plaque image and the store image features of preset stores. Since the image features adopted by the image-based store identification method disclosed in the embodiment of the present application take into account both the global features of the image and the local fine-grained regions, the image features are comprehensive and distinguishable, which can improve the image-based store identification. Accuracy.

Description

technical field [0001] The present application relates to the field of computer technology, in particular to an image-based shop identification method, device, and electronic equipment. Background technique [0002] In the prior art, when searching for a store or identifying a store in an image through the image of a store plaque, a feature index can be established for the entire signboard area, that is, the character information of the image, texture features, and plaque content distribution information are considered, which will have better results. robustness. For example, the Chinese patent application with publication number CN104598885B uses scale-invariant feature transform (SIFT, Scale-invariant feature transform) to detect or describe local features in images, and combines HS feature components for image recognition. It is more robust to images with small viewing angle changes and small changes in lighting scenes. However, the image features extracted in the above...

AI Technical Summary

Problems solved by technology

However, the image features extracted in the above shop recognition methods in the prior art are less robust to the recognition of larger images with changes in the acquisition angle and lighting scene, and the feature expression of the shop plaque image is not clear and complete enough. As a result, the accuracy of store identification is not high

[0003] To sum up, the image-based store recognition methods in the prior art have at least the problem of low recognition accuracy

Images