| 改进的区间二型模糊聚类遥感影像变化检测 |
| |
| 引用本文： | 苏艺凡,党建武,王阳萍,杨景玉.改进的区间二型模糊聚类遥感影像变化检测[J].测绘通报,2021,0(7):44-51,58. |
| |
| 作者姓名： | 苏艺凡 党建武 王阳萍 杨景玉 |
| |
| 作者单位： | 兰州交通大学电子与信息工程学院,甘肃 兰州730070;兰州交通大学电子与信息工程学院,甘肃 兰州730070;甘肃省人工智能与图形图像处理工程研究中心,甘肃 兰州730070 |
| |
| 基金项目： | 甘肃省科技计划（18JR3RA104）；国家市场监督管理总局科技计划（2019MK150）；甘肃省教育厅科技项目（2017D-08） |
| |
| 摘 要： | 遥感影像的复杂模糊性问题会干扰影像变化检测的结果，可引入区间二型模糊C均值聚类算法解决此问题，但算法参数的随机性导致检测结果不稳定。本文首先利用局部最优解优化萤火虫算法中的候选解，引入可变步长因子，以此自适应寻优区间二型模糊C均值聚类算法的模糊因子；然后结合寻优得到的模糊因子进行区间二型模糊C均值聚类，迭代更新隶属区间提取变化信息；最后通过基于复合梯形法则的加权Karnik-mendel算法降型和解模糊优化聚类中心，依据最大隶属度原则判断变化类型。通过试验验证，本文方法得到更优模糊因子和更精确的聚类中心，具有更好的稳健性，变化检测精度得到提高，检测得到的变化区域更加精细。
|
| 关 键 词： | 遥感影像变化检测 萤火虫算法 区问二型模糊C均值聚类 模糊因子 降型算法 |
| 收稿时间： | 2020-08-10 |
Remote sensing image change detection based on improved interval type-2 fuzzy clustering |
| |
| SU Yifan,DANG Jianwu,WANG Yangping,YANG Jingyu.Remote sensing image change detection based on improved interval type-2 fuzzy clustering[J].Bulletin of Surveying and Mapping,2021,0(7):44-51,58. |
| |
| Authors: | SU Yifan DANG Jianwu WANG Yangping YANG Jingyu |
| |
| Institution: | 1. School of Electronics and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China;2. Gansu Provincial Engineering Research Center for Artificial Intelligence and Graphics & Image Processing, Lanzhou 730070, China |
| |
| Abstract: | The complex fuzziness of remote sensing image can interfere with the result of image change detection so that the interval binary fuzzy C-means clustering algorithm is introduced to solve the problem. But the randomness of algorithm parameters will affect the accuracy of change detection. In this paper, the candidate solution of firefly algorithm is optimized by using local optimal solution, and the variable step size factor is introduced, so as to find the fuzzy factors of interval type-2 fuzzy C-means clustering algorithm adaptively. The interval type-2 fuzzy C-means clustering is carried out combining with the fuzzy factors obtained by optimization and the image change information is extracted by iteratively updating the membership degree. Finally, the weighted Karnik-mendel algorithm based on compound trapezoid rule is used to reduce type and resolve fuzzy to optimize clustering centre. And the change types are determined according to the principle of maximum membership. Through experimental verification, the method in this paper obtains the better fuzzy factors and more accurate clustering centers, has better robustness, and improves the change detection accuracy, and the detected change area is more elaborate. |
| |
| Keywords: | remote sensing image change detection firefly algorithm interval type-2 fuzzy C-means clustering fuzzy factors type-reduction algorithm |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《测绘通报》浏览原始摘要信息 |
| 点击此处可从《测绘通报》下载免费的PDF全文 |
|
