Object-based Karst wetland vegetation classification using UAV images

doi:10.13474/j.cnki.11-2246.2020.0346

Abstract

Abstract: This study aims to classify Karst wetland vegetation on Huixian National Wetland Park, located in Guilin, Guangxi province using object-based image analysis technique, random forest algorithm, image thresholding approach and Boruta all-related features selection algorithm based on UAV images. Results are as follows: the contribution of different feature variables is described as follows: spectral feature (DOM spectral > DSM spectral) > texture feature (DOM texture > DSM texture) > geometric feature > contextual feature; the overall classification accuracy of two UAV data sets is above 85 % as well as Kappa coefficient. This study provides insights into feature variable selection, segmentation parameter setting and classification method selection for karst wetland vegetation classification using high spatial resolution UAV RGB images.

Key words: objected-based, Karst wetland, UAV images, multi-resolution segmentation, feature selection

CLC Number:

P237

GENG Renfang, FU Bolin, JIN Shuanggen, CAI Jiangtao, GENG Wanxuan, LOU Peiqing. Object-based Karst wetland vegetation classification using UAV images[J]. Bulletin of Surveying and Mapping, 2020, 0(11): 13-18.

References

[1] BENDING J, YU K, AASEN H, et al. Combining UAV-based plant height from crop surface models, visible, and near infrared vegetation indices for biomass monitoring in barley[J]. International Journal of Applied Earth Observation and Geoinformation, 2015, 39:79-87.
[2] ZHANG C H, KOVACS J M. The application of small unmanned aerial systems for precision agriculture:a review[J]. Precision Agriculture, 2012, 13(6):693-712.
[3] LI Q S, WONG F K K, FUNG T. Assessing the utility of UAV-borne hyperspectral image and photogrammetry derived 3D data for wetland species distribution quick mapping[J]. The International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences, 2017, 42:209-215.
[4] CHABOT D, DILLON C, SHEMROCK A, et al. An object-based image analysis workflow for monitoring shallow-water aquatic vegetation in multispectral drone imagery[J]. ISPRS International Journal of Geo-Information, 2018, 7(8):294.
[5] PANDE-CHHETRI R, ABD-ELRAHMAN A, LIU T, et al. Object-based classification of wetland vegetation using very high-resolution unmanned air system imagery[J]. European Journal of Remote Sensing, 2017, 50(1):564-576.
[6] 邵亚,蔡崇法,赵悦,等. 桂林会仙湿地沉积物中磷形态及分布特征[J].环境工程学报,2014, 8(12):5311-5317.
[7] 耿仁方,付波霖,蔡江涛,等.基于无人机影像和面向对象随机森林算法的岩溶湿地植被识别方法研究[J].地球信息科学学报,2019, 21(8):1295-1306.
[8] BELGIU M, DRAGUT L. Comparing supervised and unsu-pervised multiresolution segmentation approaches for extracting buildings from very high resolution imagery[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014, 96:67-75.
[9] 刘金丽,陈钊.Landsat8多光谱数据辅助下的高分影像多尺度分割[J]. 测绘通报,2019(9):38-43.
[10] DRAGUT L, CSILLIK O, EISANK C, et al. Automated parameterisation for multi-scale image segmentation on multiple layers[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2014, 88:119-127.
[11] DRAGUT L, TIEDE D, LEVICK 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] KURSA M B, RUDNICKI W R. Feature selection with the Boruta package[J]. Journal of Statistical Software, 2010, 36(11):1-13.
[13] RAJBHANDARI S, ARYAL J, OSBORN J, et al. Leveraging machine learning to extend ontology-driven geographic object-based image analysis (O-GEOBIA):a case study in forest-type mapping[J]. Remote Sensing, 2019, 11(5):503.
[14] 张志强,张新长,辛秦川,等.结合像元级和目标级的高分辨率遥感影像建筑物变化检测[J]. 测绘学报, 2018, 47(1):102-112.
[15] GISLASON P O, BENEDIKTSSON J A, SVEINSSON J R. Random forests for land cover classification[J]. Pattern Recognition Letters, 2006, 27(4):294-300.