本文的内容是基于红外图像的特征提取，在基于个数判断噪声的中值滤波器进行平滑处理的基础上,提出了一种改进的过渡区

提取方法,并将该方法应用于对比度低、宽谷直方图的复杂背景红外图像的分割。Matlab仿真

实验表明,该方法减少了图像分割的运算量,并能将具有复杂背景的红外图像目标分割得较

好。-This content is based on the infrared image feature extraction, the number of judgments based on the median noise smoothing filter based on an improved transition region extraction method and the method is applied to low contrast, wide Valley histogram segmentation of infrared image with complex background. Matlab simulation results show that the method reduces the computation of image segmentation, and can have a complex background object segmentation of infrared image better.