Intelligent identification of remnant ridge edges in region west of Yongxing Island, South China Sea
Intelligent identification of remnant ridge edges in region west of Yongxing Island, South China Sea
复制标题
南海永兴岛以西地区残脊边缘智能识别
DOI：
10.1007/s11802-018-3508-8
复制
发表时间：
2018-02
影响因子：
1.6
通讯作者：
Wang Dawei
中科院分区：
文献类型：
--
作者：
Wang Weiwei;Guo Jing;Cai Quanqiang;Wang Dawei
边缘检测能够识别地貌单元边界，从而有助于地貌制图。本文提出了一种智能边缘识别方法，并将图像处理技术应用于多波束测深数据。为此，通过水深测量生成彩色图像，并使用加权方法将彩色图像转换为灰色图像。由于图像的质量对边缘检测有很大的影响，因此采用不同的滤波方法对灰度图像进行去噪。计算峰值信噪比和均方误差以评估哪种滤波器方法最适合于深度图像滤波，并且随后使用图像二值化方法检测边缘。针对传统的图像二值化方法无法处理复杂不平海底的问题，提出了一种基于图像像素值差值的二值化方法，根据水平和垂直方向上相邻两个像素值差值的概率分布，分别估计图像二值化的合适阈值。最后，采用八邻域框架对二值图像进行细化，连接间断边缘，实现轮廓提取。实验结果表明，该方法能有效地识别地貌单元的主要边界。此外，本文提出的自动边缘识别方法避免了使用主观判断，减少了时间和人力成本。
Edge detection enables identification of geomorphologic unit boundaries and thus assists with geomorphical mapping. In this paper, an intelligent edge identification method is proposed and image processing techniques are applied to multi-beam bathymetry data. To accomplish this, a color image is generated by the bathymetry, and a weighted method is used to convert the color image to a gray image. As the quality of the image has a significant influence on edge detection, different filter methods are applied to the gray image for de-noising. The peak signal-to-noise ratio and mean square error are calculated to evaluate which filter method is most appropriate for depth image filtering and the edge is subsequently detected using an image binarization method. Traditional image binarization methods cannot manage the complicated uneven seafloor, and therefore a binarization method is proposed that is based on the difference between image pixel values; the appropriate threshold for image binarization is estimated according to the probability distribution of pixel value differences between two adjacent pixels in horizontal and vertical directions, respectively. Finally, an eight-neighborhood frame is adopted to thin the binary image, connect the intermittent edge, and implement contour extraction. Experimental results show that the method described here can recognize the main boundaries of geomorphologic units. In addition, the proposed automatic edge identification method avoids use of subjective judgment, and reduces time and labor costs.
登录
查看更多内容
影响因子：
4.4
作者：
V. Raghavan;S. Masumoto;K. Koike;Shuichi Nagano
通讯作者：
V. Raghavan;S. Masumoto;K. Koike;Shuichi Nagano
影响因子：
1.2
作者：
A. Heap;Peter T. Harris
通讯作者：
A. Heap;Peter T. Harris
DOI：
10.1007/978-3-642-15549-9_1
发表时间：
2010-09
影响因子：
23.6
作者：
Kaiming He;Jian Sun-;Xiaoou Tang
通讯作者：
Kaiming He;Jian Sun-;Xiaoou Tang
影响因子：
10.6
作者：
Lei Zheng;Andrew K. Chan
通讯作者：
Lei Zheng;Andrew K. Chan
DOI：
--
发表时间：
2007
期刊：
--
影响因子：
--
作者：
Sun Chun-yan
通讯作者：
Sun Chun-yan