An artificial intelligence image recognition system based on reinforcement learning

By constructing a multi-objective Q-learning closed-loop framework for the entire system, the color distribution mismatch problem in image recognition of multi-source heterogeneous devices is solved, achieving efficient cross-device recognition and adaptive optimization, and improving recognition accuracy and system performance.

CN122367769APending Publication Date: 2026-07-10SHANDONG UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610495300.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-04-15
Publication Date
2026-07-10

AI Technical Summary

Technical Problem

Existing AI image recognition systems based on reinforcement learning cannot effectively handle image input from multiple heterogeneous devices. The mismatch in color distribution across devices leads to a decrease in recognition accuracy, and existing technologies have not been able to effectively solve this problem.

Method used

By employing an optimal transmission color alignment preprocessing module, a Q-learning convolution kernel dynamic configuration feature extraction module, an enhancement learning recognition decision module, a performance monitoring module, an adaptive adjustment module, and a self-optimization module, a multi-objective Q-learning closed-loop framework for the entire system is constructed to achieve adaptive optimization of cross-device image recognition.

Benefits of technology

It improves cross-device recognition accuracy, achieves a dynamic balance between system versatility and recognition efficiency, reduces system maintenance costs, and achieves continuous performance improvement through a self-optimizing framework.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122367769A_ABST
    Figure CN122367769A_ABST
Patent Text Reader

Abstract

This invention discloses an image recognition system for artificial intelligence based on reinforcement learning. The invention relates to the field of artificial intelligence image recognition technology and includes a multi-device image input module, an optimal transmission color alignment preprocessing module, a Q-learning convolutional kernel dynamic configuration feature extraction module, a reinforcement learning recognition decision module, a performance monitoring module, an adaptive adjustment module, and a self-optimization module. The multi-device image input module receives image data from multiple heterogeneous devices and detects and identifies the device type and color space type corresponding to the image data. The advantages of this invention are: it adopts optimal transmission color alignment preprocessing technology, fundamentally solving the problem of cross-device color distribution mismatch; it is compatible with image input from multiple heterogeneous devices such as surveillance cameras, industrial inspection cameras, consumer mobile phones, and remote sensing satellites; the cross-device recognition accuracy is improved compared to existing normalization schemes; and the system's versatility is significantly improved.
Need to check novelty before this filing date? Find Prior Art