Multi-agent collaborative evolution-based severe weather image restoration method and related device

By constructing a multi-agent co-evolutionary framework, the problems of insufficient generalization ability and unreliable learning signals in severe weather image restoration methods are solved. Adaptive learning and continuous optimization under unlabeled real data are achieved, improving the stability and effectiveness of image restoration.

CN122243824APending Publication Date: 2026-06-19SOUTH CHINA UNIV OF TECH
View PDF -1 Cites -1 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SOUTH CHINA UNIV OF TECH
Filing Date
2026-05-18
Publication Date
2026-06-19

AI Technical Summary

Technical Problem

Existing methods for restoring severe weather images rely on training with synthetic data, which has insufficient generalization ability. When the learning signal is unreliable in the absence of reference ground values, the fixed optimization objective lacks an effective learning mechanism with unlabeled real data, making it difficult to continuously optimize in dynamic environments.

Method used

A closed-loop collaborative mechanism is constructed, consisting of a recovery agent, a prompting agent, and an evaluation agent. Preference signals are generated through multi-dimensional quality assessment, and reinforcement learning strategies are used to optimize the parameters of the recovery agent. Furthermore, online and offline training are combined to form an adaptive learning mechanism.

Benefits of technology

It achieves stable and reliable image restoration without paired supervision, improves the model's adaptability and restoration effect in complex environments, maintains semantic information and structural consistency, and has continuous optimization capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122243824A_ABST
    Figure CN122243824A_ABST
Patent Text Reader

Abstract

This application provides a method and related equipment for image restoration in severe weather based on multi-agent co-evolution. The method includes: acquiring a degraded image; generating guidance information for restoration through a prompting agent; generating multiple candidate restoration results through a restoration agent; performing multi-dimensional quality evaluation on the candidate results through an evaluation agent to generate a preference signal reflecting relative superiority or inferiority; optimizing the restoration agent based on the preference signal using a reinforcement learning strategy, and storing the interaction data in an experience replay pool to offline optimize the prompting agent and / or the evaluation agent, forming a closed-loop co-evolutionary mechanism of "generation-evaluation-optimization". This application achieves adaptive learning without paired supervision, can stably improve the image restoration quality in real severe weather scenarios, and supports continuous optimization of video streams, and can be applied to fields such as autonomous driving and intelligent monitoring.
Need to check novelty before this filing date? Find Prior Art