Technology of classifying lithology of typical karst areas based on machine learning: Taking the Pingguo area, Guangxi as an example
-
摘要: 快速准确识别碳酸盐岩对于岩溶区的基础设施建设和重大工程实施十分重要，通过遥感岩性分类实现碳酸盐岩的快速提取目前仍然是最高效的途径之一。文章基于Landsat和AW3D 30 DSM遥感数据，以广西平果地区典型岩溶区为研究对象，采用碳酸盐岩的可见光到短波红外的多光谱信息、熵和角二阶矩等纹理信息及曲率和坡度等地形特征，对平果地区岩溶分布区的碳酸盐岩、碎屑岩、第四系及水体进行岩性分类，在选取606个总体样本并验证303个分类样本的基础上，采用最大似然分类方法对区域岩性进行快速分类。结果表明：碳酸盐岩的生产者精度和用户精度分别达到94.54%和97.64%，基本能够实现碳酸盐岩的快速提取和准确识别的需求，在典型岩溶区的岩性分类方法中具有准确率高、实现路径简单、所需数据源易获取的特点，将为典型岩溶区的岩性快速分类提供一种新的思路。Abstract:
Karst is strongly developed in Southwest China, and the geological hazards specific to karst areas may cause serious damage to the local construction facilities; therefore, the rapid identification of carbonate rocks is of great significance for the planning of infrastructure construction such as electric power, transportation, etc. At present, the classification of lithology through remote sensing is still one of the most efficient ways. In this study, on the basis of visible to short-wave infrared multispectral information, texture information such as the second order moments of entropy and angular, and topographic features such as curvature and slope, we proposed a method to classify carbonate rocks, clasolite rocks, Quaternary sediments and water bodies by remote sensing in the Pingguo area to realize automatic extraction of carbonate rocks. This method presents its advantages of obtaining data sources with high accuracy in simple realization path in terms of lithology classification of typical karst areas, and this study can provide a new idea for rapid lithology classification in karst areas. Located in the southwest of Guangxi with subtropical monsoon climate, the Pingguo area is one of the most developed karst areas in China, in which mainly develops limestone, dolomite and their interbedded and interlayered layers of the Triassic Beisi Group and Luolou Group. Karst landforms are distributed in the north, central and southwest of the Pingguo area, and are dominated by the karst peaks and depressions, significantly different from the morphological characteristics of clasolite rocks in the same area. The process of lithology classification and carbonate rock identification in this study mainly included the following steps: data source selection and processing, texture feature extraction, terrain feature extraction, sample sketching and training, and final mapping. The data was mainly from Landsat 8 OLI and AW3D 30 DSM available for the public. The 8-channel data was obtained, based on the synthesis of five multispectral bands including BLUE, GREEN, RED, NIR, SWIR bands, texture features of second-order moments of entropy and angular with Principal Component Analysis, and topographic features such as curvature and slope. Then the maximum likelihood classification was used to carry out the lithology classification of the study area, and the distribution of carbonate rocks was finally obtained. The results show that the overall accuracy of the method in the study area is over 90%, and the kappa coefficient is more than 0.85, which proves that the classification model is effective and practical for the extraction of carbonate rocks in the study area. In addition, results of accuracy evaluation show that the accuracy is high for classification of carbonate rocks, with an accuracy more than 94%, but the accuracy of classifying water bodies and Quaternary is relatively poor. The main reason for this difference is that the Quaternary sediments are scattered in the carbonate and clasolite areas, and part of Quaternary sediments may exist in the area covered with vegetation which is easy to be judged as the area of carbonate or clasolite rocks. The main reason for the slightly lower accuracy in classification of water bodies is that some water bodies and the areas shaded by carbonate rocks are easily confused, and there are also cases where carbonate rocks are misclassified as water bodies. Based on multi-source remote sensing data, we conducted the maximum likelihood estimation with the consideration of spectral, topographic, textural and other multi-factors in order to develop a fast and accurate automatic extraction method for carbonate rocks in typical karst areas. In addition, we used this method to automatically classify carbonate rocks, clasolite rocks, Quaternary sediments and water bodies in the Pingguo area and evaluated the classification accuracy. We drew the conclusions as follows. (1) Based on data of Landsat and AW3D 30 DSM, we adopted multi-factor synergistic analysis of spectral information, textural features and topographic features of typical carbonate areas, and carried out automatic extraction of carbonate rocks, clasolite rocks, etc. in the Pingguo area, a typical karst area. The overall classification accuracy was greater than 93%. (2) There was a great difference in the surface morphology of carbonate rocks and clasolite rocks in the study area, which is one of the important factors for distinguishing these two types of rocks. Meanwhile, the slope has a significant effect on the accuracy of carbonate rock extraction, so it is suggested that a slope should be used as one of the factors in the classification in the process of comprehensive extraction of carbonate rocks. (3) Lithology is difficult to be automatically extracted with multi-spectral remote sensing information, due to the complex multi-solution of extraction and the spatial heterogeneity of lithology. Most of the current studies differentiated lithology through the inversion of mineral compositions from the perspective of mineralization and alteration. With the gradual development and improvement of machine learning and deep learning, the comprehensive extraction method integrated with spectral bands, texture and morphology will provide a new idea for the significant improvement of the accuracy of lithology extraction, which may be one of the important directions for future research on lithology extraction. -
Key words:
- carbonate rocks /
- remote sensing /
- maximum likelihood classification /
- information extraction /
- Pingguo
-
图 2 碳酸盐岩和碎屑岩的典型影像图（a.碳酸盐岩典型遥感影像图，b.碎屑岩典型遥感影像图）
Figure 2. Typical images of carbonate rocks and clastic rocks (a. remote sensing image of carbonate rock; b. remote sensing image of clasolite rock)
图 4 平果地区遥感影像图(左)和地形渲染图（右）
Figure 4. Map of remote sensing image (left) and rendering map of topography (right) of the Pingguo area
图 5 GLCM纹理特征量及不同波段组合遥感图像
Figure 5. GLCM texture feature quantities and remote sensing images in different band composites
图 6 碳酸盐岩、碎屑岩等选取样本示例
Figure 6. Sample selection examples of carbonate rock, clasolite rock, etc.
图 9 岩性分类问题区域遥感影像与提取结果对比图
Figure 9. Comparison of regional remote sensing images and extraction results for lithological classification
图 10 岩性分类精度与参与分类波段数的相关关系图
Figure 10. Correlation plot of lithology classification accuracy with the number of bands participating in the classification
表 1 样本可分离度
Table 1. Sample separability
样本类别 碳酸盐岩 碎屑岩 第四系覆盖物 水体 碳酸盐岩 — 1.80 1.98 1.99 碎屑岩 1.80 — 1.99 1.99 第四系覆盖物 1.98 1.99 — 1.96 水体 1.99 1.99 1.96 — 
下载: 导出CSV
表 2 平果地区岩性分类结果精度评价表
Table 2. Accuracy evaluation of lithology classification in the Pingguo area
类别 碎屑岩 第四系 碳酸盐岩 水体 生产者精度/% 92.02 93.71 94.54 87.09 用户精度/% 89.49 82.02 97.64 93.81 总体分类精度/% 93.79 卡帕系数 0.8797
下载: 导出CSV
-
[1] 袁道先. 岩溶石漠化问题的全球视野和我国的治理对策与经验[J]. 草业科学, 2008, 25(9):19-25. doi: 10.3969/j.issn.1001-0629.2008.09.009YUAN Daoxian. Global view on karst rock desertification and integrating control measures and experiences of China[J]. Pratacultural Science, 2008, 25(9): 19-25. doi: 10.3969/j.issn.1001-0629.2008.09.009 [2] 姚长宏, 蒋忠诚, 袁道先. 西南岩溶地区植被喀斯特效应[J]. 地球学报, 2001, 22(2):159-164. doi: 10.3321/j.issn:1006-3021.2001.02.013YAO Changhong, JIANG Zhongcheng, YUAN Daoxian. Vegetation karst effects on the karst area of Southwest China[J]. Acta Geoscientia Sinica, 2001, 22(2): 159-164. doi: 10.3321/j.issn:1006-3021.2001.02.013 [3] 蒙彦, 雷明堂. 岩溶塌陷研究现状及趋势分析[J]. 中国岩溶, 2019, 38(3):411-417. doi: 10.11932/karst20190311MENG Yan, LEI Mingtang. Analysis of situation and trend of sinkhole collapse[J]. Carsologica Sinica, 2019, 38(3): 411-417 doi: 10.11932/karst20190311 [4] 戴建玲, 雷明堂, 蒋小珍, 罗伟权. 极端气候与岩溶塌陷[J]. 中国矿业, 2020, 29(Suppl.2):402-404. doi: 10.12075/j.issn.1004-4051.2020.S2.082DAI Jianling, LEI Mingtang, JIANG Xiaozhen, LUO Weiquan. Extreme climate and sinkhole[J]. China Mining Magazine, 2020, 29(Suppl.2): 402-404. doi: 10.12075/j.issn.1004-4051.2020.S2.082 [5] 冯亚伟. 山东省岩溶塌陷分布规律及成因机制[J]. 中国岩溶, 2021, 40(2):205-214. doi: 10.11932/karst2021y01FENG Yawei. Distribution and genesis of karst collapse in Shandong Province[J]. Carsologica Sinica, 2021, 40(2): 205-214. doi: 10.11932/karst2021y01 [6] 罗小杰, 沈建. 我国岩溶地面塌陷研究进展与展望[J]. 中国岩溶, 2018, 37(1):101-111. doi: 10.11932/karst20180106LUO Xiaojie, SHEN Jian. Research progress and prospect of karst ground collapse in China[J]. Carsologica Sinica, 2018, 37(1): 101-111. doi: 10.11932/karst20180106 [7] 韩啸. 贵阳院岩溶中心在ARMS11展览会精彩亮相[EB/OL]. 2021-10-27. https://www. powerchina.cn/art/2021/10/27/art_7448_1246384.html. [8] 方晴. 浅谈岩溶地区特高压输电线路选线定位原则[J]. 科技资讯, 2015, 13(7):57. doi: 10.3969/j.issn.1672-3791.2015.07.046FANG Qing. Discussion on the principle of UHV transmission line location in karst area[J]. Science & Technology Information, 2015, 13(7): 57. doi: 10.3969/j.issn.1672-3791.2015.07.046 [9] 曹文庆, 王海, 黄河. 浅析特高压输电线路岩溶地区岩土工程勘测[J]. 资源环境与工程, 2013, 27(6):761-764, 811.CAO Wenqing, WANG Hai, HUANG He. Preliminary analysis of UHV transmission line projects in karst region[J]. Resources Environment & Engineering, 2013, 27(6): 761-764, 811 [10] 舒守荣. 碳酸盐岩石最佳遥感波段选择的叠合光谱图方法[J]. 中国岩溶, 1982, 1(2):152-157.SHU Shourong. The coincident spectral plot method for selecting the optimal remote sensing bands of carbonate rocks[J]. Carsologica Sinica, 1982, 1(2): 152-157. [11] 刘超群. 碳酸盐岩地区遥感岩性信息提取方法研究[M]. 桂林:中国地质科学院, 2007.LIU Chaoqun. The study on remote sensing lithologic information mapping method in carbonate terrane[M]. Guilin: Chinese Academy of Geological Sciences, 2007. [12] 莫源富. 西南岩溶地区植被覆盖条件下的碳酸盐岩岩性遥感识别研究[D]. 长沙:中南大学, 2010.MO Yuanfu. Lithological discrimination of carbonate rocks covered by vegetation using remote sensing data in southwestern karst area, China[D]. Changsha: Central South University, 2010. [13] 莫源富, 奚小双. 植被覆盖茂密区碳酸盐岩岩性的遥感识别:以灌江流域为例[J]. 桂林理工大学学报, 2010, 30(1):41-46.MO Yuanfu, XI Xiaoshuang. Carbonate rock lithological discrimination by remote sensing data for areas with flourishing vegetation: A case from Guanjiang drainage area[J]. Journal of Guilin University of Technology, 2010, 30(1): 41-46. [14] 谢相建. 地表裸露碳酸盐岩组分比例遥感估算研究:以云南建水县为例[J]. 测绘学报, 2018, 47(10):1427. doi: 10.11947/j.AGCS.2018.20170615XIE Xiangjian. Estimation of exposed carbonate rock fraction with remote sensing imagery: A case study of Jianshui county[J]. Acta Geodaetica et Cartographica Sinica, 2018, 47(10): 1427. doi: 10.11947/j.AGCS.2018.20170615 [15] 谢相建. 地表裸露碳酸盐岩组分比例遥感估算研究:以云南建水县为例[D]. 南京:南京大学, 2016.XIE Xiangjian. Estimation of exposed carbonate rock fraction with remote sensing imagery: A case study of Jianshui county[D]. Nanjing: Nanjing University, 2016. [16] 杨云. 机器学习支持下多源遥感数据的岩性分类研究[D]. 成都:成都理工大学,2019.YANG Yun. Research on lithology classification of multi-source remote sensing data supported by machine learning[D]. Chengdu: Chengdu University of Technology,2019. [17] 覃小群, 邓艳, 蓝芙宁, 侯满福. 基于GIS技术的典型岩溶石山区土壤侵蚀危险性评价:以广西平果县果化示范区为例[J]. 安全与环境工程, 2005, 12(4):69-72. doi: 10.3969/j.issn.1671-1556.2005.04.020QIN Xiaoqun, DENG Yan, LAN Funing, HOU Manfu. Assessment on the soil erosion in the karst mountainous region based on GIS: Taking Guohua Ecological Target Area for example[J]. Safety and Environmental Engineering, 2005, 12(4): 69-72. doi: 10.3969/j.issn.1671-1556.2005.04.020 [18] 李晓青, 阳倩妮, 周楷淳, 罗为群. 喀斯特地区不同岩性上农村居民点分布特征:以平果市为例[J]. 中国岩溶, 2021, 40(2):355-362. doi: 10.11932/karst20210212LI Xiaoqing, YANG Qianni, ZHOU Kaichun, LUO Weiqun. Distribution characteristics of rural settlement on different lithology in karst area: A case study of Pingguo City[J]. Carsologica Sinica, 2021, 40(2): 355-362. doi: 10.11932/karst20210212 [19] 王增林, 朱大明. 基于遥感影像的最大似然分类算法的探讨[J]. 河南科学, 2010, 28(11):458-461. doi: 10.3969/j.issn.1004-3918.2010.11.024WANG Zenglin, ZHU Daming. A study of maximum likelihood classification algorithm based on remote sensing image[J]. Henan Science, 2010, 28(11): 458-461. doi: 10.3969/j.issn.1004-3918.2010.11.024 [20] John A Richards, Jia Xiuping. Remote Sensing Digital Image Analysis[M]. Berlin: Springer-Verlag, 2006: 196-197. [21] 张斌, 张志, 帅爽, 张耀明. 利用Landsat-8和Worldview-2数据进行协同岩性分类[J]. 地质科技情报, 2015, 34(3):208-229.ZHANG Bin, ZHANG Zhi, SHUAI Shuang, ZHANG Yaoming. Lithological mapping by using the synergestic Landsat-8 and Worldview-2 images[J]. Geological Science and Technology Information, 2015, 34(3): 208-229. [22] 杜小锋, 冯稳, 杨青雄. 基于资源三号卫星影像的岩性监督分类研究[J]. 资源环境与工程, 2018, 32(2):291-295.DU Xiaofeng, FENG Wen, YANG Qingxiong. The supervised classification of lithology based on ZY-3 image[J]. Resources Environment & Engineering, 2018, 32(2): 291-295. -
点击查看大图
图(10) / 表(2)
计量
- 文章访问数: 47
- HTML浏览量: 18
- PDF下载量: 24
- 被引次数: 0
