Abstract: According the characteristic of tensor filed can externalize the local exposure information in image，this paper introduces an image de-noising method which uses the local exposure information of tensor field to control filter.The method can maximizing holding the structure information in image like line，border and so on.The experimental results show that the peak signal-to-noise ratio of the de-noised image and the visual effect are improved via the proposed algorithm.

摘要： 利用张量场能体现图像各个局部的朝向信息的特点，提出利用张量场局部朝向信息控制滤波器对图像进行去噪处理的方法。以达到最大限度地保持图像中原有的边缘、线条等机构信息的目的。实验结果表明，该算法在去除噪声的同时很好地保持了图像中的细节特征，去噪后的图像在峰值信噪比和主观视觉效果上都有显著提高。