基于小波变换的分水岭图像分割方法

图像分割技术在数字图像处理中占有重要地位.提出了一种基于小波变换的图像分割方法,有效地将小波分析、小波包分解与数学形态学中的分水岭方法相结合.首先,通过小波包对图像有效降噪,在一定程度上减少了分水岭方法的过分割现象.然后利用小波变换得到的梯度向量进行分水岭变换,有效保持边缘信息.实验结果证明该算法是可行的,与基于形态梯度的分割结果相比,得到了较好的分割效果.

Watershed Image Segmentation Based on Wavelet Transform

Image segmentation plays a very important role in digital image processing. A novel image segmentation method based on wavelet transform is proposed, which combines wavelet analysis, wavelet packet decomposition with watershed algorithm in mathematical morphology efficiently. First, wavelet packet de-noising is used to reduce the over segmentation. Then, the gradient vector obtained through wavelet transform instead of the morphological gradient is obtained to preserve the edge information and perform watershed algorithm. Experiments indicate that it is feasible. The segmentation results of wavelet gradient are much better than that of morphological gradient.