摘要
将红外图像中的小目标检测视为基于多图像特征的像素分类问题,提出了一种基于D-S证据理论的红外小目标融合检测方法。该算法首先提取双色红外成像系统中各传感器图像的图像特征,采用D-S证据理论中的基本概率分配函数对传感器图像中的像素进行基于多图像特征的分类,得到各传感器的目标检测基本可信度图;然后应用正交和规则复合来自各传感器的目标检测判决证据,获得整个系统的目标检测基本可信度图;最后根据决策规则输出最终目标检测结果。实验结果显示,该算法能在较大程度上降低目标检测过程中的不确定性,提高了系统的检测性能。
Taking the of small target detection in IR images as the classification problem of pixels based on multiple image features, a method of IR small target fusion detection based on the D-S evidence theory is proposed. Firstly the features of the sensor images inthe dual-band IR system are extracted and all pixcls in the sensor images are classified based on the multiple image features using the basic probability assignment function of the D-S evidence theory to get the target detection basic confidence images of the sensors. Then the evidences for target detection decision from different sensors are combined using the orthogonal sum rule to get the target detection basic confidence image for the whole system. Finally, the final target detection result is obtained according to the decision rule. The experimental result shows that the method can reduce the uncertainty during the target detection to a large degree and improve the target detection performance of the whole system.
出处
《系统工程与电子技术》
EI
CSCD
北大核心
2006年第1期26-30,共5页
Systems Engineering and Electronics
关键词
双色红外
目标检测
信息融合
D-S证据理论
dual-band IR
target detection
information fusion
D-S evidence theory
