Airport runway multi-sensor collaborative target intrusion identification method and device

By constructing a perspective compression rate heatmap and a depth slicing mechanism, combined with radar and photoelectric sensors, the pixel-level coordinate collapse problem caused by perspective projection was solved, enabling accurate separation and identification of distant targets and improving the identification accuracy and positioning precision of runway safety monitoring.

CN122200373BActive Publication Date: 2026-07-17HANGZHOU LIANFEI TECH CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2026-05-12
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing technologies fail to effectively address the pixel-level coordinate collapse problem caused by perspective projection under small downward angles when dealing with deep runway scenarios, resulting in missing target counts and incorrect type identification, especially with insufficient robustness in distant areas.

Method used

By constructing a perspective compression rate heatmap and a depth slicing mechanism, and combining the range resolution advantage of radar with the semantic recognition capability of photoelectric sensors, the visual proportion features of distant targets are restored using inverse perspective equivalent scale unfolding technology, thereby achieving accurate separation, independent counting, and attribute fusion of multiple targets.

Benefits of technology

In extreme perspective environments, it enables independent counting, accurate positioning, and type determination of multiple intrusion targets, improving the accuracy of target identification and positioning precision across the entire depth range.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122200373B_ABST
    Figure CN122200373B_ABST
Patent Text Reader

Abstract

This invention relates to the field of image processing and target recognition technology, and discloses a multi-sensor collaborative target intrusion recognition method and apparatus for airport runways. The method includes: generating a perspective compression rate heatmap based on the calibration parameters of a dual-spectral camera and the geographic coordinate contour of the runway; subsequently excluding known aircraft-occupied areas within the high-compression region of the heatmap, and generating multiple independent depth slice sub-images and associated records by depth segmentation; then performing inverse perspective equivalent scale expansion on each sub-image and performing independent target detection and classification recognition; finally, mapping the results back to the GIS geographic coordinate system and fusing radar measurement attributes to generate target recognition signs. This invention solves the problem of multiple target coordinate collapse and recognition adhesion caused by perspective compression at the far end of the runway by spatial decoupling of the depth dimension and equivalent restoration of image scale, thereby improving the accuracy of target separation and recognition in complex perspective environments.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Airport runway intrusion detection methods, devices, storage media and electronic equipment

    CN113344900B

  • Airport runway intrusion identification method based on ESNB algorithm

    CN121121672A

  • Marine ship intelligent detection and three-dimensional positioning method based on multi-modal data cooperation

    CN121259288A

  • Data processing method and device based on multi-sensor fusion, and multi-sensor fusion method

    US20210012165A1