1. School of Soil and Water Conservation, Beijing Forestry University, Beijing 100083, China
2. Key Laboratory of Soil and Water Conservation & Desertification Combating of Ministry of Education, Beijing Forestry University, Beijing 100083, China
3. Jinyun National Positioning Observation and Research Station of Forest Ecosystem in Three Gorges Reservoir Area, School of Soil and Water Conservation, Beijing Forestry University, Beijing 100083, China
4. Experimental Center of Desert Forestry, Chinese Academy of Forestry, Dengkou 015200, China
Abstract:Shelterbelt systems are the main type of vegetation in the desert oasis regions, which provide a strong guarantee for wind-break and sand fixation, salt-water regulation and water-heat balance. It is important to investigate the spatial distribution information of shelterbelts. However, precisely mapping shelterbelts systems on a large scale are difficult due to narrow strips, small patches and wide & scattered distribution. This study aims to accurately map shelterbelts using object-oriented extraction based on GF-2 satellite imagery in Dengkou oasis. Firstly, the optimal scale parameter of SF segmentation was determined by local variance (LV) and rate of change (ROC) curve, and then the features space and classifier’s parameters were optimized by Out of bag error (OOB error) and Gini index through Random Forest (RF) algorithm prior to classification. Finally, Random Forest, CART decision tree, Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) were compared and validated for shelterbelt systems extraction. The results showed that: (1) the ROC-LV curve method can obtain the possible value of optimal scale parameter more objective and more efficiently than iterating all scale parameter values. (2) OOB error and Gini index through RF algorithm can effectively eliminate the redundant features among spectral, shape and texture. The processing time was sharply reduced and ensuring the accuracy of the classification. (3) The classification results were verified based on the measured data sets, and the results showed that the feature optimization based on the RF algorithm combined with the SVM classifier was the best method for extracting the desert oasis shelterbelt systems, with the highest producer accuracy of 97.14%. Meanwhile, the extracted area of shelterbelt systems was 208.99 km2, which was close to reality (210 km2). The SVM classifier performs better than the other three classifiers while zooming in a small areas; (4) Due to the high resolution of GF-2 images and the near-infrared band, sub-meter information can be obtained through appropriate band fusion. Based on the object-oriented method, a single shelterbelt can be used as the basic unit to explore the attributes and characteristics of the shelterbelts net. For example, the broken shelterbelts information could be extracted. All these conclusions will provideimportant technical support for the shelterbeltextraction in the desert oasis areas.
Key words:Shelterbelt extraction; GF-2 remote-sensing imagery; Random Forest; Object-oriented
[1] LEI Si-jun, SUN Hua, LIU Hua, et al(雷思君, 孙 华, 刘 华, 等). Journal of Central South University of Forestry & Technology(中南林业科技大学学报), 2020, 40(4): 57.
[2] YANG Yi-tian, ZHENG Du, ZHANG Xue-qin, et al(杨依天, 郑 度, 张雪芹, 等). Acta Geographica Sinica(地理学报), 2013, 68(6): 813.
[3] HE Bao-zhong, DING Jian-li, ZHANG Zhe, et al(何宝忠, 丁建丽, 张 喆, 等). Acta Geographica Sinica(地理学报), 2016, 71(11): 1948.
[4] WU Jin-zhou, ZHENG Xiao, GAO Tian, et al(武金洲, 郑 晓, 高 添, 等). Chinese Journal of Ecology(生态学杂志), 2020, 39(11): 3567.
[5] Wiseman G, Kort J, Walker D. Agriculture, Ecosystems and Environment, 2009, 131(1): 111.
[6] XING Ze-feng, LI Ying, DENG Rong-xin, et al(幸泽峰, 李 颖, 邓荣鑫, 等). Scientia Silvae Sinicae(林业科学), 2016, 52(4): 11.
[7] GAO Meng-jie, JIANG Qun-ou, ZHAO Yi-yang, et al(高梦婕, 姜群鸥, 赵一阳, 等). Journal of China Agricultural University(中国农业大学学报), 2018, 23(8): 125.
[8] SUN Pan, DONG Yu-sen, CHEN Wei-tao, et al(孙 攀, 董玉森, 陈伟涛, 等). Remote Sensing for Natural Resources(国土资源遥感), 2016, 28(4): 108.
[9] TANG Huai-zhi, TANG Min, GUAN Ming-wen, et al(汤怀志, 汤 敏, 关明文, 等). Journal of China Agricultural University(中国农业大学学报), 2021, 26(4): 157.
[10] WU Ya-juan, LIU Ting-xi, TONG Xin, et al(邬亚娟, 刘廷玺, 童 新, 等). Arid Zone Research(干旱区研究), 2020, 37(4): 1026.
[11] Wulder Michael A, White Joanne C, Hay Geoffrey J, et al. The Forestry Chronicle, 2008, 84(02): 221.
[12] Drǎguţ L, Csillik O, Eisank C, et al. ISPRS Journal of Photogrammetry and Remote Sensing, 2014, 88: 119.
[13] Habermann Mateus, Fremont Vincent, Shiguemori Elcio Hideiti. International Journal of Remote Sensing, 2019, 40(10): 3900.
[14] Breiman L. Machine Learning, 2001, 45(1): 5.
[15] ZHAN Guo-qi, YANG Guo-dong, WANG Feng-yan, et al(詹国旗, 杨国东, 王凤艳, 等). Journal of Geo-Information Science(地球信息科学学报), 2018, 20(10): 1520.
[16] ZENG Wen, LIN Hui, LI Xin-yu, et al(曾 文, 林 辉, 李新宇, 等). Journal of Central South University of Forestry & Technology(中南林业科技大学学报), 2020, 40(7): 32.
[17] Pal M, Mather P M. International Journal of Remote Sensing, 2005, 26(5): 1007.
[18] HUANG Shuang-yan, YANG Liao, CHEN Xi, et al(黄双燕, 杨 辽, 陈 曦, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2018, 38(10): 3169.
[19] PENG Jia-yi, WANG Xin-jun, ZHU Lei, et al(彭佳忆, 王新军, 朱 磊, 等). Arid Zone Research(干旱区研究), 2019, 36(3): 771.