In order to remove the salt and pepper noise in an image,an window adaptive filtering algorithm is proposed. A 3×3 window is used in the algorithm to detect noise. If center point is noise point,the statistics of non⁃noise points in the win⁃dow is performed. When the number is greater than 2,the median⁃mean filtering algorithm is used. When the number is less than or equal to 2,the filtering window size is extended to 5×5,and then the median⁃mean filtering algorithm is used. If center point is signal point,the original value is maintained and output directly. Simulation experiment results demonstrate the new al⁃gorithm has strong removing noise ability for images polluted by different degree salt and pepper noise,and can keep the details of the image.%为了有效地去除图像中的椒盐噪声，提出一种窗口自适应的滤波算法。算法先采用3×3窗口进行噪声检测，如果中心点为噪声点，则统计窗口内为非噪声点的数量。当非噪声点的数量大于2时，采用中值均值滤波算法；当非噪声点的数量小于等于2时，将窗口尺寸扩大至5×5，采用中值均值滤波算法。如果中心点为信号点，则保持原值不变直接输出。仿真实验结果证明，这种算法对不同程度椒盐噪声污染的图像具有较强的去噪能力，同时较好地保持了图像的细节。
展开▼
