Commodity image retrieval method, device, electronic device, and storage medium

By obtaining the historical search records of product logos and adjusting the inlet method of image feature information using the inlet strategy, the huge problem of image search database affecting the search efficiency is solved, and efficient and accurate product image retrieval is achieved.

CN119226554BActive Publication Date: 2025-04-01ZHUHAI AOSHI INNOVATION TECH CO LTD
2 Cites 0 Cited by

Patent Information

Application Number
CN202411075112.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-07
Publication Date
2025-04-01
Estimated Expiration
2044-08-07

AI Technical Summary

Technical Problem

With the increase of product categories, the image search database becomes larger and larger, affecting the search result accuracy of the search system. At the same time, the existing technology improves the search accuracy by introducing a variety of algorithms and models, but leads to low retrieval efficiency.

Method used

By obtaining the historical search records of each product logo, the search attribute information is extracted using the preset classification model, the image feature information is adjusted based on the entry conditions and entry strategies, the storage amount of the image search database is reduced, and the feature information is processed through the clustering algorithm to improve the search efficiency.

Benefits of technology

It effectively improves the efficiency of product image retrieval, while ensuring the accuracy of search results, and reducing the storage performance requirements for image retrieval database.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present application discloses a commodity image retrieval method, device, electronic device, and storage medium, the method comprising obtaining a first historical retrieval record of each first commodity identification; inputting all first image feature information corresponding to the first commodity identification into a preset classification model to obtain retrieval attribute information corresponding to each first commodity identification; selecting a target entry strategy from an entry strategy set based on target attribute information determined from all retrieval attribute information based on entry conditions, and saving the first image feature information of the corresponding first commodity identification to an image retrieval database based on the target entry strategy; and performing commodity image retrieval based on the image retrieval database. The present application uses the retrieval attribute information and entry strategy corresponding to each commodity identification to adjust the amount of image feature information stored in the image retrieval database, and performs commodity image retrieval based on the image retrieval database adjusted by the technical solution of the present application, which can effectively improve the efficiency of commodity image retrieval.
Need to check novelty before this filing date? Find Prior Art