基于局部梯度和局部熵的红外小目标融合检测

提出了一种基于局部梯度强度和局部熵的红外小目标融合检测的方法。小目标检测是整个自动目标识别系统的一个非常重要的子系统，可以说，检测子系统的性能在很大程度上决定了自动目标识别系统的性能。由于弱小目标的存在会引起其所在区域的内部细节和局部纹理与背景相比发生较大的变化，而局部梯度强度和局部熵非常好地刻画这种变化。文中通过提取序列图像中每帧的局部梯度强度和局部熵的特征图像，进行自适应的融合。为了充分利用目标的运动信息，连续多帧图像的融合特征图像进行能量累积，达到进一步提高检测概率，降低虚警率。实验结果证明了方法的有效性。

Fusion Detection of Small Infrared Target Based on Local Entropy and Local Gradient Strength

A novel method based on gradient strength and local entropy for fusion detection of weak small targets with low contrast is presented in this paper.Small target detection plays an important role in automatic target recognition(ATR) system.The performance of target detection subsystem mostly determines the capability of overall ATR system.The local details and local texture will be changed because of the existence of small target of interest compared with the complicated background,which can be depicted by local gradient strength and local entropy.The two feature images of each frame in image sequences,i.e.local gradient strength and local entropy,are extracted to fuse adaptively.In order to sufficiently utilize the motion information,the feature images of multi-frame are cumulated to improve the detection probability and reduce the false alarm rate.The experimental results show that the proposed algorithm is effective.