Please wait a minute...
img

Wechat

Remote Sensing Technology and Application  2021, Vol. 36 Issue (4): 926-935    DOI: 10.11873/j.issn.1004-0323.2021.4.0926
    
Researches on Grass Species Fine Identification based on UAV Hyperspectral Images in Three-River Source Region
Yina Hu1(),Ru An1(),Zetian Ai2,Weibing Du3
1.College of Hydrology and Water Resources,Hohai University,Nanjing 211100,China
2.School of Geographic Information and tourism,Chuzhou University,Chuzhou 239000,China
3.School of Surveying and mapping and Land Information Engineering,Henan Polytechnic University,Jiaozuo 454003,China
Download:  HTML  PDF (6085KB) 
Export:  BibTeX | EndNote (RIS)      
Abstract  

Fine identification of grass species is of great significance for grassland ecosystem degradation monitoring in the Three Rivers Source Region. Based on the UAV hyperspectral remote sensing system, the hyperspectral image of the typical grassland degradation area of Three-River Source Region was obtained. Firstly, using the obtained UAV hyperspectral image, the optimal bands combination were selected using XGBoost, the extended morphological attribute profile features were extracted and were combined with the selected spectral features. Secondly, sparse multinomial logistic regression and adaptive sparse representation methods were adopted to identify different grass species. Finally the shape adaptive based post-processing method was proposed to smooth the identification results. The results showed that: (1) Using the XGBoost method to select important spectral features can improve the identification result and save running time; (2) the spatial-spectral feature based method can effectively improve the identification result of grass species and the overall accuracy were improved by 4%~5% compared with the method of using only spectral features; (3) using two sparse representation methods,the overall accuracy of fine identification of grass species in the case of limited samples was 94.07% and 93.15% respectively, and the identification accuracy of various poison weed species was improved effectively by using shape adaptive post-processing method, which improved the overall accuracy by about 1.64% and 1.12%, respectively. The feature mining based sparse representation classification methods can achieve high-precision grass species fine identification of UAV hyperspectral images, and provide technical support for a wider range of grassland species fine identification.

Key words:  UAV Hyperspectral Images      Grass species fine identification      Feature mining      Shape adaptive      Sparse representation      Three-River Source     
Received:  30 April 2020      Published:  26 September 2021
ZTFLH:  P237  
Corresponding Authors:  Ru An     E-mail:  yinahu7@163.com;anrunj@163.com
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
Yina Hu
Ru An
Zetian Ai
Weibing Du

Cite this article: 

Yina Hu,Ru An,Zetian Ai,Weibing Du. Researches on Grass Species Fine Identification based on UAV Hyperspectral Images in Three-River Source Region. Remote Sensing Technology and Application, 2021, 36(4): 926-935.

URL: 

http://www.rsta.ac.cn/EN/10.11873/j.issn.1004-0323.2021.4.0926     OR     http://www.rsta.ac.cn/EN/Y2021/V36/I4/926

Fig.1  Geographical location of the study area and UAV hyperspectral image
Fig.2  Diagram of sample spots and the photographs of grass species
Fig.3  Location and distribution of samples
Fig.4  The spectral curves of grass species
Fig.5  Flowchart of the research methods
Fig.6  The scatter diagram of number of shape adaptive neighborhood and MMSAD
Fig.7  The importance of different spectral bands
Fig.8  The overall identification accuracy and running time of different spectral bands
Fig.9  The grass species identification result of SMLR and ASR
类别

训练样本/

验证样本

SMLR-EMAPSMLR-EMAP-SAASR-EMAPASR-EMAP-SA
PAUAPAUAPAUAPAUA
矮火绒8/830.891 60.840 91.000 00.846 90.879 50.973 30.963 90.833 3
棘豆17/1630.975 50.649 01.000 00.795 11.000 00.610 51.000 00.871 7
苔藓18/1680.982 10.896 70.994 00.933 00.988 10.873 70.928 60.725 6
细叶亚菊7/690.579 70.975 60.768 10.688 30.724 60.781 30.811 60.965 5
苔草19/1820.978 00.983 40.923 11.000 00.989 00.923 10.972 51.000 0
藏嵩草11/1 1111.000 00.848 71.000 00.886 01.000 00.952 81.000 00.776 9
小嵩草3/390.615 41.000 00.666 71.000 00.512 81.000 00.384 60.882 4
针茅3/330.636 41.000 00.666 71.000 00.636 41.000 00.424 20.823 5
裸地15/1 6580.956 60.995 60.968 60.999 40.935 50.995 50.959 60.995 0
OA0.940 70.957 10.931 50.942 7
Kappa0.893 30.922 50.878 50.896 7
平均用时/s111152646692
Table 1  The grass species identification accuracy of SMLR and ASR
1 Liu Jiyuan, Shao Quanqin,Fan Jiangwen.The integrated assessment indicator system of grassland ecosystem in the Three-River Headwaters Region[J]. Geographical Research,2009,28(2):273-283.
1 刘纪远,邵全琴,樊江文.三江源区草地生态系统综合评估指标体系[J].地理研究,2009,28(2):273-283.
2 Liu Zhengjia, Shao Quanqin. Vegetation coverage change and its response to climate change in Three-River Headwaters Region[J]. Research of Soil and Water Conservation,2014, 21(6): 334-339.
2 刘正佳,邵全琴.三江源地区植被覆盖度变化及其与气候因子的关系[J].水土保持研究,2014,21(6):334-339.
3 Shen X, An R, Feng L,et al. Vegetation changes in the Three-River Headwaters Region of the Tibetan Plateau of China[J]. Ecological Indicators, 2018, 93: 804-812.
4 Zhang Y,Zhang C B,Wang Z Q,et al.Comprehensive research on remote sensing monitoring of grassland degradation: a case study in the Three-River Source Region,China[J].Sustainability,2019,11(7):1845.DOI:.
doi: 10.3390/su11071845
5 Wu Hong, An Ru, Li Xiaoxue,et al. Remote sensing monitoring of grassland degradation based on NPP change in the maduo county of the sources region of Yellow River[J]. Pratacultural Science,2011,28(4): 536-542.
5 吴红,安如,李晓雪,等.基于净初级生产力变化的草地退化监测研究[J].草业科学,2011,28(4):536-542.
6 An Ru, Sun Mengqiu, Lu Ling,et al.Analysis grassland degradation of the region of the sources of the Yangtze River, the Yellow River, and the Lancang River based on vegetation coverage estimated by TM imagery[J]. Science Technology and Engineering,2015,15(3): 11-18.
6 安如,孙梦秋,陆玲,等.三江源中东部地区1990~2008草地覆盖度变化遥感估算与分析[J].科学技术与工程,2015,15(3):11-18.
7 Xiao Tong, Wang Changzuo, Feng Min,et al.Dynamic characteristic of vegetation coverage in the Rhree-River Source Region from 2000 to 2011[J]. Acta Agrestia Sinica,2014, 22(1): 39-45.
7 肖桐,王昌佐,冯敏,等.2000~2011年青海三江源地区草地覆盖度的动态变化特征[J].草地学报,2014,22(1):39-45.
8 Yang Wencai, Wu Xinhong, Zhang Degang,et al.Evaluation of alpine grassland degradation of Three-River Source Area in Chengduo Country based on MODIS-NDVI[J]. Grassland and Turf,2011,31(5): 50-54.
8 杨文才,吴新宏,张德罡,等.基于MODIS-NDVI的三江源区称多县高寒草地退化现状评价[J].草原与草坪,2011,31(5):50-54.
9 Liu X F, Zhang J S, Zhu X F, et al. Spatiotemporal changes in vegetation coverage and its driving factors in the Three-River Headwaters Region during 2000~2011[J]. Journal of Geogra-phical Sciences, 2014, 24(2): 288-302.
10 Liu Jiyuan, Xu Xinliang, Shao Quanqin.Grassland degradation in the "Three-River Headwaters" Region, Qinghai province[J]. Acta Geographica Sinica,2008,63(4):364-376.
10 刘纪远,徐新良,邵全琴.近30年来青海三江源地区草地退化的时空特征[J].地理学报,2008,63(4):364-376.
11 Jiao Quanjun Zhang Bing, Zhao Jingjing,et al. Landscape pattern analysis of alpine steppe based on airborne Hyperspectral imagery in Maduo county, Qinghai province[J]. Acta Prataculturae Sinica,2012,21(2): 43-50.
11 焦全军,张兵,赵晶晶,等.基于航空高光谱影像的青海省玛多县高寒草原景观格局特征分析[J].草业学报,2012,21(2):43-50.
12 Li Shuang, Xu Xinliang, Fu Yin,et al. A study on classification of different degradation level alpine meadows based on Hyperspectral image data in Three-River Headwater Region[J]. Remote Sensing Technology and Application,2015,30(1):50-57.
12 李双,徐新良,付颖.基于高光谱影像的三江源区不同退化程度高寒草甸分类研究 [J].遥感技术与应用,2015,30(1):50-57.
13 Qu Ran, Li Shuang, Xu Xinliang,et al. Progress on methods of grassland degradation and weed invasion monitoring based on remote sensing[J].Journal of Geo-Information Science,2013,15(5): 761-767.
13 屈冉,李双,徐新良,等.草地退化杂类草入侵遥感监测方法研究进展[J].地球信息科学学报,2013,15(5):761-767.
14 Qian Yurong, Yang Feng, Li Jianlong,et al.Estimation of photosynthetic pigment of festuca arundinacea using Hyper-spectral data[J]. Acta Prataculturae Sinica,2009 18(4): 94-102.钱育蓉,杨峰,李建龙,等.利用高光谱数据快速估算高羊茅牧草光合色素的研究[J].草业学报,2009,18(4):94-102.
15 Yamano H,Chen J,Tamura M.Hyperspectral identification of grassland vegetation in silinhot, Inner Mongolia, China[J].International Journal of Remote Sensing,2003,24(15):3171-3178.
16 Hao Fangfang, Chen Yanmei, Gao Jixi,et al.Identification of Hyperspectra characteristic bands of grassland degradation indicator plant species in bashang region of Hebei province[J].Journal of Ecology and Rural Environment, 2016,32(6): 1024-1029.
16 郝芳芳,陈艳梅,高吉喜,等.河北坝上地区草地退化指示种的高光谱特征波段识别[J].生态与农村环境学报,2016,32(6):1024-1029.
17 An Ru, Jiang Danping, Li Xiaoxue,et al.Using hyperspectral data to determine spectral characteristics of grassland vegetation in central and eastern parts of Three-River Source[J]. Remote Sensing Technology and Application,2014,29(2): 202-211.
17 安如,姜丹萍,李晓雪,等.基于地面实测高光谱数据的三江源中东部草地植被光谱特征研究[J].遥感技术与应用,2014,29(2):202-211.
18 Wang Huanjiong, Fan Wenjie, Cui Yaokui,et al.Hyperspectral remote sensing monitoring of grassland degradation[J]. Spectroscopy and Spectral Analysis,2010,30(10): 2734-2738.
18 王焕炯,范闻捷,崔要奎,等.草地退化的高光谱遥感监测方法[J].光谱学与光谱分析,2010,30(10):2734-2738.
19 Li X H, Wang H, Li X B, et al. Identifying degraded grass species in Inner Mongolia based on measured Hyperspectral data[J]. Ieee Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2019, 12(12): 5061-5075.
20 Liao Xiaohan, Xiao Qing,Zhang Hao. UAV remote sensing: popularization and expand application development trend[J].Journal of Remote Sensing,2019,23(6): 1046-1052.
20 廖小罕,肖青,张颢.无人机遥感:大众化与拓展应用发展趋势[J].遥感学报,2019,23(6):1046-1052.
21 Wang Limin, Liu Jia, Yang Lingbo,et al.Applications of unmanned aerial vehicle images on agricultural remote sensing monitoring[J]. Transactions of the Chinese Society of Agricultural Engineering,2013,2013, 29(18):136-145.王利民,刘佳,杨玲波,等.基于无人机影像的农情遥感监测应用 [J].农业工程学报,2013,29(18):136-145.
22 Schmidt J,Fassnacht F E Ewald,Neff C,et al.Adapting a natura 2000 field guideline for a remote sensing-based assessment of heathland conservation status[J].International Journal of Applied Earth Observation and Geoinformation,2017,60:61-71. DOI:,04.005.
doi: 10.1016/j.jag.2017
23 Sun Shize, Wang Chuanjian, Yin Xiaojun,et al.Estimating above round biomass of natural grassland based on multispectral images of Unmanned Aerial Vehicles[J]. Journal of Remote Sensing,2018,22(5): 848-856,
23 孙世泽,汪传建,尹小君,等.无人机多光谱影像的天然草地生物量估算[J].遥感学报,2018,22(5):848-856.
24 Yang Hongyan, Du Jinmin, Wang Yuan,et al.Classification method of grassland species based on Unmanned Aerial Vehicle remote sensing and convolutional neural network[J]. Transactions of the Chinese Society for Agricultural Machinery,2019,50(4): 188-195.
24 杨红艳,杜健民,王圆,等.基于无人机遥感与卷积神经网络的草原物种分类方法[J].农业机械学报,2019,50(4):188-195.
25 Lu B, He Y H. Species classification using Unmanned Aerial Vehicle (UAV)-acquired high spatial resolution imagery in a heterogeneous grassland[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2017, 128: 73-85.
26 Du Peijun, Xia Junshi, Xue Zhaohui,et al.Review of Hyperspectral remote sensing image classification[J]. Journal of Remote Sensing,2016,20(2): 236-256
26 杜培军,夏俊士,薛朝辉,等.高光谱遥感影像分类研究进展[J].遥感学报,2016,20(2):236-256.
27 Jia X, Kuo B, Crawford MM. Feature mining for Hyperspectral image classification[J]. Proceedings of the IEEE, 2013,101(3):676-697.
28 Fauvel M,Tarabalka Y,Benediktsson J A,et al.Advances in spectral-spatial classification of Hyperspectral images[J].Proceedings of the IEEE,2013,101(3):652-675.
29 Reshma S,Veni S,George J E,et al.Hyperspectral crop classification using fusion of spectral, spatial features and vegetation indices:approach to the big data challenge[M]. International Conference on Advances in Computing, Communications and Informatics (ICACCI), IEEE, 2017:380-386.
30 Wei L,Yu M,Zhong Y,et al.Spatial-spectral fusion based on conditional random fields for the fine classification of crops in uav-borne Hyperspectral remote sensing imagery[J]. Remote Sensing,2019,11(7):780-799. DOI:.
doi: 10.3390/rs11070780
31 Shang Kun, Zhang Xia, Sun Yanli,et al. Sophisticated vegetation classification based on feature band set using Hyperspectral image[J]. Spectroscopy and Spectral Analysis,2015,35(6): 1669-1676.
32 Shi Feifei,Gao Xiaohong,Yang Lingyu,et al.Research on typical crop classification based on HJ-1A Hyperspectral data in the Huangshui River Basin[J]. Remote Sensing Technology and Application, 2017,32(2): 206-217.
32 史飞飞,高小红,杨灵玉,等.基于HJ-1A高光谱遥感数据的湟水流域典型农作物分类研究[J].遥感技术与应用,2017,32(2):206-217.
33 Zhang Miao,Jiang Zhirong,Ma Mingguo,et al.Fine classification of planting structure in the middle reaches of Heihe river basin based on Hyperspectral Compact Airborne Spectrographic Imager(CASI) data[J]. Remote Sensing Technology and Application, 2013,28(2): 283-289.
33 张苗,蒋志荣,马明国,等.基于CASI影像的黑河中游种植结构精细分类研究[J].遥感技术与应用,2013,28(2):283-289.
34 Cao J,Leng W,Liu K,et al.Object-based bangrove species classification using unmanned aerial vehicle Hyperspectral images and digital surface models[J].Remote Sensing,2018,10(1):89-108. DOI: .
doi: 10.3390/rs10010089
35 Dong Xinyu,Li Jiaguo,Chen Hanyue,et al. Extraction of individual tree information based on remote sensing images from an Unmanned Aerial Vehicle[J]. Journal of Remote Sensing,2019,23(6): 1269-1280.
35 董新宇,李家国,陈瀚阅,等.无人机遥感影像林地单株立木信息提取 [J].遥感学报,2019,23(6):1269-1280.
36 Gao Xia,Shen Xin,Dai Jinsong,et al. Tree species classification in urban forests based on LiDAR point cloud segmentation and Hyperspectral metrics extraction[J]. Remote Sensing Technology and Application, 2018,33(6):1073-1083.
36 皋厦,申鑫,代劲松,等.结合LiDAR单木分割和高光谱特征提取的城市森林树种分类 [J].遥感技术与应用,2018,33(6):1073-1083.
37 Li Xiaomei, Tan Bingxiang,LiZengyuan,et al.Comparation of forest types classification methods using CHRIS Hyperspectral image[J]. Remote Sensing Technology and Application, 2010,25(2):227-234.
37 李小梅,谭炳香,李增元,等.CHRIS高光谱图像森林类型分类方法比较研究[J].遥感技术与应用,2010,25(2):227-234.
38 Chen T Q,Guestrin C,Assoc Comp M.XGBoost: a scalable tree boosting system[C]∥ Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining,2016: 785-794.
39 Zhang A W, Dong Z,Kang X Y. Feature selection algorithms of airborne LiDAR combined with Hyperspectral images based on XGBoost[J]. Chinese Journal of Lasers, 2019,46(4): 150-158.
40 Yang B H,Qi L,Wang M X, et al.Cross-category tea polyphenols evaluation model based on feature fusion of electronic nose and Hyperspectral imagery[J]. Sensors,2020,20(1):50-67. DOI:.
doi: 10.3390/s20010050
41 Dalla Mura M,Benediktsson J A,Waske B,et al. Morphological attribute profiles for the analysis of very high resolution images[J].IEEE Transactions on Geoscience and Remote Sensing,2010,48(10):3747-3762.
42 Wright J, Yang A Y, Ganesh A, et al. Robust face recognition via sparse representation[J]. IEEE Transactions on Pattern Analysis And Machine Intelligence,2009,31(2):210-227.
43 Bioucas-Dias J M,IEEE. A variable splitting augmented lagrangian approach to linear spectral unmixing[M].First Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing, IEEE, 2009: 1-4.
44 Foi A,Katkovnik V,Egiazarian K.Pointwise shape-adaptive dct for high-quality denoising and deblocking of grayscale and color images[J].IEEE Transactions on Image Processing,2007,16(5):1395-1411.
No Suggested Reading articles found!