Journal of Geo-information Science ›› 2019, Vol. 21 ›› Issue (9): 1430-1443.doi: 10.12082/dqxxkx.2019.180628
Previous Articles Next Articles
ZHANG Yindan1,2,WANG Miaomiao1,LU Haixia1,LIU Yong1,*()
Received:
2018-12-04
Revised:
2019-05-21
Online:
2019-09-25
Published:
2019-09-24
Contact:
LIU Yong
E-mail:liuy@lzu.edu.cn
Supported by:
ZHANG Yindan,WANG Miaomiao,LU Haixia,LIU Yong. Comparing Supervised and Unsupervised Segmentation Evaluation Methods for Extracting Specific Land Cover from High-Resolution Remote Sensing Imagery[J].Journal of Geo-information Science, 2019, 21(9): 1430-1443.DOI:10.12082/dqxxkx.2019.180628
Add to citation manager EndNote|Reference Manager|ProCite|BibTeX|RefWorks
Tab. 5
Classification features and experiment parameters"
典型地物 | 分类特征 | 分类器参数 | |
---|---|---|---|
耕地/居民住宅 | 光谱特征 | 亮度值、最大最小值差、各波段的均值、标准差 | 树的个数分别为250、300;树的最大深度为20,每个节点上对象的最小数量为10,其他参数保持默认 |
纹理特征 | 灰度共生矩阵的同质性、对比度、相异性、信息熵、角度二阶距、均值、标准差和相关性 | ||
几何特征 | 面积、长宽比、不对称性、椭圆拟合度、矩形拟合度和形状指数 | ||
指数 | 归一化差分植被指数、归一化水指数 | ||
坑塘 | 光谱特征 | 亮度值、最大最小值差、各波段的均值、标准差 | 树的个数为100;树的最大深度为15;每个节点上对象的最小数量为5;其他参数保持默认 |
指数 | 归一化差分植被指数、归一化水指数 |
Tab. 6
Statistics of the training and testing sample datasets"
地物类型 | 影像 | 训练样本/个 | 验证样本/个 | ||
---|---|---|---|---|---|
目标类 | 其他类 | 目标类 | 其他类 | ||
耕地 | QB_MS | 314 | 2117 | 169 | 1229 |
WV2_MS | 311 | 828 | 381 | 337 | |
ALOS_MS | 66 | 293 | 55 | 231 | |
居民住宅 | QB_MS | 106 | 556 | 30 | 334 |
WV2_MS | 246 | 633 | 74 | 339 | |
ALOS_MS | 18 | 278 | 50 | 233 | |
坑塘 | QB_MS | 24 | 133 | 52 | 273 |
WV2_MS | 23 | 68 | 17 | 51 | |
ALOS_MS | 60 | 143 | 72 | 211 |
Tab. 7
Statistics of segmentation assessment for the typical land covers based the unsupervised method (ESP2)"
地物类型 | 影像 | 最优分割参数组合 | 分割精度评价 | ||||
---|---|---|---|---|---|---|---|
Scale | Shape | Cpt | USeg | OSeg | AFI | ||
耕地 | QB_MS | 78 | 0.1 | 0.1 | 0.9399 | 0.3651 | -9.5709 |
WV2_MS | 69 | 0.6 | 0.4 | 0.5821 | 0.0953 | -1.1649 | |
ALOS_MS | 47 | 0.1 | 0.1 | 0.3613 | 0.0563 | -0.4774 | |
居民住宅 | QB_MS | 76 | 0.7 | 0.5 | 0.9176 | 0.0256 | -10.8215 |
WV2_MS | 64 | 0.6 | 0.5 | 0.9511 | 0.1927 | -15.4960 | |
ALOS_MS | 39 | 0.3 | 0.4 | 0.7709 | 0.1180 | -2.8503 | |
坑塘 | QB_MS | 107 | 0.5 | 0.7 | 0.6616 | 0.0615 | -1.7730 |
WV2_MS | 102 | 0.3 | 0.6 | 0.0183 | 0.0907 | 0.0737 | |
ALOS_MS | 45 | 0.3 | 0.3 | 0.8862 | 0.0983 | -6.9251 |
Tab. 8
Statistics of classification assessment for the typical land covers based on the unsupervised method (ESP2)"
地物类型 | 影像 | 分类精度评价 | |||
---|---|---|---|---|---|
UA | PA | OA | KIA | ||
耕地 | QB_MS | 0.9830 | 0.5678 | 0.9715 | 0.7060 |
WV2_MS | 0.9985 | 0.9057 | 0.9208 | 0.7639 | |
ALOS_MS | 0.9839 | 0.8905 | 0.9453 | 0.8880 | |
居民住宅 | QB_MS | 0.7518 | 0.8272 | 0.9716 | 0.7725 |
WV2_MS | 0.9753 | 0.8537 | 0.9008 | 0.7902 | |
ALOS_MS | 0.9950 | 0.5515 | 0.9460 | 0.6826 | |
坑塘 | QB_MS | 1.0000 | 0.8406 | 0.9424 | 0.8708 |
WV2_MS | 1.0000 | 0.9282 | 0.9507 | 0.8900 | |
ALOS_MS | 0.9953 | 0.8859 | 0.9806 | 0.8790 |
Tab. 9
Statistics of the typical land cover segmentation assessment based on the supervised method (ED2)"
地物类型 | 影像 | 最优分割参数 | 分割精度评价 | ||||
---|---|---|---|---|---|---|---|
Scale | Shape | Cpt | USeg | OSeg | AFI | ||
耕地 | QB_MS | 17 | 0.1 | 0.1 | 0.2499 | 0.2873 | 0.0499 |
WV2_MS | 25 | 0.6 | 0.4 | 0.1080 | 0.0826 | -0.0285 | |
ALOS_MS | 27 | 0.1 | 0.1 | 0.1634 | 0.0985 | -0.0775 | |
居民住宅 | QB_MS | 24 | 0.7 | 0.5 | 0.2099 | 0.1421 | -0.0858 |
WV2_MS | 25 | 0.6 | 0.5 | 0.3272 | 0.3317 | 0.0068 | |
ALOS_MS | 31 | 0.3 | 0.4 | 0.6158 | 0.1375 | -1.2452 | |
坑塘 | QB_MS | 50 | 0.5 | 0.7 | 0.2167 | 0.0629 | -0.1963 |
WV2_MS | 154 | 0.3 | 0.6 | 0.0417 | 0.0968 | 0.0575 | |
ALOS_MS | 28 | 0.3 | 0.3 | 0.6613 | 0.1328 | -1.5603 |
Tab. 10
Statistics of the typical land cover classification assessment based on the supervised method (ED2)"
地物类型 | 影像 | 分类精度评价 | |||
---|---|---|---|---|---|
UA | PA | OA | KIA | ||
耕地 | QB_MS | 0.9968 | 0.7145 | 0.9814 | 0.8228 |
WV2_MS | 0.9939 | 0.9333 | 0.9400 | 0.8113 | |
ALOS_MS | 0.9910 | 0.9100 | 0.9567 | 0.9133 | |
居民住宅 | QB_MS | 0.9288 | 0.7312 | 0.9793 | 0.8074 |
WV2_MS | 0.9909 | 0.8987 | 0.9103 | 0.8473 | |
ALOS_MS | 0.9288 | 0.7264 | 0.9497 | 0.7091 | |
坑塘 | QB_MS | 1.0000 | 0.8953 | 0.9622 | 0.9161 |
WV2_MS | 0.9628 | 0.8642 | 0.9271 | 0.8351 | |
ALOS_MS | 0.9713 | 0.8953 | 0.9810 | 0.8880 |
[1] | Blaschke T, Hay G J, Kelly M , et al. Geographic object-based image analysis towards a new paradigm[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014,87(100):180-191. |
[2] | Baatz M and Schäpe A . Multiresolution segmentation: An optimization approach for high quality multi-scale image segmentation[J]. Angewandte Geographische Informationsverarbeitung XII, 2000,58:12-23. |
[3] | Belgiu M and Dr Gut L . Comparing supervised and unsupervised multiresolution segmentation approaches for extracting buildings from very high-resolution imagery[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014,96:67-75. |
[4] | Kim M, Madden M, Warner T . Estimation of optimal image object size for the segmentation of forest stands with multispectral IKONOS imagery[M]. Berlin: Springer Berlin Heidelberg, 2008: 291-307. |
[5] | 黄慧萍 . 面向对象影像分析中的尺度问题研究[D]. 北京:中国科学院遥感应用研究所, 2003: 103-121. |
[ Huang H P . Scale issues in object-oriented image analysis[D]. Beijing: Institute of Remote Sensing Applications Chinese Academy of Sciences, 2003: 103-121. ] | |
[6] | 张俊, 汪云甲, 李妍 , 等. 一种面向对象的高分辨率影像最优分割尺度选择算法[J]. 科技导报, 2009,27(921):91-94. |
[ Zhang J, Wang Y J, Li Y , et al. An object-oriented optimal scale choice method for high spatial resolution remote sensing image[J]. Science & Technology, 2009,27(921):91-94. ] | |
[7] | 何敏, 张文君, 王卫红 . 面向对象的最优分割尺度计算模型[J]. 大地测量与地球动力学, 2009,29(1):106-109. |
[ He M, Zhang W J, Wang W H . Optimal segmentation scale model based on object-oriented analysis method[J]. Journal of Geodesy and Geodynamics, 2009,29(1):106-109. ] | |
[8] | Johnson B A, Bragais M, Endo I , et al. Image segmentation parameter optimization considering within and between segment heterogeneity at multiple scale levels: Test case for mapping residential areas using landsat imagery[J]. ISPRS International Journal of Geo-Information, 2015,4(4):2292-2305. |
[9] | 王志华, 孟樊, 杨晓梅 , 等. 高空间分辨率遥感影像分割尺度参数自动选择研究[J]. 地球信息科学学报, 2016,18(5):639-648. |
[ Wang Z H, Meng F, Yang X M , et al. Study on the automatic selection of segmentation scale parameters for high spatial resolution remote sensing images[J]. Journal of Geo-information Science, 2016,18(5):639-648. ] | |
[10] | Wang, Y J, Qi Q W, Liu Y , et al. Unsupervised segmentation parameter selection using the local spatial statistics for remote sensing image segmentation[J]. International Journal of Applied Earth Observation and Geoinformation, 2019,81:98-109. |
[11] | Drǎguţ L, Tiedec D, Levickd S R . ESP: A tool to estimate scale parameter for multiresolution image segmentation of remotely sensed data[J]. International Journal of Geographical Information Science, 2010,24(6):859-871. |
[12] | Drăguţ L, Csillik O, Eisank C , et al. Automated parameterization for multi-scale image segmentation on multiple layers[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014,88:119-127. |
[13] | Sonka M, Hlavac V, Boyle R . Image processing, analysis, and machine vision[M]. United States: Cengege Learning, 2014: 236-240. |
[14] | Achanccaray P, Ayma V, Jimenez L , et al. SPT 3.1: A free software for automatic tuning of segmentation parameters in optical, hyperspectral and SAR images [C]. IEEE International IGARSS Geoscience and Remote Sensing Symposium. 2015. |
[15] | Anders N S, Seijmonsbergen A C, Bouten W Segmentation optimization and stratified object-based analysis for semi-automated geomorphological mapping[J]. Remote Sensing of Environment, 2011,115:2976-2985. |
[16] | Zhou Y N, Li J, Feng L , et al. Adaptive scale selection for multiscale segmentation of satellite images[J]. IEEE Journal of Selected Topics in Applied Earth Observations & Remote Sensing, 2017,10(8):3641-3651. |
[17] | Troya-Galvis A, Gancarski P, Passat N , et al. Unsupervised quantification of Under and over segmentation for Object-Based remote sensing image analysis[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2015,8(5):1936-1945. |
[18] | Bialas J, Oommen T, Havens T C . Optimal segmentation of high spatial resolution images for the classification of buildings using random forests[J]. International Journal of Applied Earth Observation and Geoinformation, 2019,32:82 |
[19] | Liu Y, Zhang Y D, Huang Z, et al. Segmentation optimization via recognition of the PSE-NSR-ED2 patterns along with the scale parameter in object-based image analysis[C]. GEOBIA, University of Twenty Faculty of Geoinformation and Earth Observation (ITC), Amsterdam, Netherlands, 2016. |
[20] | Wang Z, Yang X, Lu C , et al. A scale self-adapting segmentation approach and knowledge transfer for automatically updating land use/cover change databases using high spatial resolution images[J]. International Journal of Applied Earth Observation and Geoinformation, 2018,69:88-98. |
[21] | 陈扬洋, 明冬萍, 徐录 , 等. 高空间分辨率遥感影像分割定量实验评价方法综述[J]. 地球信息科学学报, 2017,19(6):818-830. |
[ Chen Y Y, Ming D P, Xu L , et al. Scale parameter estimation based on the spatial and spectral statistics in high spatial resolution image segmentation. Journal of Geo-information Science, 2017,19(6):622-631. ] | |
[22] | Clinton N, Holt A, Scarborough J , et al. Accuracy assessment measures for object-based image segmentation goodness[J]. Photogrammetric Engineering and remote sensing, 2010,76(3):289-299. |
[23] | Lucieer A and Stein A . Existential uncertainty of spatial objects segmented from satellite sensor imagery[J]. IEEE Transactions on Geoscience & Remote Sensing, 2002,40(11):2518-2521. |
[24] | Yang J, Li P, He Y . A multi-band approach to unsupervised scale parameter selection for multi-scale image segmentation[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014,94:13-24. |
[25] | Yue A Z, Yang J Y, Zhang C , et al. The optimal segmentation scale identification using multispectral WorldView-2 Images[J]. Sensor Letters, 2012,10(1-2):285-291. |
[26] | Breiman L . Random Forest[J]. Machine Learning, 2001,45:5-32. |
[27] | Congalton R G . A review of assessing the accuracy of classification of remotely sensed data[J]. Remote Sensing of Environment, 1991,37(2):35-46. |
[28] | Wang, Z H, Yang X M, Lu C , et al. A scale self-adapting segmentation approach and knowledge transfer for automatically updating land use/cover change databases using high spatial resolution images[J]. International Journal Applied Earth Obser vation Geoinformation, 2018,69:88-98. |
[1] | Fangming WU, Miao ZHANG, Bingfang WU. Object-oriented Rapid Estimation of Rice Acreage from UAV Imagery [J]. Journal of Geo-information Science, 2019, 21(5): 789-798. |
[2] | ZHANG Tao,SU Fenzhen,YANG Xiaomei,SUN Xiaoyu. A Method for Detecting Red Tide Information from MODIS Data and Its Application in Pearl River Estuary [J]. , 2009, 11(2): 244-249. |
[3] | LUO Cailian, XU Hanqiu. Auto-Detection of Land Cover Changes Based on Satellite Image Fusion Technique [J]. , 2005, 7(1): 111-115. |
[4] | WANG Chunlan, CHEN Jianfei. Extracting the Information of Concrete/Urban Development from High Spectrum Resolution Image of ASTER [J]. , 2004, 6(2): 106-108,124. |
|