基于改进加权局部对比度的红外小目标检测

在复杂背景的红外图像中弱小目标通常淹没在高亮边缘与强杂波处，提出一种基于改进加权局部对比度的红外小目标检测方法。利用小目标的局部特性建立一种加权函数将目标与其背景邻域的差异点乘凸显目标，进而与相接背景邻域作比值运算达到抑制复杂背景的效果；通过目标的各向同性和背景的各向异性，采用六方向梯度决策法创建背景抑制模型进一步抑制高亮边缘，实现降低虚警率，提高检测率的目的；最后，通过卷积计算将两者结合，采用自适应阈值分割检测真实目标。实验结果表明，该算法在复杂背景及强杂波干扰下有较强的鲁棒性。

Infrared small object detection based on improved weighted local contrast

In the infrared image of complex background,weak and small targets are usually submerged in the highlight edge and strong clutter.In this paper,a method for detecting small infrared targets based on improved weighted local contrast is proposed.Using the local characteristics of the small target,a weighting function is established to multiply the difference point between the target and its background neighborhood to highlight the target,which in turn is compared with the adjacent background neighborhood to suppress the effect of complex background.Through the isotropy of the target and the anisotropy of the background,the background suppression model is created by the six direction gradient decision method to further suppress the highlighted edges,so as to reduce the false alarm rate and improve the detection rate.Finally,the two are combined by convolution calculation,and the real target is detected by adaptive threshold segmentation.The experimental results show that the proposed algorithm has strong robustness under complex background and strong clutter interference.