Real-time infrared small target detection method based on linear global scanning network

By using the linear global scanning network LGSNet, the problems of computational complexity versus real-time performance and low efficiency in spatial dependency modeling in infrared small target detection are solved, achieving efficient infrared small target detection and improving detection accuracy and robustness.

CN121788809AActive Publication Date: 2026-04-03NANJING UNIV OF POSTS & TELECOMM
View PDF 6 Cites 1 Cited by

Patent Information

Application Number
CN202610275174.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-09
Publication Date
2026-04-03
Estimated Expiration
2046-03-09

AI Technical Summary

Technical Problem

Existing infrared small target detection methods suffer from a trade-off between computational complexity and real-time performance, and their spatial dependency modeling is inefficient, making it difficult to achieve efficient infrared small target detection on embedded hardware platforms.

Method used

A linear global scan network (LGSNet) based on recurrent neural networks is adopted. By combining the linear global scan module and the linear context aggregator with the spatial scan GRU unit and the channel attention branch, a lightweight network is constructed to capture global dependencies and enhance feature extraction capabilities.

Benefits of technology

While reducing computational overhead, it improves the accuracy and robustness of infrared small target detection, achieves real-time response speed, and is suitable for embedded hardware platforms.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121788809A_ABST
    Figure CN121788809A_ABST
Patent Text Reader

Abstract

The invention discloses a real-time infrared small target detection method based on a linear global scanning network. In order to solve the problems that an existing model is high in calculation complexity and poor in real-time performance, a U-Net-like lightweight architecture giving consideration to both detection precision and reasoning speed is constructed. In the encoding stage, shallow local textures are extracted through Stem and ResBlock, a linear global scanning (LGS) module is introduced into a deep layer, and the core of the LGS module captures anisotropic long-distance semantic dependency with linear complexity by utilizing space scanning GRU. The coding end multiplexes the GRU by using a linear context aggregator (LCA) and combines a channel reweighting enhancement feature. In the decoding stage, jump connection semantics are aligned through PlainBlock, and upsampling features are fused through a StandFusion module and are refined through cascade convolution. And finally, an Inception module is combined with depth supervision to generate a high-precision prediction result. The method is low in calculation overhead, high in detection precision and suitable for high-frame-rate real-time infrared monitoring scenes.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Design method of interpretable multi-scale infrared weak and small target detection network

    CN114998566A

  • Lightweight sand dune form type identification model based on partial convolution and Transform

    CN119295890A

  • Water body classification method and system based on multi-scale feature fusion and attention mechanism

    CN120126009A

  • Intensive small target detection method based on fuzzy perception

    CN120318493A

  • Methods and systems for automated image segmentation of anatomical structure

    EP4300365A2

Cited By

  • Steel bar detection model, method and system

    CN122073010A