《计算机应用研究》|Application Research of Computers

一种用细胞神经网提取遥感图像边缘的新方法

Novel CNN algorithm for extracting remote sensing image edge

免费全文下载 (已被下载 次)  
获取PDF全文
作者 徐国保,洪丽兰,郝彦爽,尹怡欣,沈玉利
机构 1.北京科技大学 信息工程学院,北京 100083;2.广东海洋大学 信息学院,广东 湛江 524088;3.仲恺农业技术学院,广州 510225
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2008)11-3504-03
DOI 10.3969/j.issn.1001-3695.2008.11.092
摘要 针对灰度遥感图像具有噪声多、图像亮度均匀、边缘模糊等特点,提出了基于细胞神经网遥感图像边缘检测的新方法。该算法主要是利用细胞神经网先后对遥感图像进行图像滤波、灰度阈值化、膨胀腐蚀、边缘检测等模板操作。实验结果表明,与传统的Sobel和Canny边缘检测算法相比,本算法不仅能有效地去除噪声对边缘检测的影响,而且能够快速完整地提取图像边缘。
关键词 细胞神经网;遥感图像;边缘提取;模板
基金项目 广东省自然科学基金资助项目(7010116);湛江市科技攻关资助项目(2008C08014)
本文URL http://www.arocmag.com/article/1001-3695(2008)11-3504-03.html
英文标题 Novel CNN algorithm for extracting remote sensing image edge
作者英文名 XU Guo-bao, HONG Li-lan, HAO Yan-shuang, YIN Yi-xin, SHEN Yu-li
机构英文名 1. School of Information Engineering, University of Science & Technology Beijing, Beijing 100083, China; 2. School of Information, Guangdong Ocean University, Zhanjiang Guangdong 524088, China; 3. Zhongkai University of Agriculture & Technology, Guangzhou 510225, China
英文摘要 As gray remote sensing image had the characteristics of much noise, image brightness uniformity, and fuzzy edge, a novel edge detection method based on cellular neural networks (CNN)was presented. In the algorithm, image filtering, gray threshold segmentation, dilation and erosion, and edge detection using CNN were performed for remote sensing image successively. The experimental results show that the proposed algorithm, compared with the triditional edge detection algorithms of Sobel and Canny can not only effectively eliminate the influence of noise on edge detection, but quickly detect the complete image edge.
英文关键词 cellular neural networks; remote sensing image; edge detection; template
参考文献 查看稿件参考文献
 
收稿日期
修回日期
页码 3504-3506
中图分类号
文献标志码 A