摘要
医学影像技术广泛应用于临床和科研领域,而医学数字图像数据量大、噪声情况复杂。图像分割是图像处理基本步骤之一,也是其应用和发展的瓶颈问题,提出了一种改进的分水岭分割算法,数字图像通过多小波变换分解、能量计算、分水岭变换和区域融合等步骤,有效的克服了分水岭变换严重的过分割问题,实现了有意义的区域分割。
Medical image is very popular in clinical medicine, which is very difficult for image processing. The paper presents a M-band wavelet-based watershed image segmentation method for medical digital image. The method is based on a multiresolution application of a M-band wavelet and watershed transformation, followed by a wavelet coefficient-based energy computation and region merging procedure. Experimental results show that the method is useful of the reduction of over-segmentation and can be applied to the segmentation of digital images.
出处
《数理医药学杂志》
2007年第3期298-300,共3页
Journal of Mathematical Medicine
关键词
多小波变换
分水岭变换
M-band wavelets transform
watershed transform
