摘要
针对现有的图像分割方法无法准确地分割声纳图像的问题,提出了一种改进的水平集声纳图像分割方法。介绍了LBF能量模型,借鉴其无重初始化的水平集演化思想。为克服声纳图像中复杂背景带来的负面效应,利用形态学顶帽—底帽变换对声纳图像进行预处理,并在此基础上进行无需初始化的水平集分割。进行仿真对比实验,实验结果显示:与LBF能量模型相比,改进的水平集分割方法更加适应于背景不均匀的声纳图像分割。
Aiming at the problem that the existing image segmentation methods cannot be accurately applied in sonar image segmentation, an improved level set sonar image segmentation method is proposed. LBF energy model is introduced and used for reference for its idea of level set evolution without re-initialization. To overcome negative effects caused by the complex background in sonar image, sonar image is preprocessed by morphological top-hat and bottom-hat transformation, and on these basis, the level set segmentation without re-initialization is carried up. Compare the improved level set method with the LBF energy model method in the simulation experiment, and the results show that the improved level set segmentation method is more suitable for sonar image segmentation with uneven background.
出处
《传感器与微系统》
CSCD
北大核心
2012年第1期29-31,共3页
Transducer and Microsystem Technologies
基金
国家自然科学基金资助项目(50909025
60772128)
关键词
声纳图像分割
LBF能量模型
水平集
形态学运算
sonar image segmentation
LBF energy model
level set
morphological operations
