摘要
传统单谱段探测技术因光谱信息单一、环境适应性差,难以满足复杂动态环境下的高精度识别需求。为此,提出一种基于多谱段融合的光电探测识别技术。利用棱栅-电荷耦合元件系统生成多谱段图像,结合亮度-色调-饱和度(intensity-hue-saturation,IHS)颜色模型与改进小波变换实现图像融合,提高目标的光谱与空间信息完整性。结合最大值投影统计量与多谱段一致性连通域提取方法,在复杂背景下实现弱小目标的可靠识别。试验结果表明,该方法最大耗时仅为10 ms,星掩复杂场景下目标识别率也能达到90%,使用该技术在复杂场景下表现出较强的抗干扰能力。
Traditional single-spectral detection technology struggles to meet high-precision recognition demands in complex dynamic environments due to its limited spectral information and poor environmental adaptability.In this regard,a photoelectric detection and recognition technology based on multi-spectral fusion is proposed.The multi-spectral image is generated by using the grismcharge coupled device system,and the image fusion is realized by combining the intensity-hue-saturation(IHS)color model and the improved wavelet transform,enhancing the integrity of both spectral and spatial information of the target.Combined with the maximum projection statistic and the multi-spectral consistency connected domain extraction method,the reliable recognition of weak and small targets in complex background is realized.The test results show that the maximum time consumption of this method is only 10 ms,and the target recognition rate can reach 90%in complex star-masking scenario.This technology exhibits strong antiinterference capability in complex scenarios.
作者
王振东
WANG Zhendong(Luoyang Electro-Optical Equipment Research Institute,Luoyang 471000,Henan,China)
出处
《光源与照明》
2026年第1期86-88,共3页
Lamps & Lighting
关键词
多谱段融合
光电探测
目标识别
IHS颜色模型
最大值投影
multi-spectral fusion
photoelectric detection
target recognition
IHS color model
maximum projection
