光子学报, 2009, 38 (9): 2444, 网络出版: 2010-05-10
基于多特征融合的复杂背景下弱小多目标检测和跟踪算法
Detecting and Tracking Arithmetic for Infrared Faint Targets under Complex Background Based on Multi-Characteristics Fusion
多特征融合 红外弱小多目标 多向梯度 信噪比 检测概率 Multi-characteristics fusion Infrared faint targets Multi-direction grads SNR Detection probability
摘要
针对红外弱小多目标的检测和跟踪难题,提出一种基于多特征融合的复杂背景下弱小多目标检测和跟踪算法.融合红外弱小运动目标的灰度特征、梯度特征、运动特征等多个典型特性,进行复杂背景下弱小多目标的检测和跟踪.实验证明:该算法应用于复杂背景下低信噪比的红外弱小多目标图像序列能得到较理想的结果,算法检测概率高、检测速度快、具有较强鲁棒性.
Abstract
Aiming at detection and tracking puzzle for infrared faint targets,a detecting and tracking arithmetic for infrared faint targets under complex background is put forward based on multi-characteristics fusion. Several typical characteristics of infrared faint targets such as grey, grads and movement characteristic, accomplish detection and tracking for faint targets are fused.Experiment results show that the arithmetic can obtain good result toward low SNR infrared image sequence under complex background. The arithmetic have high detection probability, high detecting speed and stronger stability.
陈皓, 马彩文, 陈岳承, 岳鹏. 基于多特征融合的复杂背景下弱小多目标检测和跟踪算法[J]. 光子学报, 2009, 38(9): 2444. CHEN Hao, MA Cai-wen, CHEN Yue-cheng, YUE Peng. Detecting and Tracking Arithmetic for Infrared Faint Targets under Complex Background Based on Multi-Characteristics Fusion[J]. ACTA PHOTONICA SINICA, 2009, 38(9): 2444.