【摘要】 用分水岭方法进行图像分割时,容易造成图像的过度分割。为了克服这种缺点,提出了一种基于小波变换和分水岭算法的图像分割方法,该方法首先利用小波变换产生多分辨率图像,然后对最低分辨率图像进行应用标记的分水岭分割,得到初始的分割区域,最后利用区域标记和小波反变换,得到高分辨率图像的分水岭分割结果,从而较好地解决了分水岭变换方法中的过度分割问题。更多还原

【Abstract】 Watershed based image segmentation has the drawback of over-segmentation.Combining the wavelet and watersheds,this paper proposes an image segmentation method which solves this problem.Wavelet transform is first used to create multi-resolution images.Then the marker-controlled watershed segmentation algorithm is applied to segment the lowest-resolution image and get the initial watershed segmentation.At last,labeled images and inverse wavelet transform are used together to get the full-resolution image.The experiment results show that this method is a solution for the over-segmentation problem.更多还原