医学图像存在病变区域和背景区域,病变区域是分割的重点.针对传统分水岭算法对噪声敏感和易于严生过分割的问题,提出了一种将形态学滤波、多尺度形态学梯度和控制标记符分水岭相结合的分割策略.首先利用数学形态闭-开运算完成预处理以滤除原始医学图像中的噪声和非感知信息.其次做闭运算以平滑图像,并对平滑的图像计算多尺度的形态学梯度.再次对形态学梯度图像进行重建,然后采用控制标记符的分水岭变换算法对重建后的梯度图像进行分割.最后将分割结果变换回原始尺度.仿真实验结果表明,这种改进的方法不但使经典分水岭算法中的过分割现象得到了很好的抑制,医学图像中的病变区域被有效分割出来；而且分割算法简单,同时具有多尺度的特点,能够适应医学图像分类与信息提取的需求.%The medical images have pathological change region and background. The pathological change region that is so-called the region of interest (ROI) is the emphasis of image segmentation. Aimed at resolving the problems of sensitivity to noise and over-segmentation existing in traditional watershed algorithm, an image segmentation strategy on the combination of morphological filtering and multi-scale morphology gradient and marker controlled watershed is proposed. Firstly, use of mathematical morphological closed-open operation to filter out complete pretreatment original medical image to filter the noise and the non-perceptional information, in order to preserve the structural information of the original images; secondly, do closed operations smooth image, and calculate the image multi-scale morphology gradient, thirdly, watershed algorithm is reconstruced it and applied marker controlled to segment the reconstructed gradient image. Finally, the segmentation results is transfored to original scale. Simulation results show that the improvement can not only suppress the over-segmented phenomena properly, but also segment the pathological change regions in medical images efectively and segmentation algorithm is simple , and has the characteristics of multiple scales, thus is quite qualified for classification and information extraction of medical sensed imagery.
展开▼
