Please wait a minute...
img

Wechat

Remote Sensing Technology and Application  2020, Vol. 35 Issue (3): 623-633    DOI: 10.11873/j.issn.1004-0323.2020.3.0623
    
Selection and Evaluation of the Optimal Scale in Multi-scale Segmentation of Remote Sensing Images
Fang Wang1,2(),Wunian Yang1(),Jian Wang3,Bin Xie4,Jintong Ren1
1.Key Laboratory of Geo-spatial Information Technology of Ministry of Land and Resources of P. R. China, Chengdu University of Technology, Chengdu 610059, China
2.College of Geography and Resources Science, Neijiang Normal University, Neijiang 641100, China
3.Department of Civil Engineering, Neijiang Vocational&Technical Cllege, Neijiang 641000, China
4.Department of Surveying and Mapping Engineering, Sichuan College of Architectural Technology, Deyang 618000, China
Download:  HTML  PDF (14731KB) 
Export:  BibTeX | EndNote (RIS)      
Abstract  

Multi-scale segmentation is the premise and key step of Object-Based Image Analysis (OBIA). The quality of multi-scale segmentation directly affects the accuracy of object-oriented classification. However, scale selection and evaluation remains a challenge in multi-scale segmentation. According to the fact that the optimal segmentation scale of the remote sensing image is closely related to the complexity of the objects of the image, a top-down method to select the optimal scale based on the complexity of segmented objects is proposed. In the top-down segmentation process, image features of each segmented object are extracted to construct the complexity function, and the optimal scale of each object is determined by setting a threshold value and iterating calculation. Then, the segmentation results with the best scale are obtained and applied to the ZY-3 satellite multispectral image and the GF-2 fusion image to obtain segmentation and classification results. Qualitative visual evaluation method, unsupervised evaluation method and supervised classification evaluation method were used to compare them with results obtained by the optimal single-scale segmentation and the unsupervised evaluation method. The experimental results show that the method can accurately obtain the scale matching with the ground targets, and improve segmentation effect and the classification accuracy, it is of practical value.

Key words:  Mean-shift segmentation      Object-Based Image Analysis (OBIA)      Object complexity      Optimal segmentation scale      Scale selection and evaluation     
Received:  17 January 2019      Published:  10 July 2020
ZTFLH:  TP751  
Corresponding Authors:  Wunian Yang     E-mail:  398321192@qq.com;ywn@cdut.edu.cn
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
Fang Wang
Wunian Yang
Jian Wang
Bin Xie
Jintong Ren

Cite this article: 

Fang Wang,Wunian Yang,Jian Wang,Bin Xie,Jintong Ren. Selection and Evaluation of the Optimal Scale in Multi-scale Segmentation of Remote Sensing Images. Remote Sensing Technology and Application, 2020, 35(3): 623-633.

URL: 

http://www.rsta.ac.cn/EN/10.11873/j.issn.1004-0323.2020.3.0623     OR     http://www.rsta.ac.cn/EN/Y2020/V35/I3/623

Fig.1  Flow diagram of the study
Fig. 2  ZY-3 multispectral image, GF-2fusion image and the distribution of reference objects in the study area
影像获取时间空间分辨/m波段面积/km2地物类型地理位置
GF-22016.6.161.0蓝、绿、红、近红9.81建筑、水体、裸地、绿地、道路、耕地四川省隆昌县中东部
ZY-32017.8.275.8蓝、绿、红、近红34.11建筑、水体、裸地、绿地、道路四川省泸州市区
Table 1  Information about satellite images of the study area
Fig. 3  Multi-scale segmentation results based on ZY-3 multi-spectral image and GF-2 image using the proposed method
Fig.4  Segmentation results
Fig.5  Classification results
Fig.6  Local segmentation and classification results
影像ZY-3GF-2
方法SE-OSU-ES-OSE-OSU-ES-O
GS0.5800.6220.6440.5890.6320.660
Table 2  GS values of segmentation results obtained with the SE-OS, U-E, and SS-OS methods
精度土地利用类型ZY-3多光谱影像GF-2影像
SE-OSU-EO-SSE-OSU-EO-S
制图精度/%建筑用地90.1487.3288.3892.0485.5083.19
耕地///90.9082.8375.76
林地///96.9084.5493.81
绿地99.9899.7599.2492.7792.7792.77
道路87.7467.9265.1091.7891.7883.56
裸地88.4675.0067.3085.7185.7185.71
水体85.7176.6275.3293.5593.5593.55
用户精度/%建筑用地92.0981.5882.3095.4195.2490.38
耕地///95.7483.6794.94
林地///92.1689.1385.05
绿地90.4186.8482.9186.5275.5076.24
道路86.9285.7181.1891.7890.4187.14
裸地85.1976.4779.5585.7199.9685.71
水体99.9899.9799.9799.9799.9799.97
总体精度/%91.1084.4983.5692.7588.0486.08
Kappa0.8770.7820.7680.9120.8550.832
Table 3  The object-oriented classification precision
1 Ma Yanni, Ming Dongping, Yang Haiping. Scale Estimation of Object-oriented Image Analysis based on Spectral-spatial Statistics[J]. Journal of Remote Sensing, 2017,21(4):566-578.
1 马燕妮,明冬萍,杨海平. 面向对象影像多尺度分割最大异质性参数估计[J]. 遥感学报, 2017,21(4): 566-578.
2 Zhu Chengjie, Yang Shizhi, Cui Shengcheng, et al. Accuracy Evaluating Method for Object-based Segmentation of High Resolution Remote Sensing Image[J]. High Power Laser and Particle Beams, 2015,27(6):43-49.
2 朱成杰,杨世植,崔生成,等. 面向对象的高分辨率遥感影像分割精度评价方法[J]. 强激光与粒子束, 2015,27(6): 43-49.
3 Zhou Y, 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 and Remote Sensing, 2017, 10(8): 3641-3651. .
doi: 10.1109/JSTARS. 2017.2693993
4 Wang Fang, Yang Wunian, Deng Xiaoyu, et al. Discussion on Urban Ecological Land Classification Method based on GF-2 Data[J]. Science of Surveying and Mapping, 2018,43(3):71-76.
4 王芳,杨武年,邓晓宇,等. 高分二号数据的城市生态用地分类方法探讨[J]. 测绘科学, 2018,43(3):71-76.
5 Ming D P, Li J, Wang J Y, et al. Scale Parameter Selection by Spatial Statistics for GeOBIA: Using Mean-shift based Multi-scale Segmentation as an Example[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2015, 106:28-41. .
doi: 10.1016/j.isprsjprs.2015.04.010
6 Zhou Y N, Luo J C, Shen Z F, et al. Multiscale Water Body Extraction in Urban Environments from Satellite Images[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2014, 7(10): 4301-4312. .
doi: 10.1109/JSTARS.2014.2360436
7 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. DOI:.
doi: 10.1016/j.isprsjprs. 2013.09.014
8 Yang Haiping, Ming Dongping. Optimal Scales based Segmentation of High Spatial Resolution Remote Sensing Data[J].Journal of Geo-information Science,2016,18(5):632-638.
8 杨海平,明冬萍. 综合多层优选尺度的高分辨率影像分割[J]. 地球信息科学学报, 2016,18(5): 632-638.
9 Konik M, Bradtke K. Object-oriented Approach to Oil Spill Detection Using ENVISAT ASAR Images[J]. ISPRS Journal of Photogrammetry and Remote Sensing,2016,118:37-52. .
doi: 10.1016/j.isprsjprs.2016.04.006
10 Martha T R, Kerle N, Van Westen C J, et al. Object-oriented Analysis of Multi-temporal Panchromatic Images for Creation of Historical Landslide Inventories[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2012, 67(2): 105-119.
doi: 10.1016/j.isprsjprs.2011.11.004
11 Blaschke H. Object-oriented Image Analysis and Scale-space: Theory and Methods for Modeling and Evaluating Multi-scale Landscape Structure[J]. International Archives of Photogrammetry and Remote Sensing, 2001, 34: 22-29.
12 Wang Fang, Wang Jian, Xie Bin, et al. Discussion on Classification Method of Adaptive Scale based on Remote Sensing Image[J]. Science of Surveying and Mapping, 2019,44(11):156-163.
12 王芳,王建,谢兵,等. 一种遥感影像自适应分割尺度的分类方法[J]. 测绘科学, 2019,44(11): 156-163.
13 Bai Tao, Yang Guodong, Wang Fengyan, et al. Object-Oriented Optimal Segmentation Scale Calculation Model [J]. Journal of Jilin University (Earth Science Edition), 2020,50(1):304-312.
13 白韬,杨国东,王凤艳,等. 一种面向对象的最优分割尺度计算模型[J]. 吉林大学学报(地球科学版), 2020,50(1): 304-312.
14 Wang F, Yang W N, Ren J T. Adaptive Scale Selection in Multiscale Segmentation based on the Segmented Object Complexity of GF-2 Satellite Image[J]. Arabian Journal of Geosciences, 2019, 12(22).
doi: 10.1007/s12517-019-4879-x
15 Vieira M A, Formaggio A R, Rennó C D, et al. Object Based Image Analysis and Data Mining Applied to a Remotely Sensed Landsat Time-series to Map Sugarcane over Large Areas[J]. Remote Sensing of Environment, 2012, 123(8): 553-562.
doi: 10.1016/j.rse.2012.04.011
16 Blaschke T. Object based Image Analysis for Remote Sensing[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2010, 65(1): 2-16.
doi: 10.1016/j.isprsjprs.2009.06.004
17 Dey V, Zhang Y, Zhong M. A Review on Image Segmentation Techniques with Remote Sensing Perspective[J]. Pattern Recognition,2010,38(9):1277-1294.
18 Lin Wenjie, Li Yu. Parallel Regional Segmentation Method of High-resolution Remote Sensing Image based on Minimum Spanning Tree[J]. Remote Sensing, 2020, 12(5): 783. DOI:.
doi: 10.3390/rs12050783
19 Wang F, Yang W N, Ren J T. Selection and Evaluation of Optimal Segmentation Scale for High-resolution Remote Sensing Images based on Prior Thematic Maps and Image Features[J].Journal of Applied Remote Sensing, 2019, 13(1): 1-23, 23. .
doi: 10.1117/1.JRS.13.016507
20 Duro D C, Franklin S E, Dubé M G. A Comparison of Pixel-based and Object-based Image Analysis with Selected Machine Learning Algorithms for the Classification of Agricultural Landscapes Using SPOT-5 HRG Imagery[J]. Remote Sensing of Environment, 2012, 118(6): 259-272. .
doi: 10.1016/j.rse.2011.11.020
21 Zhang H, Fritts J E, Goldman S A. Image Segmentation Evaluation: A Survey of Unsupervised Methods[J]. Computer Vision & Image Understanding, 2008, 110(2): 260-280. .
doi: 10.1016/j.cviu.2007.08.003
22 Huang X, Zhang L. An Adaptive Mean-shift Analysis Approach for Object Extraction and Classification from Urban Hyperspectral Imagery[J]. IEEE Transactions on Geoscience and Remote Sensing, 2008, 46(12): 4173-4185. .
doi: 10.1109/TGRS.2008.2002577
23 Zhang Yifei, Lü Ke, Dai Shuangfeng, et al. A Sea-land Segmentation Algorithm for Remote Sensing Images based on Mean-Shift [J]. Optical Technique, 2016,42(1):39-45.
23 张毅飞,吕科,代双凤,等. 基于均值漂移的遥感图像海陆边界分割算法[J]. 光学技术, 2016,42(1): 39-45.
24 Wang Wehong, Xu Wentao, Xia Liegang, et al. Study for Parallel Segmentation Optimization Algorithm of Mean Shift in Large-scale Remote Sensing Image [J]. Journal of Chinese Computer Systems, 2015, 36(5): 1085-1090.
24 王卫红, 徐文涛, 夏列钢, 等. 大规模遥感影像Mean Shift并行分割优化算法研究[J]. 小型微型计算机系统, 2015,36(5): 1085-1090.
25 Comaniciu D, Meer P. Mean Shift: A Robust Approach Toward Feature Space Analysis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(5): 603-619. .
doi: 10.1109/34.1000236
26 Adelson E H, Anderson C H, Bergen J R, et al. Pyramid Methods in Image Processing[J]. Rca Engineer, 1983, 29(6). 33-41.
27 Lindeberg T. Scale-Space Theory in Computer Vision[M]. Springer Berlin, 1994, 256: 349-382. .
doi: 10.1007/978-1-4757-6465-9
28 Gotlieb C C, Kreyszig H E. Texture Descriptors based on Co-occurrence Matrices[J]. Computer Vision Graphics and Image Processing, 1990, 51(1): 70-86. 10.1016/S0734-189X(05)80063-5.
doi: 10.1016/S0734-189X(05)80063-5
29 Johnson B, Xie Z. Unsupervised Image Segmentation eEaluation and Refinement Using a Multi-scale Approach[J]. ISPRS Journal of Photogrammetry & Remote Sensing, 2011, 66(4): 473-483. .
doi: 10.1016/j.isprsjprs.2011.02.006
30 Haralick R M, Shapiro L G. Survey: Image Segmentation Techniques[J]. Computer Vision Graphics and Image Processing, 1985, 29(1): 100-132. 10.1016/S0734-189X(85)90153-7.
doi: 10.1016/S0734-189X(85)90153-7
31 Liu Jinli, Chen Zhao, Gao Jinping, et al. Research on the Method of Determining the Optimal Segmentation Scale for Tree Species Classification[J]. Scientia Silvae Sinicae, 2019, 55(11):95-104.
31 刘金丽,陈钊,高金萍,等. 高分影像树种分类的最优分割尺度确定方法[J]. 林业科学, 2019,55(11): 95-104.
[1] Rui YANG Su Yang. U-Net neural networks and its application in high resolution satellite image classification[J]. Remote Sensing Technology and Application, 0, (): 0 .
[2] . [J]. Remote Sensing Technology and Application, 1986, 1(1): 16 .
[3] . [J]. Remote Sensing Technology and Application, 1986, 1(1): 22 -32 .
[4] . [J]. Remote Sensing Technology and Application, 1986, 1(1): 8 -10 .
[5] . [J]. Remote Sensing Technology and Application, 1986, 1(1): 40 -45 .
[6] . [J]. Remote Sensing Technology and Application, 1986, 1(2): 22 -24 .
[7] . [J]. Remote Sensing Technology and Application, 1987, 2(2): 42 -48 .
[8] . [J]. Remote Sensing Technology and Application, 1987, 2(2): 70 .
[9] . [J]. Remote Sensing Technology and Application, 1987, 2(3): 67 .
[10] . [J]. Remote Sensing Technology and Application, 1987, 2(4): 1 -8 .