| 摘要 |
一种影像杂讯估测方法，包括下列步骤：决定一影像的多个取样区块；为每一上述取样区块，计算该取样区块的至少一色彩分量的平均值与至少一色彩分量的标准差；根据上述平均值将上述多个取样区块划分为多个区间；以及为每一上述区间，根据至少一门槛值计算该区间的所有取样区块的标准差的加权平均值。上述门槛值是根据该区间的所有取样区块的标准差其中的最小值而决定。上述加权平均值可用于影像的杂讯消除、边缘侦测、或移动侦测。; for each sample block, calculate a mean of at least one color component of the sample block and a standard deviation of at least one color component of the sample block; dividing the sample blocks into a plurality of segments according to the means of the sample blocks; and, for each segment, calculate a weighted average of the standard deviations of all of the sample blocks of the segment according to at least one threshold values. The at least one threshold value is determined according to the minimum among the standard deviations of all of the sample blocks of the segment. The weighted average may be used in noise reduction, edge detection or motion detection of the image. |
