复杂背景下的风力机组叶片图像分割方法研究
|
摘要
针对单一的图像分割方法对复杂背景下的风力机组叶片图像进行分割时,得到的结果不具备准确性的问题,提出一种将Canny边缘检测与形态学分析相结合的方法,对Canny算法处理之后的二值图像进行腐蚀膨胀,去除背景区域。在分割一种具有颜色差异的风机叶片图像时,将RGB图像转化为HSV图像,再分别对HSV图像的色度、饱和度、和明度分量进行处理,并运用开运算去除图像噪声,得到理想的分割结果。实验结果表明,提出的图像分割方法能得到完整的风机叶片图像分割结果。
When a single image segmentation method is used to segment a wind turbine blade image under a complex background, the results obtained are not accurate and complete. This paper proposes a method of combinating Canny edge detection and morphological analysis, in which the binary image processed by the Canny algorithm is etched and expanded to remove the background area. When it segments a fan blade image with color difference, firstly, the RGB image is converted into a HSV image, and then the chroma, saturation, and brightness components of the HSV image are processed respectively. Finally, an open operation is used to remove image noise and an ideal segmentation result is obtained. Experimental results show that the image segmentation method proposed in this paper can obtain a complete fan blade image segmentation results.
When a single image segmentation method is used to segment a wind turbine blade image under a complex background, the results obtained are not accurate and complete. This paper proposes a method of combinating Canny edge detection and morphological analysis, in which the binary image processed by the Canny algorithm is etched and expanded to remove the background area. When it segments a fan blade image with color difference, firstly, the RGB image is converted into a HSV image, and then the chroma, saturation, and brightness components of the HSV image are processed respectively. Finally, an open operation is used to remove image noise and an ideal segmentation result is obtained. Experimental results show that the image segmentation method proposed in this paper can obtain a complete fan blade image segmentation results.
引文
[1] 周莉莉,姜枫.图像分割方法综述研究[J].计算机应用研究,2017,34(7):1921-1928.
[2] 钱青,臧冬菊.一种改进的Sobel算子图像清晰度评价函数[J].计算机与数字工程,2015,43(10):1865-1870.
[3] 许宏科,秦严严,陈会茹.一种基于改进Canny的边缘检测算法[J].红外技术,2014,36(3):210-214.
[4] 袁小翠,吴禄慎,陈华伟.基于Otsu方法的钢轨图像分割[J].光学精密工程,2016,24(7):1772-1781.
[5] 王淑青,姚伟,陈进,等.基于直方图均衡化与形态学处理的边缘检测[J].计算机应用与软件,2016,33(3):193-196.
[6] 陈虎,凌朝东,张浩,等.基于FPGA的实时彩色图像边缘检测算法的实现[J].液晶与显示,2015,30(1):143-150.
[7] LIU J Y,TANG Q J,WANG Y,et al. Defects’ geometric feature recognition based on infrared image edge detection[J]. Infrared Physics and Technology, 2014(67): 387-390.
[8] GONG M G, TIAN D Y, SU L ZH, et al. An efficient bi-convex fuzzy variational image segmentation method[J]. Information Sciences, 2015(293): 351-369.
[9] 贺振东,王耀南,刘洁,等.基于背景差分的高铁钢轨表面缺陷图像分割[J].仪器仪表学报,2016,37(3):640-649.
[10] 雷俊,王立辉,何芸倩,等.适用于机器人视觉的图像分割方法[J].系统工程与电子技术,2017,39(7):1653-1659.
[11] 王文豪,周静波,高尚兵,等.基于HSV空间改进的多尺度显著性检测[J].计算机工程与科学,2017,39(2):364-370.
[12] 王江明,余烨,金强,等.多梯度融合的RGBD图像边缘检测[J].电子测量与仪器学报,2017,31(3):436-442.
[13] 陶强,刘莉.基于双区域演化水平集的图像分割方法[J].电子测量技术,2016,39(9):91-95,107.
[2] 钱青,臧冬菊.一种改进的Sobel算子图像清晰度评价函数[J].计算机与数字工程,2015,43(10):1865-1870.
[3] 许宏科,秦严严,陈会茹.一种基于改进Canny的边缘检测算法[J].红外技术,2014,36(3):210-214.
[4] 袁小翠,吴禄慎,陈华伟.基于Otsu方法的钢轨图像分割[J].光学精密工程,2016,24(7):1772-1781.
[5] 王淑青,姚伟,陈进,等.基于直方图均衡化与形态学处理的边缘检测[J].计算机应用与软件,2016,33(3):193-196.
[6] 陈虎,凌朝东,张浩,等.基于FPGA的实时彩色图像边缘检测算法的实现[J].液晶与显示,2015,30(1):143-150.
[7] LIU J Y,TANG Q J,WANG Y,et al. Defects’ geometric feature recognition based on infrared image edge detection[J]. Infrared Physics and Technology, 2014(67): 387-390.
[8] GONG M G, TIAN D Y, SU L ZH, et al. An efficient bi-convex fuzzy variational image segmentation method[J]. Information Sciences, 2015(293): 351-369.
[9] 贺振东,王耀南,刘洁,等.基于背景差分的高铁钢轨表面缺陷图像分割[J].仪器仪表学报,2016,37(3):640-649.
[10] 雷俊,王立辉,何芸倩,等.适用于机器人视觉的图像分割方法[J].系统工程与电子技术,2017,39(7):1653-1659.
[11] 王文豪,周静波,高尚兵,等.基于HSV空间改进的多尺度显著性检测[J].计算机工程与科学,2017,39(2):364-370.
[12] 王江明,余烨,金强,等.多梯度融合的RGBD图像边缘检测[J].电子测量与仪器学报,2017,31(3):436-442.
[13] 陶强,刘莉.基于双区域演化水平集的图像分割方法[J].电子测量技术,2016,39(9):91-95,107.