Infrared simulation and detection method and system for low-altitude threat target

Through infrared simulation and detection methods of low-altitude threat targets, precise modeling of terrain and targets, and the use of multi-scale dynamic kernel fusion technology, the problems of detection blind spots in low-altitude environments and high-speed target detection in complex backgrounds are solved, and real-time and robust target detection is achieved.

CN120672984APending Publication Date: 2025-09-19NANJING UNIV OF SCI & TECH
0 Cites 0 Cited by

Patent Information

Application Number
CN202510810831.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-06-17
Publication Date
2025-09-19

AI Technical Summary

Technical Problem

Existing active protection systems have blind spots at low altitude and close range, poor intuitive visibility, difficulty in quickly and effectively detecting high-speed incoming targets in complex backgrounds, and high field testing costs.

Method used

An infrared simulation and detection method for low-altitude threat targets is designed. This method involves accurately modeling terrain, vegetation, and threat targets to generate infrared images. High-speed moving targets are detected using multi-scale dynamic kernel fusion technology. Relative total variation and Gaussian weighted kernels are used to process background clutter interference to achieve target detection.

Benefits of technology

It achieves accurate simulation and real-time detection of scenes and targets in low-altitude environments, improves the robustness and real-time performance of detection, reduces computing power requirements, and adapts to high-speed moving target detection in complex and changing scenarios.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses an infrared simulation and detection method and system for a low-altitude threat target. The method comprises the following steps: creating a terrain according to a topographic map of a real scene; modeling vegetation, buildings and threat targets on the terrain; utilizing python to simulate an accurate motion trail of the threat target; each model is rendered to generate a simulated infrared sequence, and the simulated infrared sequence is output to a processing module; carrying out preprocessing such as relative total variation on the simulated infrared sequence to suppress clutters; and calculating local gray statistical information by adopting a multi-scale sliding window, dynamically adjusting the size of a filtering kernel, fusing multi-scale information, and detecting and selecting a high-speed moving target with violent scale change by utilizing a comparison relationship between the target and a background kernel. According to the invention, an accurate diversified low-altitude infrared scene is constructed; for a dynamic threat target, simulating an accurate motion track of the target through python; and a multi-scale convolution scheme is adopted in a detection algorithm, so that the detection precision is improved, and meanwhile, the computing power requirement is reduced. High-speed real-time performance can be achieved, and low-altitude environment is adapted.
Need to check novelty before this filing date? Find Prior Art