改进的分水岭算法在医学图像分割中的应用
|
摘要
分水岭算法是一种常用的图像分割方法,由于分水岭算法是基于的图像灰度梯度,对噪声很敏感,直接运用分水岭算法分割图像,很容易产生过分割。为了去除过度分割,为此本文提出了一种改进的分水岭分割方法,首先利用中值滤波消除噪声,再用形态学基本运算得到梯度图像,然后利用形态学开、闭操作重建梯度图像,最后通过实验证明,基于数学形态学的分水岭算法分割医学图像效果优于传统的分水岭算法。
Watershed algorithm is a commonly used image segmentation. Watershed algorithm is based on image gray gradient and sensitive to noise.It is easy to generate over-segmentation by directly using watershed algorithm to segment images. In order to remove the over-segmentation, an improved watershed segmentation method is proposed in this paper. Firstly, the median filter is used to eliminate noise, and the gradient image is obtained by morphological basic operations, then the gradient image is reconstructed by morphological open and close operations.Finally, experiments show that the watershed algorithm based on mathematical morphology is better than the traditional watershed algorithm in medical image segmentation.
Watershed algorithm is a commonly used image segmentation. Watershed algorithm is based on image gray gradient and sensitive to noise.It is easy to generate over-segmentation by directly using watershed algorithm to segment images. In order to remove the over-segmentation, an improved watershed segmentation method is proposed in this paper. Firstly, the median filter is used to eliminate noise, and the gradient image is obtained by morphological basic operations, then the gradient image is reconstructed by morphological open and close operations.Finally, experiments show that the watershed algorithm based on mathematical morphology is better than the traditional watershed algorithm in medical image segmentation.
引文
[1]杨玚,谢华成.基于马尔科夫随机场与模拟退火算法的图像分割[J].软件,2015,36(4):40-43.
[2]郑金志,郑金敏,汪玉琳.基于优化初始聚类中心的改进WFCM图像分割算法[J].软件,2015,36(4):136-142.
[3]郭璇,郑菲,赵若晗,等.基于阈值的医学图像分割技术的计算机模拟及应用[J].软件,2018,39(3):12-15.
[4]董默,赵若晗,周志尊,等.医学图像三维重建系统设计与应用[J].软件,2018,39(1):87-90.
[5]戴青云,余英林.数学形态学在图像处理中的应用进展[J].控制理论与应用,2001,18(4):478-482.
[6]李玲玲,余文勇.基于数学形态学的灰度线形态识别研究与开发[J].计算机工程与应用,2002,11:104-106.
[7]盛仲飙.基于对比度增强的分水岭分割算法[J].计算机技术与发展.2012(12):90-92.
[8]陈洁,胡永,刘泽国.基于标记的分水岭图像分割算法研究[J].软件,2012,33(9):115-117
[9]Vincent L.Soille P.Watersheds in digital spaces:an efficient algorithm based on immersion simulations.IEEE Trans.Patt.Anal.Mach.Intell.1991,13(6):583-598.
[10]蔡迪,段汕.形态分水岭变换在图像分割中的应用[J].中南民族大学学报,2005,24(4):100-103.
[2]郑金志,郑金敏,汪玉琳.基于优化初始聚类中心的改进WFCM图像分割算法[J].软件,2015,36(4):136-142.
[3]郭璇,郑菲,赵若晗,等.基于阈值的医学图像分割技术的计算机模拟及应用[J].软件,2018,39(3):12-15.
[4]董默,赵若晗,周志尊,等.医学图像三维重建系统设计与应用[J].软件,2018,39(1):87-90.
[5]戴青云,余英林.数学形态学在图像处理中的应用进展[J].控制理论与应用,2001,18(4):478-482.
[6]李玲玲,余文勇.基于数学形态学的灰度线形态识别研究与开发[J].计算机工程与应用,2002,11:104-106.
[7]盛仲飙.基于对比度增强的分水岭分割算法[J].计算机技术与发展.2012(12):90-92.
[8]陈洁,胡永,刘泽国.基于标记的分水岭图像分割算法研究[J].软件,2012,33(9):115-117
[9]Vincent L.Soille P.Watersheds in digital spaces:an efficient algorithm based on immersion simulations.IEEE Trans.Patt.Anal.Mach.Intell.1991,13(6):583-598.
[10]蔡迪,段汕.形态分水岭变换在图像分割中的应用[J].中南民族大学学报,2005,24(4):100-103.