热带地理 ›› 2020, Vol. 40 ›› Issue (2): 289-302.doi: 10.13284/j.cnki.rddl.003233
赵馨1, 周忠发1(), 王玲玉1, 骆剑承2, 孙营伟2, 刘巍2, 吴田军3
收稿日期:
2019-11-29
修回日期:
2020-03-16
出版日期:
2020-03-10
发布日期:
2020-05-15
通讯作者:
周忠发
E-mail:fa6897@163.com
作者简介:
赵馨(1996—),女,贵州人,硕士研究生,主要研究方向为地理信息系统与遥感,(E-mail) 1303873722@qq.com。
基金资助:
Zhao Xin1, Zhou Zhongfa1(), Wang Lingyu1, Luo Jiancheng2, Sun Yingwei2, Liu Wei2, Wu Tianjun3
Received:
2019-11-29
Revised:
2020-03-16
Online:
2020-03-10
Published:
2020-05-15
Contact:
Zhou Zhongfa
E-mail:fa6897@163.com
摘要:
选取北盘江镇与花江镇作为研究区,利用谷歌高精度遥感影像,结合分区分层分级思想,基于深度学习与传统约束方法对研究区耕地进行精准提取。结果表明:1)在数量精度上,以视觉形态差异对研究区进行分区并选取不同模型训练获得的精细地块,面积约为9 867 hm 2,与实际数量基本一致,F-Measure主要分布在[0.82, 0.98]之间,受到地形和岩石裸露率的影响,石漠化严重地区耕地提取精度较低。2)在形态精度上,预测耕地与实际耕地的GIOU主要分布在[0.7, 1]之间,分割正确率>0.85,表明预测耕地边界与实际地块边界吻合度高,提取结果符合研究区实际情况。3)利用地貌坡度等约束条件对耕地进行划分发现,研究区以喀斯特耕地为主,占比74%,并且石漠化程度较严重,其中轻度石漠化耕地与中度石漠化耕地占总耕地的32%。在石漠化地区,耕地的狭长程度、破碎程度受人类影响较大;距离居民地越近、可达性越高的地区,其土地利用率越高,斑块越破碎。文章所提出方法可用于耕地破碎、地形复杂地区的耕地提取,能够为地区的发展、耕地治理与研究、环境保护与决策等提供精准数据支持。
中图分类号:
赵馨, 周忠发, 王玲玉, 骆剑承, 孙营伟, 刘巍, 吴田军. 喀斯特山区石漠化耕地遥感精准提取与分析——以贵州省北盘江镇与花江镇为例[J]. 热带地理, 2020, 40(2): 289-302.
Zhao Xin, Zhou Zhongfa, Wang Lingyu, Luo Jiancheng, Sun Yingwei, Liu Wei, Wu Tianjun. Extraction and Analysis of Cultivated Land Experiencing Rocky Desertification in Karst Mountain Areas Based on Remote Sensing—A Case Study of Beipanjiang Town and Huajiang Town in Guizhou Province[J]. Tropical Geography, 2020, 40(2): 289-302.
表1
精度验证数据分布1"
TA/块 | TP/块 | EP/块 | FP/块 | P | R | F | TA/块 | TP/块 | EP/块 | FP/块 | P | R | F |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|
33 | 23 | 4 | 6 | 0.85 | 0.79 | 0.82 | 77 | 67 | 6 | 4 | 0.92 | 0.94 | 0.93 |
37 | 26 | 6 | 5 | 0.81 | 0.84 | 0.83 | 32 | 28 | 4 | 0 | 0.88 | 1.00 | 0.93 |
33 | 24 | 4 | 5 | 0.86 | 0.83 | 0.84 | 32 | 28 | 1 | 3 | 0.97 | 0.90 | 0.93 |
27 | 20 | 3 | 4 | 0.87 | 0.83 | 0.85 | 57 | 50 | 5 | 2 | 0.91 | 0.96 | 0.93 |
37 | 28 | 3 | 6 | 0.90 | 0.82 | 0.86 | 33 | 29 | 0 | 4 | 1.00 | 0.88 | 0.94 |
33 | 25 | 4 | 4 | 0.86 | 0.86 | 0.86 | 73 | 65 | 4 | 4 | 0.94 | 0.94 | 0.94 |
30 | 23 | 3 | 4 | 0.88 | 0.85 | 0.87 | 101 | 90 | 7 | 4 | 0.93 | 0.96 | 0.94 |
48 | 37 | 6 | 5 | 0.86 | 0.88 | 0.87 | 66 | 59 | 3 | 4 | 0.95 | 0.94 | 0.94 |
34 | 27 | 3 | 4 | 0.90 | 0.87 | 0.89 | 58 | 52 | 2 | 4 | 0.96 | 0.93 | 0.95 |
25 | 20 | 2 | 3 | 0.91 | 0.87 | 0.89 | 126 | 113 | 7 | 6 | 0.94 | 0.95 | 0.95 |
31 | 25 | 0 | 6 | 1.00 | 0.81 | 0.89 | 45 | 41 | 3 | 1 | 0.93 | 0.98 | 0.95 |
42 | 34 | 3 | 5 | 0.92 | 0.87 | 0.89 | 34 | 31 | 0 | 3 | 1.00 | 0.91 | 0.95 |
48 | 39 | 4 | 5 | 0.91 | 0.89 | 0.90 | 34 | 31 | 0 | 3 | 1.00 | 0.91 | 0.95 |
27 | 22 | 0 | 5 | 1.00 | 0.81 | 0.90 | 35 | 32 | 2 | 1 | 0.94 | 0.97 | 0.96 |
96 | 80 | 7 | 9 | 0.92 | 0.90 | 0.91 | 47 | 43 | 2 | 2 | 0.96 | 0.96 | 0.96 |
30 | 25 | 3 | 2 | 0.89 | 0.93 | 0.91 | 61 | 56 | 2 | 3 | 0.97 | 0.95 | 0.96 |
48 | 40 | 5 | 3 | 0.89 | 0.93 | 0.91 | 74 | 68 | 2 | 4 | 0.97 | 0.94 | 0.96 |
59 | 50 | 6 | 3 | 0.89 | 0.94 | 0.92 | 27 | 25 | 2 | 0 | 0.93 | 1.00 | 0.96 |
29 | 25 | 3 | 1 | 0.89 | 0.96 | 0.93 | 105 | 98 | 5 | 2 | 0.95 | 0.98 | 0.97 |
29 | 25 | 1 | 3 | 0.96 | 0.89 | 0.93 | 28 | 27 | 1 | 0 | 0.96 | 1.00 | 0.98 |
表2
精度验证数据分布2"
序号 | 实际斑块数/块 | 预测斑块数/块 | 实际面积/hm2 | 预测面积/ hm2 | 地貌类型 |
---|---|---|---|---|---|
1 | 111 | 105 | 9 | 9 | 无明显石漠化 |
2 | 72 | 103 | 4.717 2 | 8.391 1 | 轻度石漠化 |
3 | 116 | 124 | 9 | 9 | 无明显石漠化 |
4 | 56 | 72 | 6.431 2 | 6.562 4 | 无明显、中度石漠化 |
5 | 30 | 40 | 7.572 4 | 7.666 2 | 轻度石漠化 |
6 | 18 | 10 | 7.489 3 | 5.362 3 | 轻度、中度石漠化 |
7 | 18 | 24 | 2.761 7 | 2.875 | 轻度、中度石漠化 |
8 | 56 | 70 | 8.937 1 | 8.424 2 | 轻度石漠化 |
9 | 24 | 30 | 7.572 4 | 7.666 2 | 轻度、中度石漠化 |
10 | 55 | 64 | 5.119 5 | 5.562 8 | 轻度石漠化 |
11 | 30 | 40 | 3.141 2 | 3.024 5 | 中度石漠化 |
12 | 4 | 3 | 0.809 9 | 0.743 5 | 中度、重度石漠化 |
13 | 4 | 3 | 0.383 3 | 0.225 4 | 中度石漠化 |
14 | 25 | 33 | 5.080 9 | 5.269 3 | 非喀斯特地区 |
15 | 17 | 19 | 1.283 9 | 1.147 3 | 中度石漠化 |
16 | 4 | 5 | 0.528 | 0.504 8 | 中度石漠化 |
17 | 13 | 15 | 0.918 3 | 0.902 5 | 轻度、中度石漠化 |
18 | 50 | 51 | 3.734 6 | 3.171 6 | 无明显石漠化 |
19 | 59 | 68 | 4.966 | 4.739 2 | 中度石漠化 |
20 | 70 | 95 | 5.308 9 | 5.437 | 中度石漠化 |
总数 | 832 | 974 | 94.755 8 | 95.675 3 | — |
[1] |
Bartholomé E and Belward A S . 2005. GLC2000: A New Approach to Global Land Cover Mapping from Earth Observation Data. International Journal of Remote Sensing, 26(9):1959-1977.
doi: 10.1080/01431160412331291297 |
[2] | Bengio Y and Delalleau O . 2011. On the Expressive Power of Deep Architectures. In: Tapio Elomaa, Jaakko Hollmén and Heikki Mannila. Proceedings of the 14th International Conference on Discovery Science. Berlin: Springer-Verlag, 18-36. |
[3] | Bontemps S, Defourny P, Brockmann C, Schouten L, Vancutsem C, Huc M, Bontemps S, Leroy M, Frédéric A, Herold M, Ranera F and Arino O . 2011. GLOBCOVER2009 Products Description and Validation Report. ( 2011-02-18) [2019-11-25]. . |
[4] | Chaurasia A and Culurciello E . 2017. LinkNet: Exploiting Encoder Representations for Efficient Semantic Segmentation. St Petersburg, USA: IEEE Visual Communications and Image Processing (VCIP). |
[5] | 曹雪, 柯长青 . 2006. 基于对象级的高分辨率遥感影像分类. 遥感信息,(5):27-30,51,73. |
[ Cao Xue and Ke Changqing . 2006. Classification of High-Resolution Remote Sensing Images Using Object-Oriented Method. Remote Sensing Information,( (5):27-30,51,73. ] | |
[6] |
查勇, 倪绍祥, 杨山 . 2003. 一种利用TM图像自动提取城镇用地信息的有效方法. 遥感学报, 7(1):37-40.
doi: 10.11834/jrs.20030107 |
[ Zha Yong, Ni Shaoxiang and Yang Shan . 2003. An Effective Approach to Automatically Extract Urban Land-Use from TM Imagery. Journal of Remote Sensing, 7(1):37-40. ]
doi: 10.11834/jrs.20030107 |
|
[7] | 陈云浩, 冯通, 史培军, 王今飞 . 2006. 基于面向对象和规则的遥感影像分类研究. 武汉大学学报(信息科学版), 31(4):316-320. |
[ Chen Yunhao, Feng Tong, Shi Peijun and Wang Jinfei . 2006. Classification of Remot Sensing Image Based on Object Oriented and Class Rules. Geomatics and Information Science of Wuhan University, 31(4):316-320. ] | |
[8] |
Dong Wen, Wu Tianjun, Luo Jiancheng, Sun Yingwei and Xia Liegang . 2019. Land Parcel-Based Digital Soil Mapping of Soil Nutrient Properties in an Alluvial-Diluvia Plain Agricultural Area in China. Geoderma, 340:234-248.
doi: 10.1016/j.geoderma.2019.01.018 |
[9] |
Everingham M, Van Gool L, Williams C K I, Winn J and Zisserman A . 2010. The Pascal Visual Object Classes (VOC) Challenge. International Journal of Computer Vision, 88(2):303-338.
doi: 10.1007/s11263-009-0275-4 |
[10] |
Friedl M A, Sulla Menashe D, Tan B, Schneider A, Ramankutty N and Sibley A . 2010. Modis Collection 5 Global Land Cover: Algorithm Refinements and Characterization of New Datasets. Remote Sensing of Environment, 114(1):168-182.
doi: 10.1016/j.rse.2009.08.016 |
[11] |
Fritz S, See L, You L, Justice C, Becker-Reshef I, Bydekerke L and Woodcock C . 2013. The Need for Improved Maps of Global Cropland. Eos, Transactions American Geophysical Union, 94(3):31-32.
doi: 10.1002/2013EO030006 |
[12] |
傅泽强, 蔡运龙, 杨友孝, 戴尔阜 . 2001. 中国粮食安全与耕地资源变化的相关分析. 自然资源学报, 16(4):313-319.
doi: 10.11849/zrzyxb.2001.04.003 |
[ Fu Zeqiang, Cai Yunlong, Yang Youxiao and Dai Erfu . 2001. Research on the Relationship of Cultivated Land Change and Food Security in China. Journal of Natural Resources, 16(4):313-319. ]
doi: 10.11849/zrzyxb.2001.04.003 |
|
[13] |
Gong Peng, Wang Jie, Yu Le, Zhao Yongchao, Zhao Yuanyuan, Liang Lu, Niu Zhenguo, Huang Xiaomeng, Fu Haohuan, Liu Shuang, Li Congcong, Li Xueyan, Fu Wei, Liu Caixia, Xu Yue, Wang Xiaoyi, Cheng Qu, Hu Luanyun, Yao Wenbo, Zhang Han, Zhu Peng, Zhao Ziying, Zhang Haiying, Zheng Yaomin, Ji Luyan, Zhang Yawen, Chen Han, Yan An, Guo Jianhong, Yu Liang, Wang Lei, Liu Xiaojun, Shi Tingting, Zhu Menghua, Chen Yanlei, Yang Guangwen, Tang Ping, Xu Bing, Giri Chandra, Clinton Nicholas, Zhu Zhiliang, Chen Jin and Chen Jun . 2013. Finer Resolution Observation and Monitoring of Global Land Cover: First Mapping Results with Landsat TM and ETM+ Data. International Journal of Remote Sensing, 34(7):2607-2654.
doi: 10.1080/01431161.2012.748992 |
[14] |
Hansen M C and Reed B C . 2000 a. A Comparison of the IGBP DISCover and University of Maryland Global Land Cover Products. International Journal of Remote Sensing, 21(6/7):1365-1373.
doi: 10.1080/014311600210218 |
[15] |
Hansen M C, Defries R S, Townshend J R G and Sohlberg R A . 2000 b. Global Land Cover Classi Cation at 1km Spatial Resolution Using a Classification Tree Approach. International Journal of Remote Sensing, 21(6/7):1331-1364.
doi: 10.1080/014311600210209 |
[16] | 胡潭高, 朱文泉, 阳小琼, 潘耀忠, 张锦水 . 2009. 高分辨率遥感图像耕地地块提取方法研究. 光谱学与光谱分析, 29(10):2703-2707. |
[ Hu Tangao, Zhu Wenquan, Yang Xiaoqiong, Pan Yaozhong and Zhang Jinshui . 2009. Study on Extraction of Cultivated Land in High Resolution Remote Sensing Images. Spectroscopy and Spectral Analysis, 29(10):2703-2707. ] | |
[17] |
Liu Yun, Cheng Mingming, Hu Xiaowei, Bian Jiawang, Zhang Le, Bai Xiang and Tang Jinhui . 2019. Richer Convolutional Features for Edge Detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 41(8):1-8.
doi: 10.1109/TPAMI.2018.2878849 pmid: 30387723 |
[18] | 刘逸竹 . 2018. 大区域灌溉耕地制图技术研究. 北京:中国农业科学院. |
[ Liu Yizhu . 2018. Study on Mapping Technology of Irrigation Cultivated Land in Large Area. Beijing: Chinese Academy of Agricultural Sciences. ] | |
[19] | 骆剑承, 王钦敏, 马江洪, 周成虎, 梁怡 . 2002. 遥感图像最大似然分类方法的EM改进算法. 测绘学报, 31(3):234-239. |
[ Luo Jiancheng, Wang Qinmin, Ma Jianghong, Zhou Chenghu Lian Yi , The EM-Based Maximum Likelihood Classifier for Remotely Sensed Data. Acta Geodaetica Et Cartographic Sinica, 31(3):234-239. ] | |
[20] |
Moran M S, Inoue Y and Barnes E M . 1997. Opportunities and Limitations for Image-Based Remote Sensing in Precision Crop Management. Remote Sensing of Environment, 61(3):319-346.
doi: 10.1016/S0034-4257(97)00045-X |
[21] | Rezatofighi H, Tsoi N, Gwak J. Sadeghian A, Reid I and Savarese S . 2019. Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression. Long Beach, USA: IEEE Conference on Computer Vision and Pattern Recognition. |
[22] | Simonyan K and Zisserman A . 2014. Very Deep Convolutional Networks for Large-Scale Image Recognition. Computer Science, 9(4):1-14. |
[23] | 孙家波 . 2014. 基于知识的高分辨率遥感影像耕地自动提取技术研究. 北京:中国农业大学. |
[ Sun Jiabo . 2014. Research on Automatic Extraction of Cultivated Land Based on Knowledge-based High-Resolution Remote Sensing Images. Beijing: China Agricultural University. ] | |
[24] | 王世杰 . 2003. 喀斯特石漠化——中国西南最严重的生态地质环境问题. 矿物岩石地球化学通报, 22(2):120-126. |
[ Wang Shijie . 2003. The Most Serious Eco-Geologically environmental Problem In Southwestern China—Karst Rocky Desertification. Bulletin of Mineralogy , Petrology and Geochemistry, 22(2):120-126. ] | |
[25] |
Xie S and Tu Z . 2017. Holistically-Nested Edge Detection. International Journal of Computer Vision, 125(1/3):3-18.
doi: 10.1007/s11263-017-1004-z |
[26] | 熊康宁, 黎平, 周忠发 . 2002. 喀斯特石漠化的遥感-GIS典型研究. 北京: 地质出版社. |
[ Xiong Kangning, Li Ping and Zhou Zhongfa. 2002. The RS and GIS Representative Study on Karst Rock Desertification. Beijing: Geology Press. ] | |
[27] | 于利峰, 乌兰吐雅, 乌云德吉, 许洪滔, 包珺玮, 任婷婷 . 2018. 基于纹理特征与MODIS-NDVI时间序列的耕地面积提取研究. 中国农业资源与区划, 39(11):169-177. |
[ Yu Lifeng, Wulan Tuya, Wuyun Deji, Xu Hongtao, Bao Junwei and Ren Tingting . 2018. Study on Extaction of Aable Land Area Based on Texture Features and MODIS-NDVI Time Series. Chinese Journal of Agricultural Resources and Regional Planning, 39(11):169-177. ] | |
[28] | Zhou L, Zhang C and Wu M . 2018. D-LinkNet: LinkNet with Pretrained Encoder and Dilated Convolution for High Resolution Satellite Imagery Road Extraction. Salt Lake City, Utah: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW). |
[1] | 刘文祥, 舒远仲, 唐小敏, 刘金梅. 采用双注意力机制Deeplabv3+算法的遥感影像语义分割[J]. 热带地理, 2020, 40(2): 303-313. |
[2] | 孙中宇, 荆文龙, 乔曦, 杨龙. 基于无人机遥感的盛花期薇甘菊爆发点识别与监测[J]. 热带地理, 2019, 39(4): 482-491. |
[3] | 黄登红,周忠发,吴跃,朱孟,尹林江,崔亮. 基于无人机可见光影像的高原丘陵盆地区山药植株识别[J]. 热带地理, 2019, 39(4): 571-582. |
[4] | 张凤太, 邵技新, 苏维词. 贵州喀斯特山区乡村景观生态优化模式研究[J]. 热带地理, 2009, 29(5): 418-422. |
[5] | 吴良林, 周永章, 陈子燊, 宋书巧, 阎小平. 广西喀斯特山区原生态旅游资源脆弱性及其安全保护[J]. 热带地理, 2008, 28(1): 74-79. |
|