| 基于蚁群和Canny边缘检测算子混合算法的二维岩石图像裂隙特征提取与修复研究 |
| |
| 引用本文： | 张愉玲,邢会林,李三忠,逄硕,刘骏标,张熔鑫.基于蚁群和Canny边缘检测算子混合算法的二维岩石图像裂隙特征提取与修复研究[J].大地构造与成矿学,2021,45(1):242-251. |
| |
| 作者姓名： | 张愉玲 邢会林 李三忠 逄硕 刘骏标 张熔鑫 |
| |
| 作者单位： | 深海圈层与地球系统前沿科学中心,海底科学与探测技术教育部重点实验室,中国海洋大学 海洋高等研究院/海洋地球科学学院,山东 青岛 266100;中国海洋大学 海底科学与工程计算国际中心,山东 青岛 266100;深海圈层与地球系统前沿科学中心,海底科学与探测技术教育部重点实验室,中国海洋大学 海洋高等研究院/海洋地球科学学院,山东 青岛 266100;中国海洋大学 海底科学与工程计算国际中心,山东 青岛 266100;青岛海洋科学与技术国家实验室 海洋矿产资源评价与探测技术功能实验室,山东 青岛 266100 |
| |
| 摘 要： | 岩石中常含有裂缝、岩脉等细小而重要的微观结构,需要高分辨率的成像技术来提取其特征。在岩石图像处理中,特征提取可以提取出岩石的主要结构信息从而极大降低数据量,然而某些细线状岩石特征(如裂缝、裂隙等)往往无法被完整识别出来而存在间断现象。本文在采用蚁群算法修复岩石细线状特征的基础上,综合利用蚁群算法、DBSCAN聚类算法和Canny算子提出一种新的岩石图像修复算法。根据设定的阈值,对提取的图像间断部分进行搜索并连接,以修复在图像特征提取过程中缺失的部分重要信息,同时去除修复过程产生的多余特征。实际结果表明,利用混合算法处理的岩石特征提取图像得到了较好的修复,能够提高岩石裂隙特征提取的准确性。
|
| 关 键 词： | 岩石裂隙 图像修复 特征提取 蚁群算法 Canny算子 |
Fracture Extraction and Repair of 2D Rock Image Based on Hybrid Algorithm of Ant Colony and Canny Edge Detection Operator |
| |
| ZHANG Yuling,XING Huilin,LI Sanzhong,PANG Shuo,LIU Junbiao,ZHANG Rongxin.Fracture Extraction and Repair of 2D Rock Image Based on Hybrid Algorithm of Ant Colony and Canny Edge Detection Operator[J].Geotectonica et Metallogenia,2021,45(1):242-251. |
| |
| Authors: | ZHANG Yuling XING Huilin LI Sanzhong PANG Shuo LIU Junbiao ZHANG Rongxin |
| |
| Institution: | (Frontiers Science Center for Deep Ocean Multispheres and Earth System,Key Lab of Submarine Geosciences and Prospecting Techniques,MOE and College of Marine Geosciences,Ocean University of China,Qingdao 266100,Shandong,China;International Center for Submarine Geosciences and Geoengineering Computing,Ocean University of China,Qingdao 266100,Shandong,China;Laboratory for Marine Mineral Resources,Qingdao National Laboratory for Marine Science and Technology,Qingdao 266100,Shandong,China) |
| |
| Abstract: | Rocks often contain some small but important microstructures such as cracks and veins,the features of which require high-resolution imaging technology to extract.In rock image processing,feature extraction can extract the main structural information of the rock with significantly less amount of data.However,some fine linear rock features(such as cracks)are often discontinuous and hard to identify.To repair fine linear features in rocks,a new repairment algorithm for rock images which comprehensively utilizes ant colony optimization,DBSCAN clustering algorithm and Canny operator is proposed based on the ant colony optimization.This method can search and connect the intermittent parts of the extracted image according to the set threshold,repair some important information missing in the process of image feature extraction,and remove the noise features generated during the repair process.The actual results showed that the accuracy of rock fracture feature extraction on the images processed by the hybrid algorithm is significantly improved. |
| |
| Keywords: | rock cracks image repairment feature extraction ant colony optimization Canny operator |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
