【摘要】 为了提高激光医学图像的分割精度,针对传统分水岭方法存在的不足,提出一种改进分水岭和区域增长的激光医学图像分割方法。首先对采激光医学图像进行双边滤波,消除噪声,防止边缘细节信息丢失,然后分水岭方法对激光医学图像进行分割,并针对传统分水岭方法不足进行相应的改进,最后采用区域增长方法对图像分割的结果进行处理,使分割目标更加完整,并通过仿真实验分析其有效性。结果表明,相对于传统分水岭方法以及其它分割方法,本文方法获得更高了的激光医学图像分割精度,对噪声鲁棒性更强,可以为临床医学诊断提供重要依据。更多还原

【Abstract】 In order to improve the segmentation accuracy of laser medical images,a new method of laser medical image segmentation based on watershed and region growing is proposed to improve the existing problems of the traditional watershed method. Firstly,laser medical image acquisition of bilateral filtering is produced,to eliminate the noise and prevent the loss of the edge details and watershed method of laser medical image segmentation,and the corresponding improvement aiming at the deficiency of traditional watershed method,finally to deal with the results of laser medical image segmentation using region growing method,make the segmentation more complete,and through simulation analysis of the intersection. The results show that when compared with the watershed method and other segmentation methods,this method obtains higher laser medical image segmentation accuracy and noise robustness,can provide an important basis for the diagnosis of clinical medicine.更多还原