摘要
图像阴影是一种普遍存在的图像退化降质现象,它的存在严重影响了高层的图像理解。采用均值平移理论模型,先对图像进行分类,然后利用图像阴影区域具有亮度暗,方差小的特点,从分类结果中筛选出阴影区域,通过实验验证表明,提出的算法对彩色影像上阴影的检测和提取是有效的。
The shadow phenomenon on the imagery is treated as a special kind of image degradation, its existing will ill influence the high-level image understanding. This paper brings forth a new shadow detection algorithm based on mean shift model. First, the paper uses mean shift model to classify images. Then, based on shadow area characteristic of low brightness and small covariance, determines which image classified area is shadow area. After the experiments with several real images, it shows that the mean shift model based shadow detection algorithm is valid both for the color images and the black and white images.
出处
《计算机与数字工程》
2007年第11期98-101,共4页
Computer & Digital Engineering
