Abstract 【Objective】The shadows in UAV multi-spectral remote sensing images were removed to improve the accuracy of the nitrogen inversion model for apple canopy. 【Method】Using the UAV multi-spectral images collected in June 2019 at the apple orchard of Qixia city in Shandong province, as the experimental area, normalized shaded vegetation index (NSVI) and normalized canopy shadow index (NDCSI) were respectively used to remove shadow and to extract the spectral information of the canopy in non shadow area. The correlation analysis method was used to analyze the correlation between the spectral data, including the data obtained based on the original spectral images and the images after removing the shadow based on NSVI and NDCSI, and the measured leaf nitrogen content data, respectively, and then the sensitive wavelength of nitrogen content were screened and spectral parameters were constructed. Partial least squares (PLS) and support vector machine (SVM) methods were used to build the inversion model of nitrogen content and to carry out the precision inspection in the fruit tree canopy. 【Result】The results showed that the green band and red band were sensitive bands for the inversion of nitrogen content in fruit tree canopy based on UAV multi-spectral images. The spectral information of fruit tree canopy was weakened by shadow, and the spectral difference of canopy multispectral bands before and after shadow removal was significant, especially in red-edge band and near-infrared band. The accuracy of nitrogen inversion model based on two shadow indexes after shadow removal was improved, and the optimal model was the support vector machine nitrogen content inversion model based on NDCSI, the modeling set of this model R2 and RPD was 0.774 and 1.828, the validation set R2 and RPD were 0.723 and 1.819 respectively. 【Conclusion】NDCSI could effectively remove the shadow in the multi-spectral fruit tree canopy image of the UAV to improve remote sensing inversion accuracy of nitrogen content in apple canopy, so as to provide a useful reference for precise nitrogen management in orchard. Keywords:canopy shadow;shadow vegetation index;UAV;multispectral;remote sensing
PDF (1373KB)元数据多维度评价相关文章导出EndNote|Ris|Bibtex收藏本文 本文引用格式 李美炫, 朱西存, 白雪源, 彭玉凤, 田中宇, 姜远茂. 基于无人机影像阴影去除的苹果树冠层氮素含量遥感反演[J]. 中国农业科学, 2021, 54(10): 2084-2094 doi:10.3864/j.issn.0578-1752.2021.10.005 LI MeiXuan, ZHU XiCun, BAI XueYuan, PENG YuFeng, TIAN ZhongYu, JIANG YuanMao. Remote Sensing Inversion of Nitrogen Content in Apple Canopy Based on Shadow Removal in UAV Multi-Spectral Remote Sensing Images[J]. Scientia Acricultura Sinica, 2021, 54(10): 2084-2094 doi:10.3864/j.issn.0578-1752.2021.10.005
Table 1 表1 表1多光谱传感器的波段参数 Table 1Band parameters of multispectral sensor
波段 Band
中心波长 Band center (nm)
带宽 Band width (nm)
绿光(Bg)
550
40
红光(Br)
660
40
红边(Breg)
735
10
近红外(Bnir)
790
40
Bg为绿光波段,Br为红光波段,Breg为红边波段,Bnir为近红外波段。下同 Bg is the green band, Br is the red band, Breg is the red edge band, and Bnir is the near infrared band. The same below
TIANG, LI HF, TIANM, LIU XX, CHENQ, ZHU ZL, JIANG YM, GE SF. Effects of different integration of water and fertilizer modes on the absorption and utilization of nitrogen fertilizer and fruit yield and quality of apple trees Chinese Journal of Applied Ecology, 2020,31(6):1867-1874. (in Chinese) [本文引用: 1]
WANGL, ZHAO GX, ZHU XC, WANG RY, CHANG CY. Satellite remote sensing retrieval of canopy nitrogen nutritional status of apple trees at blossom stage Chinese Journal of Applied Ecology, 2013,24(10):2863-2870. (in Chinese) [本文引用: 1]
ZHAOJ, LI ZM, LU LQ, JIAP, YANG HB, LAN YB. Weed identification in maize field based on multi-spectral remote sensing of unmanned aerial vehicle Scientia Agricultura Sinica, 2020,53(8):1545-1555. (in Chinese) [本文引用: 1]
LIU CH, WANGZ, CHEN ZC, ZHOUL, YUE XZ, MIAO YX. Nitrogen monitoring of winter wheat based on unmanned aerial vehicle remote sensing image Transactions of the Chinese Society for Agricultural Machinery, 2018,49(6):207-214. (in Chinese) [本文引用: 1]
YUE JB, YANG GJ, LI CC, LI ZH, WANG YJ, FENG HK, XUB. Estimation of winter wheat above-ground biomass using unmanned aerial vehicle-based snapshot hyperspectral sensor and crop height improved models , 2017,9(7):708-726. DOI:10.3390/rs9070708URL [本文引用: 1]
CINATP, DI GENNARO SF, BERTONA, MATESEA. Comparison of unsupervised algorithms for vineyard canopy segmentation from UAV multispectral images , 2019,11(9):1023-1046. DOI:10.3390/rs11091023URL [本文引用: 1]
SHIY, GENGN, HU SJ, ZHANG ZY, ZHANGJ. Illumination distribution model of apple tree canopy based on random forest regression algorithm Transactions of the Chinese Society for Agricultural Machinery, 2019,50(5):214-222. (in Chinese) [本文引用: 1]
ZHOUK, DENG XQ, YAOX, TIAN YC, CAO WX, ZHUY, USTIN SL, CHENGT. Assessing the spectral properties of sunlit and shaded components in rice canopies with near-ground imaging spectroscopy data , 2017,17(3):578-594. DOI:10.3390/s17030578URL [本文引用: 1]
GUO JH, TIAN QJ, WU JZ. Study on multispectral detecting shadow areas and a theoretical model of removing shadows from remote sensing images National Remote Sensing Bulletin, 2006(2):151-159. (in Chinese) [本文引用: 1]
PU RL, LANDRYS. A comparative analysis of high spatial resolution IKONOS and WorldView-2 imagery for mapping urban tree species , 2012,124:516-533. DOI:10.1016/j.rse.2012.06.011URL [本文引用: 1]
TITSL, SOMERSB, STUCKENSJ, FARIFTEHJ, COPPINP. Integration of in situ measured soil status and remotely sensed hyperspectral data to improve plant Production system monitoring: concept, perspectives and limitations , 2013,128:197-211. DOI:10.1016/j.rse.2012.10.006URL [本文引用: 1]
XI ZY, LIU LJ, LU DS, GE HL, CHEN YL. Mapping of carya cathayensis spatial distribution with linear spectral mixture model Scientia Silvae Sinicae, 2015,51(10):43-52. (in Chinese) [本文引用: 1]
MILAS AS, ARENDK, MAYERC, SIMONSON MA, MACKEYM. Different colours of shadows: Classification of UAV images , 2017,38(8/10):3084-3100. DOI:10.1080/01431161.2016.1274449URL [本文引用: 1]
JIAO JN, SHIJ, TIAN QJ, GAOL, XU NX. Research on multispectral-image-based NDVI shadow-effect-eliminating model National Remote Sensing Bulletin, 2020,24(1):53-66. (in Chinese) [本文引用: 1]
ZHANG LF, SUN XJ, WU TX, ZHANG HM. An analysis of shadow effects on spectral vegetation indexes using a ground-based imaging spectrometer , 2015,12(11):2188-2192. DOI:10.1109/LGRS.2015.2450218URL [本文引用: 1]
江洪, 袁亚伟, 王森. 阴影消除植被指数(SEVI)去除地形本影和落影干扰的性能评估与应用 , 2019,21(12):1977-1986. DOI:10.12082/dqxxkx.2019.190409 [本文引用: 1] 地形校正是崎岖山区遥感图像预处理的关键步骤。为了评估基于DEM数据的经验校正模型、山地辐射传输模型和波段组合优化计算模型在去除地形阴影效应方面的性能,并将其应用于福州市植被覆盖监测,本文采用C模型(和SCS+C模型)、6S+C模型和阴影消除植被指数(SEVI)进行评估、比较。采用1999年和2014年两期Landsat 5 TM卫星数据和相关的 30 m ASTER GDEM V2高程数据,分别计算了C校正(和SCS+C校正)和6S+C校正后的归一化植被指数(NDVI)和比值植被指数(RVI)以及基于表观反射率数据的SEVI。通过目视比较、光谱特征比较以及太阳入射角余弦值(cos i)与植被指数的线性回归分析,可以看出C模型和SCS+C模型对本影具有较好的校正效果,但对落影的校正效果欠佳。NDVI和RVI的本影与邻近无阴影阳坡的相对误差分别从71.64%、52.57%降至4.80%、6.43%(C模型)和0.50%、9.94%(SCS + C模型),而落影与邻近无阴影阳坡的相对误差分别从62.01%、47.57%降至31.05%、24.40%(C模型)和33.42%、16.01%(SCS + C模型)。在NDVI的落影校正效果上,6S+C模型比C模型和SCS+C模型有一定的提升,本影与邻近无阴影阳坡之间的相对误差为8.63%,落影与邻近无阴影阳坡之间的相对误差为14.27%。而SEVI在消除本影和落影方面整体效果更好,本影和落影与邻近无阴影阳坡的相对误差分别为9.86%和10.53%。最后,基于SEVI对福州市1999-2014年的植被覆盖变化进行了监测。监测结果表明: ① 1999-2014年植被覆盖增加了893.61 km <sup>2</sup>,植被增加区域主要分布在海拔250~1250 m范围内;② SEVI均值在坡度40°附近达到峰值。 JIANGH, YUAN YW, WANGS. Shadow-eliminated vegetation index (SEVI) for removing terrain shadow effect: Evaluation and application . Journal of Geo-information Science, 2019,21(12):1977-1986. (in Chinese) DOI:10.12082/dqxxkx.2019.190409 [本文引用: 1] 地形校正是崎岖山区遥感图像预处理的关键步骤。为了评估基于DEM数据的经验校正模型、山地辐射传输模型和波段组合优化计算模型在去除地形阴影效应方面的性能,并将其应用于福州市植被覆盖监测,本文采用C模型(和SCS+C模型)、6S+C模型和阴影消除植被指数(SEVI)进行评估、比较。采用1999年和2014年两期Landsat 5 TM卫星数据和相关的 30 m ASTER GDEM V2高程数据,分别计算了C校正(和SCS+C校正)和6S+C校正后的归一化植被指数(NDVI)和比值植被指数(RVI)以及基于表观反射率数据的SEVI。通过目视比较、光谱特征比较以及太阳入射角余弦值(cos i)与植被指数的线性回归分析,可以看出C模型和SCS+C模型对本影具有较好的校正效果,但对落影的校正效果欠佳。NDVI和RVI的本影与邻近无阴影阳坡的相对误差分别从71.64%、52.57%降至4.80%、6.43%(C模型)和0.50%、9.94%(SCS + C模型),而落影与邻近无阴影阳坡的相对误差分别从62.01%、47.57%降至31.05%、24.40%(C模型)和33.42%、16.01%(SCS + C模型)。在NDVI的落影校正效果上,6S+C模型比C模型和SCS+C模型有一定的提升,本影与邻近无阴影阳坡之间的相对误差为8.63%,落影与邻近无阴影阳坡之间的相对误差为14.27%。而SEVI在消除本影和落影方面整体效果更好,本影和落影与邻近无阴影阳坡的相对误差分别为9.86%和10.53%。最后,基于SEVI对福州市1999-2014年的植被覆盖变化进行了监测。监测结果表明: ① 1999-2014年植被覆盖增加了893.61 km <sup>2</sup>,植被增加区域主要分布在海拔250~1250 m范围内;② SEVI均值在坡度40°附近达到峰值。
LIU XN, JIANGH, WANG XQ. Extraction of mountain vegetation information based on vegetation distinguished and shadow eliminated vegetation index Transactions of the Chinese Society of Agricultural Engineering, 2019,35(20):135-144. (in Chinese) [本文引用: 1]
XU ZH, LIUJ, YU KY, LIUT, GONG CH, TANG MY, XIE WJ, LI ZL. Construction of vegetation shadow index (SVI) and application effects in four remote sensing images Spectroscopy and Spectral Analysis, 2013,33(12):3359-3365. (in Chinese) [本文引用: 1]
XU ZH, LINL, WANG QF, HUANG XY, LIUJ, XU KY, CHEN CC. Construction and application effects of normalized shaded vegetation index Journal of Infrared and Millimeter Waves, 2018,37(2):154-162. (in Chinese) [本文引用: 2]
XU NX, TIANJ, TIAN QJ, XU KJ, TANG SF. Analysis of vegetation red edge with different illuminated/shaded canopy proportions and to construct normalized difference canopy shadow index , 2019,11(10):1192-1207. DOI:10.3390/rs11101192URL [本文引用: 4]
YANG GJ, WANP, YU HY, XUB, FENG HK. Automatic radiation uniformity correction of multispectral imagery acquired with unmanned aerial vehicle Transactions of the Chinese Society of Agricultural Engineering, 2015,31(9):147-153. (in Chinese) [本文引用: 1]
PUZ, YANGL, BAIJ. Shadow detection and removal based on object-oriented method in high spatial resolution remote sense image Remote Sensing Technology and Application, 2008,23(6):735-738. (in Chinese) [本文引用: 1]
JIANG ZY, HUETE AR, CHENJ, CHEN YH, LIJ, YAN GJ, ZHANG XY. Analysis of NDVI and scaled difference vegetation index retrievals of vegetation fraction , 2006,101(3):366-378. DOI:10.1016/j.rse.2006.01.003URL [本文引用: 1]
CHEN HY, ZHAO GX, CHEN JC, WANG RY, GAO MX. Remote sensing inversion of saline soil salinity based on modified vegetation index in estuary area of Yellow River Transactions of the Chinese Society of Agricultural Engineering, 2015,31(5):107-114. (in Chinese) [本文引用: 1]
FUP, MEACHAM-HENSOLDK, GUAN KY, WUJ, BERNACCHIC. Estimating photosynthetic traits from reflectance spectra: A synthesis of spectral indices, numerical inversion, and partial least square regression , 2020,43(5):1241-1258. [本文引用: 1]
ALI AM, DARVISHZADEHR, SKIDMOREA, GARA TW, O’CONNORB, ROEOESLIC, HEURICHM, PAGANINIM. Comparing methods for mapping canopy chlorophyll content in a mixed mountain forest using Sentinel-2 data , 2020,87:14. [本文引用: 1]
ZHANGZ, LONG HL, WANG CC, YANG GJ. Comparison of hyperspectral remote sensing estimation models based on photosynthetic characteristics of winter wheat leaves Scientia Agricultura Sinica, 2019,52(4):616-628. (in Chinese) [本文引用: 1]
SHIZ, WANG QL, PENGJ, JI WJ, LIU HJ, LIX, RAPHAEL A VR. Development of a national VNIR soil-spectral library for soil classification and prediction of organic matter concentrations , 2014,57(7):1671-1680. [本文引用: 1]
YAN GJ, LI LY, COYA, MU XH, CHEN SB, XIE DH, ZHANG WM, SHEN QF, ZHOU HM. Improving the estimation of fractional vegetation cover from UAV RGB imagery by colour unmixing , 2019,158:23-34. DOI:10.1016/j.isprsjprs.2019.09.017URL [本文引用: 1]
ABOUTALEBIM, TORRES-RUA AF, KUSTAS WP, NIETOH, COOPMANSC, MCKEEM. Assessment of different methods for shadow detection in high-resolution optical imagery and evaluation of shadow impact on calculation of NDVI, and evapotranspiration , 2019,37(3):407-429. DOI:10.1007/s00271-018-0613-9URL [本文引用: 1]
CHEN PF, LIANGF. Cotton nitrogen nutrition diagnosis based on spectrum and texture feature of images from low altitude unmanned aerial vehicle Scientia Agricultura Sinica, 2019,52(13):2220-2229. (in Chinese) [本文引用: 1]
ZHOUK, GUO YJ, GENG YN, ZHUY, CAO WX, TIAN YC. Development of a novel bidirectional canopy reflectance model for row-planted rice and wheat , 2014,6(8):7632-7659. DOI:10.3390/rs6087632URL [本文引用: 1]
ZENGQ, XU KY, YAOX, ZHENG WY, ZHANG JZ, AI JW, LIUJ. Research on canopy reflectance modeling of Phyllostachys pubescens forest based on the PROSAIL canopy radiative transfer model Plant Science Journal, 2017,35(5):699-707. (in Chinese) [本文引用: 1]