基于自适应邻域概念的实时视频预处理算法

针对各种实时视频应用领域的实际要求，提出了一种新颖的实时视频预处理算法。该算法将自适应邻域的概念应用到视频信号预处理中，并融合线性均值滤波和非线性中值滤波两种方法的优点，能够同时有效地抑制加性高斯白噪声和稀疏的冲击噪声，并能完好地保持图像的边缘，因此，成功地弥补了传统算法的不足．此外，该算法较低的运算复杂度使得它可广泛应用于各种实时环境．仿真和实际使用结果表明，通过预处理后的图像的确获得了良好的视觉效果．

A Real-time Video Pre-filtering Algorithm Using Adaptive-neighborhood Statistics

Video pre-filtering is an important technique in all kinds of real-time video application fields and related areas. However, many proposed algorithms are too complex to be applied under real-time circumstances, and can not effectively suppress both additive white Gauss and impulsive noise without blurring the edges. We applied the concept of adaptive-neighborhood statistics to video pre-processing and proposed a novel algorithm. It not only retains the edge information of color images quite well but also has good restraint characteristics for both additive Gauss and impulsive noise. The emulation result shows that the algorithm provides better visual effect with very low computing complexity.