scholarly journals Estimation of Leaf Nitrogen Content in Wheat Using New Hyperspectral Indices and a Random Forest Regression Algorithm

2018 ◽  
Vol 10 (12) ◽  
pp. 1940 ◽  
Author(s):  
Liang Liang ◽  
Liping Di ◽  
Ting Huang ◽  
Jiahui Wang ◽  
Li Lin ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Random Forest ◽  
Nitrogen Content ◽  
Vegetation Index ◽  
Leaf Nitrogen ◽  
Support Vector ◽  
Leaf Nitrogen Content ◽  
Random Forest Regression ◽  
First Derivative ◽  
Remote Sensing Mapping

Novel hyperspectral indices, which are the first derivative normalized difference nitrogen index (FD-NDNI) and the first derivative ratio nitrogen vegetation index (FD-SRNI), were developed to estimate the leaf nitrogen content (LNC) of wheat. The field stress experiments were conducted with different nitrogen and water application rates across the growing season of wheat and 190 measurements were collected on canopy spectra and LNC under various treatments. The inversion models were constructed based on the dataset to evaluate the ability of various spectral indices to estimate LNC. A comparative analysis showed that the model accuracies of FD-NDNI and FD-SRNI were higher than those of other commonly used hyperspectral indices including mNDVI705, mSR, and NDVI705, which was indicated by higher R2 and lower root mean square error (RMSE) values. The least squares support vector regression (LS-SVR) and random forest regression (RFR) algorithms were then used to optimize the models constructed by FD-NDNI and FD-SRNI. The p-R2 values of the FD-NDNI_RFR and FD-SRNI_RFR models reached 0.874 and 0.872, respectively, which were higher than those of the exponential and SVR model and indicated that the RFR model was accurate. Using the RFR inversion model, remote sensing mapping for the Operative Modular Imaging Spectrometer (OMIS) image was accomplished. The remote sensing mapping of the OMIS image yielded an accuracy of R2 = 0.721 and RMSE = 0.540 for FD-NDNI and R2 = 0.720 and RMSE = 0.495 for FD-SRNI, which indicates that the similarity between the inversion value and the measured value was high. The results show that the new hyperspectral indices, i.e., FD-NDNI and FD-SRNI, are the optimal hyperspectral indices for estimating LNC and that the RFR algorithm is the preferred modeling method.

scholarly journals Nitrogen Content Estimation of Apple Leaves Using Hyperspectral Analysis

2021 ◽  
Vol 2021 ◽  
pp. 1-9
Author(s):  
Weihao Sun ◽  
Dongwei Wang ◽  
Ning Jin ◽  
Shusheng Xu ◽  
Haoran Bai
Keyword(s):  
Nitrogen Content ◽  
Leaf Nitrogen ◽  
Differential Spectrum ◽  
Gradient Boosting ◽  
Support Vector ◽  
Model Parameters ◽  
Leaf Nitrogen Content ◽  
Inversion Model ◽  
Apple Leaves ◽  
Characteristic Wavelength

Leaf nitrogen content (LNC) is an important factor reflecting the growth quality of plants. We estimated the nitrogen content of apple leaves using hyperspectral wavelength analysis using the differential spectrum, differential spectrum transformation, and vegetation spectrum index with different derivative gaps. We then used the characteristic wavelengths extracted via the correlation coefficient method as the input vectors to the gradient boosting decision tree (GBDT) model for analysis and performed cross-validation to optimize the inversion model parameters. We analyzed the results with different input variables and loss functions and compared the GBDT model with other mainstream algorithm models. The results show that the R2 value of the optimized GBDT inversion model is higher than that obtained using the random forest (RF) and support vector regression (SVR) models. Thus, the GBDT model is accurate, and the characteristic wavelength analysis is helpful for the tasks of real-time monitoring and detection of apple tree health.

Predicting Winter Wheat Grain Quality Using Hyperspectral Data Based on Plant Nitrogen Status

2012 ◽  
Vol 524-527 ◽  
pp. 2132-2138 ◽  
Author(s):  
Hui Fang Wang ◽  
Ji Hua Wang ◽  
Mei Chen Feng ◽  
Qian Wang ◽  
Wen Jiang Huang ◽  
...  
Keyword(s):  
Winter Wheat ◽  
Nitrogen Content ◽  
Vegetation Index ◽  
Grain Quality ◽  
Leaf Nitrogen ◽  
Hyperspectral Data ◽  
Leaf Nitrogen Content ◽  
Wheat Grain ◽  
Quality Model ◽  
Plant Nitrogen

Quality of winter wheat from hyperspectral data would provide opportunities to manage grain harvest differently, and to maximize output by adjusting input in fields. In this study, two varieties winter wheat as the object, hyperspectral data were utilized to predict grain quality. Firstly, the leaf and stem nitrogen content at winter wheat anthesis stage was proved to be signification correctly with crude content and wet gluten. And the leaf relatedcoefficient more than stem at the anthesis. Then, spectral indices significantly correlated to plant nitrogen content at anthesis stage were potential indicators for grain qualities. The vegetation index, VI derived from the canopy spectral reflectance was signification correlated to the leaf nitrogen content at anthesis stage, and highly significantly correlated to the leaf nitrogen content. Based on above analysis, the predict grain quality model were build and the related coefficient were 0.86, 0.68, 0.84, 0.58 which were reached a very significant.The result demonstrated the model based on SIPI and RVI to predict different cultivars wheat grain quality were practical and feasible.

scholarly journals Remote sensing of LAI, chlorophyll and leaf nitrogen pools of crop- and grasslands in five European landscapes

2012 ◽  
Vol 9 (8) ◽  
pp. 10149-10205 ◽  
Author(s):  
E. Boegh ◽  
R. Houborg ◽  
J. Bienkowski ◽  
C. F. Braban ◽  
T. Dalgaard ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Land Use ◽  
Nitrogen Content ◽  
Field Data ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Leaf Nitrogen ◽  
Nitrogen Pools ◽  
Canopy Nitrogen ◽  
Use Categories

Abstract. Leaf nitrogen and leaf surface area influence the exchange of gases between terrestrial ecosystems and the atmosphere, and they play a significant role in the global cycles of carbon, nitrogen and water. Remote sensing data from satellites can be used to estimate leaf area index (LAI), leaf chlorophyll (CHLl) and leaf nitrogen density (Nl). However, methods are often developed using plot scale data and not verified over extended regions that represent a variety of soil spectral properties and canopy structures. In this paper, field measurements and high spatial resolution (10–20 m) remote sensing images acquired from the HRG and HRVIR sensors aboard the SPOT satellites were used to assess the predictability of LAI, CHLl and Nl. Five spectral vegetation indices (SVIs) were used (the Normalized Difference Vegetation index, the Simple Ratio, the Enhanced Vegetation Index-2, the Green Normalized Difference Vegetation Index, and the green Chlorophyll Index) together with the image-based inverse canopy radiative transfer modelling system, REGFLEC (REGularized canopy reFLECtance). While the SVIs require field data for empirical model building, REGFLEC can be applied without calibration. Field data measured in 93 fields within crop- and grasslands of five European landscapes showed strong vertical CHLl gradient profiles in 20% of fields. This affected the predictability of SVIs and REGFLEC. However, selecting only homogeneous canopies with uniform CHLl distributions as reference data for statistical evaluation, significant (p < 0.05) predictions were achieved for all landscapes, by all methods. The best performance was achieved by REGFLEC for LAI (r2=0.7; rmse = 0.73), canopy chlorophyll content (r2=0.51; rmse = 439 mg m−2) and canopy nitrogen content (r2 = 0.53; rmse = 2.21 g m−2). Predictabilities of SVIs and REGFLEC simulations generally improved when constrained to single land use categories (wheat, maize, barley, grass) across the European landscapes, reflecting sensitivity to canopy structures. Predictability further improved when constrained to local (10 × 10 km2) landscapes, thereby reflecting sensitivity to local environmental conditions. All methods showed different predictabilities for land use categories and landscapes. Combining the best methods, LAI, canopy chlorophyll content (CHLc) and canopy nitrogen content (CHLc) for the five landscapes could be predicted with improved accuracy (LAI rmse = 0.59; CHLc rmse = 346 g m−2; Ncrmse = 1.49 g m−2). Remote sensing-based results showed that the vegetation nitrogen pools of the five agricultural landscapes varied from 0.6 to 4.0 t km−2. Differences in nitrogen pools were attributed to seasonal variations, extents of agricultural area, species variations, and spatial variations in nutrient availability. Information on Nl and total Nc pools within the landscapes is important for the spatial evaluation of nitrogen and carbon cycling processes. The upcoming Sentinel-2 satellite mission will provide new multiple narrow-band data opportunities at high spatio-temporal resolution which are expected to further improve remote sensing predictabilities of LAI, CHLl and Nl.

scholarly journals WAVELENGTH SELECTION OF HYPERSPECTRAL LIDAR BASED ON FEATURE WEIGHTING FOR ESTIMATION OF LEAF NITROGEN CONTENT IN RICE

2016 ◽  
Vol XLI-B1 ◽  
pp. 9-13
Author(s):  
Lin Du ◽  
Shuo Shi ◽  
Wei Gong ◽  
Jian Yang ◽  
Jia Sun ◽  
...  
Keyword(s):  
Nitrogen Content ◽  
System Integration ◽  
Wide Spectrum ◽  
Green Plant ◽  
Leaf Nitrogen ◽  
Feature Weighting ◽  
Support Vector ◽  
Leaf Nitrogen Content ◽  
Wavelength Selection ◽  
Regression Methods

Hyperspectral LiDAR (HSL) is a novel tool in the field of active remote sensing, which has been widely used in many domains because of its advantageous ability of spectrum-gained. Especially in the precise monitoring of nitrogen in green plants, the HSL plays a dispensable role. The exiting HSL system used for nitrogen status monitoring has a multi-channel detector, which can improve the spectral resolution and receiving range, but maybe result in data redundancy, difficulty in system integration and high cost as well. Thus, it is necessary and urgent to pick out the nitrogen-sensitive feature wavelengths among the spectral range. The present study, aiming at solving this problem, assigns a feature weighting to each centre wavelength of HSL system by using matrix coefficient analysis and divergence threshold. The feature weighting is a criterion to amend the centre wavelength of the detector to accommodate different purpose, especially the estimation of leaf nitrogen content (LNC) in rice. By this way, the wavelengths high-correlated to the LNC can be ranked in a descending order, which are used to estimate rice LNC sequentially. In this paper, a HSL system which works based on a wide spectrum emission and a 32-channel detector is conducted to collect the reflectance spectra of rice leaf. These spectra collected by HSL cover a range of 538 nm – 910 nm with a resolution of 12 nm. These 32 wavelengths are strong absorbed by chlorophyll in green plant among this range. The relationship between the rice LNC and reflectance-based spectra is modeled using partial least squares (PLS) and support vector machines (SVMs) based on calibration and validation datasets respectively. The results indicate that I) wavelength selection method of HSL based on feature weighting is effective to choose the nitrogen-sensitive wavelengths, which can also be co-adapted with the hardware of HSL system friendly. II) The chosen wavelength has a high correlation with rice LNC which can be retrieved by using PLS and SVMs regression methods.

scholarly journals Study on the Quantitative Relationship Among Canopy Hyperspectral Reflectance, Vegetation Index and Cotton Leaf Nitrogen Content

2021 ◽  
Author(s):  
Caixia Yin ◽  
Jiao Lin ◽  
Lulu Ma ◽  
Ze Zhang ◽  
Tongyu Hou ◽  
...  
Keyword(s):  
Nitrogen Content ◽  
Nitrogen Fertilizer ◽  
Spectral Reflectance ◽  
Vegetation Index ◽  
Early Stage ◽  
Leaf Nitrogen ◽  
Nitrogen Fertilizers ◽  
Cotton Leaf ◽  
Leaf Nitrogen Content ◽  
Center Wavelength

AbstractStudy the response mechanism of Canopy spectral reflectance (CSR) to cotton nitrogen fertilizer, propose the sensitive band and center wavelength of cotton leaf nitrogen content (LNC), and compare the response characteristics of various vegetation indexes to LNC, propose a vegetation index that responds well to LNC and construct estimating model. This experiment sets five nitrogen fertilizer levels, namely N0 (control), N120 (120 kg/hm2), N240 (240 kg/hm2), N360 (360 kg/hm2), N480 (480 kg/hm2). Among them, referring to the conventional nitrogen fertilizer is applied by local farmers (N330, 330 kg/hm2). The results showed the following: (1) Visible light and near-infrared (NIR) can be used as two large ranges for precise monitoring of nitrogen, especially the CSR in the NIR range differs significantly under different nitrogen fertilizers. In the early stage of cotton growth, the CSR decreased with the nitrogen application rate increase, in a suitable nitrogen environment (360 kg/hm2), and beyond N360, vice versa. In the later growth period, the CSR increases with the increase in nitrogen fertilizer. This trend is most evident in the short-wave NIR regions;(2) the range of 690–709 nm, 717–753 nm, and 940–958, which can be remote sensed by the spectral reflectance when cotton is affected in poor or rich nitrogen. The center wavelength corresponding to the nitrogen-sensitive band, respectively, are 697 nm, 735 nm, 953 nm, the band width can maintain 5–15 nm, generally not more than 20 nm;(3) compared with the ratio vegetation index, difference vegetation index, and normalized vegetation index, the combined vegetation index of more than two bands has a better effect on cotton LNC monitoring, of which the index (R560−R670)/(R560 + R670−R450), (R700−1.7 × R670 + 0.7 × R450)/(R700 + 2.3 × R670−1.3 × R450) are significantly related to LNC in this papers, and the correlation coefficients can reach, respectively, 0.935* and 0.936*. These findings help to estimate the model of LNC. The model is as follows: Y = 19.883 × x + 42.285, where x refers to the combined vegetation index (R700−1.7 × R670 + 0.7 × R450)/(R700 + 2.3 × R670−1.3 × R450), Y is LNC, but the model accuracy will be affected in the crop different phenological stage, and the model has the highest monitoring accuracy during the bud period.

scholarly journals Corn Nitrogen Nutrition Index Prediction Improved by Integrating Genetic, Environmental, and Management Factors with Active Canopy Sensing Using Machine Learning

2022 ◽  
Vol 14 (2) ◽  
pp. 394
Author(s):  
Dan Li ◽  
Yuxin Miao ◽  
Curtis J. Ransom ◽  
G. Mac Bean ◽  
Newell R. Kitchen ◽  
...  
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Vegetation Index ◽  
Sensor Data ◽  
Support Vector ◽  
Random Forest Regression ◽  
Machine Learning Methods ◽  
Developmental Growth ◽  
N Status ◽  
Better Than

Accurate nitrogen (N) diagnosis early in the growing season across diverse soil, weather, and management conditions is challenging. Strategies using multi-source data are hypothesized to perform significantly better than approaches using crop sensing information alone. The objective of this study was to evaluate, across diverse environments, the potential for integrating genetic (e.g., comparative relative maturity and growing degree units to key developmental growth stages), environmental (e.g., soil and weather), and management (e.g., seeding rate, irrigation, previous crop, and preplant N rate) information with active canopy sensor data for improved corn N nutrition index (NNI) prediction using machine learning methods. Thirteen site-year corn (Zea mays L.) N rate experiments involving eight N treatments conducted in four US Midwest states in 2015 and 2016 were used for this study. A proximal RapidSCAN CS-45 active canopy sensor was used to collect corn canopy reflectance data around the V9 developmental growth stage. The utility of vegetation indices and ancillary data for predicting corn aboveground biomass, plant N concentration, plant N uptake, and NNI was evaluated using singular variable regression and machine learning methods. The results indicated that when the genetic, environmental, and management data were used together with the active canopy sensor data, corn N status indicators could be more reliably predicted either using support vector regression (R2 = 0.74–0.90 for prediction) or random forest regression models (R2 = 0.84–0.93 for prediction), as compared with using the best-performing single vegetation index or using a normalized difference vegetation index (NDVI) and normalized difference red edge (NDRE) together (R2 < 0.30). The N diagnostic accuracy based on the NNI was 87% using the data fusion approach with random forest regression (kappa statistic = 0.75), which was better than the result of a support vector regression model using the same inputs. The NDRE index was consistently ranked as the most important variable for predicting all the four corn N status indicators, followed by the preplant N rate. It is concluded that incorporating genetic, environmental, and management information with canopy sensing data can significantly improve in-season corn N status prediction and diagnosis across diverse soil and weather conditions.

scholarly journals WAVELENGTH SELECTION OF HYPERSPECTRAL LIDAR BASED ON FEATURE WEIGHTING FOR ESTIMATION OF LEAF NITROGEN CONTENT IN RICE

2016 ◽  
Vol XLI-B1 ◽  
pp. 9-13
Author(s):  
Lin Du ◽  
Shuo Shi ◽  
Wei Gong ◽  
Jian Yang ◽  
Jia Sun ◽  
...  
Keyword(s):  
Nitrogen Content ◽  
System Integration ◽  
Wide Spectrum ◽  
Green Plant ◽  
Leaf Nitrogen ◽  
Feature Weighting ◽  
Support Vector ◽  
Leaf Nitrogen Content ◽  
Wavelength Selection ◽  
Regression Methods

Hyperspectral LiDAR (HSL) is a novel tool in the field of active remote sensing, which has been widely used in many domains because of its advantageous ability of spectrum-gained. Especially in the precise monitoring of nitrogen in green plants, the HSL plays a dispensable role. The exiting HSL system used for nitrogen status monitoring has a multi-channel detector, which can improve the spectral resolution and receiving range, but maybe result in data redundancy, difficulty in system integration and high cost as well. Thus, it is necessary and urgent to pick out the nitrogen-sensitive feature wavelengths among the spectral range. The present study, aiming at solving this problem, assigns a feature weighting to each centre wavelength of HSL system by using matrix coefficient analysis and divergence threshold. The feature weighting is a criterion to amend the centre wavelength of the detector to accommodate different purpose, especially the estimation of leaf nitrogen content (LNC) in rice. By this way, the wavelengths high-correlated to the LNC can be ranked in a descending order, which are used to estimate rice LNC sequentially. In this paper, a HSL system which works based on a wide spectrum emission and a 32-channel detector is conducted to collect the reflectance spectra of rice leaf. These spectra collected by HSL cover a range of 538 nm – 910 nm with a resolution of 12 nm. These 32 wavelengths are strong absorbed by chlorophyll in green plant among this range. The relationship between the rice LNC and reflectance-based spectra is modeled using partial least squares (PLS) and support vector machines (SVMs) based on calibration and validation datasets respectively. The results indicate that I) wavelength selection method of HSL based on feature weighting is effective to choose the nitrogen-sensitive wavelengths, which can also be co-adapted with the hardware of HSL system friendly. II) The chosen wavelength has a high correlation with rice LNC which can be retrieved by using PLS and SVMs regression methods.

scholarly journals Estimation of Leaf Nitrogen Content in Wheat Based on Fusion of Spectral Features and Deep Features from Near Infrared Hyperspectral Imagery

Sensors ◽  
2021 ◽  
Vol 21 (2) ◽  
pp. 613
Author(s):  
Baohua Yang ◽  
Jifeng Ma ◽  
Xia Yao ◽  
Weixing Cao ◽  
Yan Zhu
Keyword(s):  
Nitrogen Content ◽  
Hyperspectral Imagery ◽  
Leaf Nitrogen ◽  
Support Vector ◽  
Growth Monitoring ◽  
Leaf Nitrogen Content ◽  
Spectral Features ◽  
Wheat Growth ◽  
Important Indicator ◽  
Model Based

Nitrogen is an important indicator for monitoring wheat growth. The rapid development and wide application of non-destructive detection provide many approaches for estimating leaf nitrogen content (LNC) in wheat. Previous studies have shown that better results have been obtained in the estimation of LNC in wheat based on spectral features. However, the lack of automatically extracted features leads to poor universality of the estimation model. Therefore, a feature fusion method for estimating LNC in wheat by combining spectral features with deep features (spatial features) was proposed. The deep features were automatically obtained with a convolutional neural network model based on the PyTorch framework. The spectral features were obtained using spectral information including position features (PFs) and vegetation indices (VIs). Different models based on feature combination for evaluating LNC in wheat were constructed: partial least squares regression (PLS), gradient boosting decision tree (GBDT), and support vector regression (SVR). The results indicate that the model based on the fusion feature from near-ground hyperspectral imagery has good estimation effect. In particular, the estimation accuracy of the GBDT model is the best (R2 = 0.975 for calibration set, R2 = 0.861 for validation set). These findings demonstrate that the approach proposed in this study improved the estimation performance of LNC in wheat, which could provide technical support in wheat growth monitoring.

scholarly journals Nitrogen fertilization management in white oat using spectral indices

2020 ◽  
Vol 50 ◽  
Author(s):  
José de Arruda Barbosa ◽  
Rogério Teixeira de Faria ◽  
Anderson Prates Coelho ◽  
Alexandre Barcellos Dalri ◽  
Luiz Fabiano Palaretti
Keyword(s):  
Remote Sensing ◽  
Nitrogen Fertilization ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
New Technology ◽  
Leaf Nitrogen ◽  
Leaf Nitrogen Content ◽  
Spectral Indices ◽  
Remote Sensing Techniques ◽  
Nitrogen Rates

ABSTRACT Remote sensing techniques have been considered a new technology in worldwide agriculture for diagnosing the plant nutritional demand. Fertilizer management efficiency is a goal to be achieved, and modern tools based on remote sensing are promising for monitoring the crop needs. This study aimed to evaluate the agronomic performance and relative economic return of white oat under nitrogen rates, as well as to verify whether the normalized difference vegetation index (NDVI) and leaf chlorophyll index (LCI) could be used for topdressing nitrogen fertilization management, in white oat. Treatments consisted of five topdressing nitrogen fertilization strategies: T1 - 160 kg ha-1 (reference rate); T2 - 90 kg ha-1 (recommended rate); T3 - 60 kg ha-1 (economic rate); T4 - 30 kg ha-1 (when NDVI < 90 % of T1); and T5 - 30 kg ha-1 (when LCI < 90 % of T1). The white oat did not respond to the topdressing nitrogen fertilization. Its temporal monitoring using spectral indices allowed dispensing the topdressing nitrogen fertilization without reducing the grain and biomass yields and the leaf nitrogen content, when compared to the recommended management (90 kg ha-1 of N as topdressing), with no differences between the evaluated spectral indices. Thus, both the NDVI and LCI spectral indices are promising tools for the topdressing nitrogen fertilization management in the white oat crop.

Sign in / Sign up

Export Citation Format

Share Document