An artificial intelligence-based automatic article classification method and system

By adopting an automatic item classification method based on RGB grouping attention mechanism and artificial intelligence training, the problem of inaccurate item recognition and classification is solved, achieving fast and accurate item classification and improving work efficiency.

CN120047931BActive Publication Date: 2026-07-21YANCHENG INST OF TECH +1
View PDF 4 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
YANCHENG INST OF TECH
Filing Date
2025-02-05
Publication Date
2026-07-21

AI Technical Summary

Technical Problem

During the automatic sorting of items, factors such as the angle and compression of the items can lead to inaccurate identification and sorting, resulting in insufficient accuracy of image recognition technology and affecting subsequent operations.

Method used

By collecting image information of various types of items, using the RGB grouping attention mechanism for feature extraction and processing, and combining artificial intelligence to train an initial classification model, a target classification model is established to achieve fast and accurate classification of items.

Benefits of technology

It improves the accuracy and efficiency of automatic item classification, reduces the time and effort required for manual classification, and provides a rich and comprehensive foundation of item information.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120047931B_ABST
    Figure CN120047931B_ABST
Patent Text Reader

Abstract

The application provides an artificial intelligence-based automatic article classification method and system, which collects various types of image information of articles, extracts main features from the various types of image information, establishes an article information library based on the main features, realizes the advance acquisition of information features of the articles in various forms, provides a rich and comprehensive article information basis for realizing automatic article classification, extracts and processes the main features in the article information library by using an RGB grouping attention mechanism to obtain effective feature images, provides accurate and effective feature images for the establishment of the training of a classification model, trains an initial classification model based on the effective feature images and in combination with artificial intelligence to obtain a target classification model, inputs the collected images of articles to be classified into the target classification model, determines the category of the articles to be classified according to the output result, quickly and accurately identifies the classified articles, greatly reduces the time and effort of manual classification, and improves work efficiency.
Need to check novelty before this filing date? Find Prior Art