Article recognition method and device, model training method and device

Abstract

The embodiment of the invention provides an article recognition method and device, a model training method and device, electronic equipment and a storage medium, and relates to the technical field of Internet and deep learning, and the method comprises the steps of obtaining target picture information corresponding to each to-be-recognized article, and inputting each target picture information to a recognition model; and generating an identification result corresponding to each to-be-identified article. The recognition model comprises a convolution layer and a clustering layer, wherein the clustering layer is used for performing clustering processing on each target picture feature output by the convolution layer, and each target picture feature is generated by performing convolution processing by the convolution layer according to each target picture information; the identification result represents whether each to-be-identified article meets a preset article requirement or not, and clustering processing is carried out on each target picture feature output by a convolution layer. Through clustering processing, the defect of low recognition efficiency caused by adopting a one-by-one recognition mode is avoided, so that the recognition efficiency is improved.

Description

Technical field [0001] The present application relates to the field of Internet and depth learning techniques, and more particularly to an article identification method, a model training method, a device, an electronic device, and a storage medium. Background technique [0002] With the rapid development of e-commerce, online shopping subverts traditional shopping. Online shopping Because it is convenient to become a favorite shopping method, in the face of a virtual environment such as a network, there is no information credibility problem, such as authenticity, etc., how to identify items The problem that needs to be solved urgently. [0003] In the prior art, it is generally used to compare the pictures of the item one by one, and the picture of the actual item is usually used, and the identification result is determined based on the similarity, if the similarity is greater than the threshold, the item corresponding to the identification result picture is to satisfy the pre-se...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of the present application provides an item recognition method, a model training method, a device, an electronic device, and a storage medium to solve the problem of low recognition efficiency

Images