Leaf Hyperspectral Data Research Articles

Comparison of leaf chlorophyll content retrieval performance of citrus using FOD and CWT methods with field-based full-spectrum hyperspectral reflectance data

Citrus is one of the most economically valuable fruit trees in the world, for which leaf chlorophyll content (LCC) serves as a crucial indicator for evaluating its growth and health status. However, the quantitative estimation of LCC using remote-sensing techniques is still challenging owing to unclear sensitive spectral ranges, baseline drift and overlapping spectrum peaks. To resolve these issues, we clarified the spectral response characteristics of citrus LCC using fractional-order derivatives (FOD) and continuous wavelet transform (CWT) methods to determine its sensitive spectral range with in situ full-spectrum leaf hyperspectral data. We proposed a novel method for estimating the LCC of citrus by combining an ensemble learning regression model based on Hyperopt optimization (H-ELR) with partial least squares regression (PLSR) using the ultra-high dimensional feature variables produced by dual- and tri-band combination strategies. We evaluated the retrieval performance of LCC between the FOD- and CWT-based optimal spectral feature variables. Besides, we further examined the feasibility of improving the estimation accuracy of LCC by the combination of their optimal feature variables. Finally, we evaluated the effect of the spectral curve trend changes and different dimensional spectral features on LCC estimation. The results showed that: (1) The FOD- and CWT-based methods improved the correlation between original spectral reflectance and LCC, with the correlation coefficient increasing by 0.046 and 0.054, respectively. We confirmed that 425–740 nm is the optimal spectral range for LCC estimation. (2) We found that the 0.9 order derivative Tri-band index (TBI2 (R571, R1697, R740)) constructed by combining the sensitive spectral bands of leaf water content achieved a high-precision LCC estimation (R2 = 0.876). In addition, the MERIS terrestrial chlorophyll index (MTCI) constructed based on the scale6 of CWT also gained good retrieval accuracy (R2 = 0.806). (3) We demonstrated that a combination of FOD-based and CWT-based sensitive spectral features improves estimation accuracies (R2 = 0.891) of LCC and revealed that the reflectance peaks and slope peaks at 550 and 750 nm are essential variables for predicting the citrus LCC. (4) The combination of PLSR and H-ELR model provided a good retrieval performance of citrus LCC with the kurtosis (γ = 3.2) and skewness (Sk = 0.066) of the residual prediction values. Our proposed method can provide a scientific basis for estimating LCC and other physiological parameters of citrus and other crop types, which is also important to optimize agricultural management practices and improve crop yield.

A leaf-level spectral library to support high-throughput plant phenotyping: predictive accuracy and model transfer.

Leaf-level hyperspectral reflectance has become an effective tool for high-throughput phenotyping of plant leaf traits due to its rapid, low-cost, multi-sensing, and non-destructive nature. However, collecting samples for model calibration can still be expensive, and models show poor transferability among different datasets. This study had three specific objectives: first, to assemble a large library of leaf hyperspectral data (n=2460) from maize and sorghum; second, to evaluate two machine-learning approaches to estimate nine leaf properties (chlorophyll, thickness, water content, nitrogen, phosphorus, potassium, calcium, magnesium, and sulfur); and third, to investigate the usefulness of this spectral library for predicting external datasets (n=445) including soybean and camelina using extra-weighted spiking. Internal cross-validation showed satisfactory performance of the spectral library to estimate all nine traits (mean R2=0.688), with partial least-squares regression outperforming deep neural network models. Models calibrated solely using the spectral library showed degraded performance on external datasets (mean R2=0.159 for camelina, 0.337 for soybean). Models improved significantly when a small portion of external samples (n=20) was added to the library via extra-weighted spiking (mean R2=0.574 for camelina, 0.536 for soybean). The leaf-level spectral library greatly benefits plant physiological and biochemical phenotyping, whilst extra-weight spiking improves model transferability and extends its utility.

Rice Variety Identification Based on the Leaf Hyperspectral Feature via LPP-SVM

Rice variety identification is important for genetic breeding classification and crop yield estimation. Traditional identification methods are time-consuming and inaccurate. This paper proposes a method for rice variety identification based on the hyperspectral characteristics of leaves. Hyperspectral data of rice leaves were collected using a geophysical spectrometer imaging system. To reduce the redundance among the hyperspectral data and save the identification cost, locality preserving projections (LPP) is first applied to extract low-dimensional representative features from the leaf hyperspectral data. Then, support vector machine (SVM) is combined for conducting the identification of rice varieties. The experimental results show that the identification rate of 10 varieties of early rice was found to be 91.67% and the identification rate of 10 varieties of late rice was 97.33%.

Inversion of Soil Heavy Metal Content Based on Spectral Characteristics of Peach Trees

There exists serious heavy metal contamination of agricultural soils in China. It is not only time- and labor-intensive to monitor soil contamination, but it also has limited scope when using conventional chemical methods. However, the method of the heavy metal monitoring of soil based on vegetation hyperspectral technology can break through the vegetation barrier and obtain the heavy metal content quickly over large areas. This paper discusses a highly accurate method for predicting the soil heavy metal content using hyperspectral techniques. We collected leaf hyperspectral data outdoors, and also collected soil samples to obtain heavy metal content data using chemical analysis. The prediction model for heavy metal content was developed using a difference spectral index, which was not highly satisfactory. Subsequently, the five factors that have a strong influence on the content of heavy metals were analyzed to determine multiple regression models for the elements As, Pb, and Cd. The results showed that the multiple regression model could better estimate the heavy metal content with stable fitting that has high prediction accuracy compared with the linear model. The results of this research provide a scientific basis and technical support for the hyperspectral inversion of the soil heavy metal content.

Rapid determination of leaf water content for monitoring waterlogging in winter wheat based on hyperspectral parameters

Waterlogging is becoming an obvious constraint on food production due to the frequent occurrence of extremely high-level rainfall events. Leaf water content (LWC) is an important waterlogging indicator, and hyperspectral remote sensing provides a non-destructive, real-time and reliable method to determine LWC. Thus, based on a pot experiment, winter wheat was subjected to different gradients of waterlogging stress at the jointing stage. Leaf hyperspectral data and LWC were collected every 7 days after waterlogging treatment until the winter wheat was mature. Combined with methods such as vegetation index construction, correlation analysis, regression analysis, BP neural network (BPNN), etc., we found that the effect of waterlogging stress on LWC had the characteristics of hysteresis and all waterlogging stress led to the decrease of LWC. LWC decreased faster under severe stress than under slight stress, but the effect of long-term slight stress was greater than that of short-term severe stress. The sensitive spectral bands of LWC were located in the visible (VIS, 400–780 nm) and short-wave infrared (SWIR, 1 400–2 500 nm) regions. The BPNN Model with the original spectrum at 648 nm, the first derivative spectrum at 500 nm, the red edge position (λr), the new vegetation index RVI (437, 466), NDVI (437, 466) and NDVI’ (747, 1 956) as independent variables was the best model for inverting the LWC of waterlogging in winter wheat (modeling set: R2=0.889, RMSE=0.138; validation set: R2=0.891, RMSE=0.518). These results have important theoretical significance and practical application value for the precise control of waterlogging stress.

Mapping leaf chlorophyll content of mangrove forests with Sentinel-2 images of four periods

Chlorophyll is a good indicator of health status and nutritional condition as plant grows. Many studies have investigated the feasibility of retrieving leaf chlorophyll content (LCC) using vegetation indices (VIs) of multiple plant species, yet very few studies have examined the multi-temporal Sentinel-2 images for mapping LCC of mangrove forests. With field collected leaf SPAD values (relative chlorophyll content), this study explored the relationship of leaf SPAD values against five types of newly-developed VIs derived from leaf hyperspectral data and Sentinel-2 data of four periods (May 2018, January 2019, August 2019, and December 2019). Linear regression with best-performing VIs and Kernel Ridge Regression (KRR) were developed to construct the SPAD retrieval model in each period. The leave-one-out cross-validation technique was employed to compare the estimation results of VIs and KRR method, and the four periods of SPAD maps were produced by the best-performing model. The results showed that the newly-developed index (ratio of single-band reflectance to the sum of two bands reflectance, RSSI) achieved the high correlation coefficient with leaf SPAD value at both leaf and canopy level. At canopy level, the linear model using RSSI (B8/(B2 + B5), B8a/(B2 + B4), B8/(B2 + B5), and B8/(B2 + B3)) outperformed than that using traditional broadband indices and KRR model with R2adjust = 0.496, 0.742, 0.681, and 0.801; RMSE = 5.75, 4.29, 4.00, and 3.46; and RE = 7.67%, 5.68%, 4.97%, and 4.63% in each period. We concluded that there are great potentials of newly-developed index of RSSI using Sentinel-2 data for regional retrieving and mapping LCC of mangrove forests across different time periods, which is essential for mangrove ecological conservation and restoration.

Estimation of heavy metal concentration in rice leaves of farmland by hyperspectral leaf reflectance and partial least square regression

Protecting people from heavy metal contamination is an important public-health concern and a major national environmental issue in China. The objective of this study was to quantitatively estimate the heavy metal concentration in rice leaves using leaf hyperspectral data and partial least squares regression (PLSR) models. 21 rice leaf samples and spectrum were gathered from farmlands in Zhangjiagang area, China. Copper (Cu), Cadmium (Cd) concentrations of rice leaves were measured within the lab. Firstly, the spectral data were treated by some methods, including, original reflectance (OS), First Derivative (FD) and Second Derivative (SD). Secondly, in order to select input variables for PLSR models, the correlation analysis between heavy metal concentration and spectral bands (OS, FD and SD), spectral indices were performed. Finally, we constructed the PLSR models between heavy metal concentration and spectrum. The results showed that correlation coefficients between Cu concentration and spectral data were higher than Cd. And that the bands significant correlation (P<0.05) with Cu concentration were far more than Cd. Ultimately, we selected 453 variables (442 bands and 11 spectral indices) and 19 variables (18 bands and 1 spectral index) as input variables of PLSR model for Cu and Cd, respectively. Moreover, we found that the Cu and Cd concentrations significantly correlated with spectral variables for (R2=0.41, RMSE=1.93) and (R2=0.38, RMSE=0.018) of PLSR models, respectively. These results indicated that they were good predicting models for estimating heavy metal concentration in rice leaves.

Detecting Sulfuric and Nitric Acid Rain Stresses on Quercus glauca through Hyperspectral Responses.

Acid rain, which has become one of the most severe global environmental issues, is detrimental to plant growth. However, effective methods for monitoring plant responses to acid rain stress are currently lacking. The hyperspectral technique provides a cost-effective and nondestructive way to diagnose acid rain stresses. Taking a widely distributed species (Quercus glauca) in Southern China as an example, this study aims to monitor the hyperspectral responses of Q. glauca to simulated sulfuric acid rain (SAR) and nitric acid rain (NAR). A total of 15 periods of leaf hyperspectral data under four pH levels of SAR and NAR were obtained during the experiment. The results showed that hyperspectral information could be used to distinguish plant responses under acid rain stress. An index (green peak area index, GPAI) was proposed to indicate acid rain stresses, based on the significantly variations in the region of 500–660 nm. Light acid rain (pH 4.5 SAR and NAR) promoted Q. glauca growth relative to the control groups (pH 5.6 SAR and NAR); moderate acid rain (pH 3.0 SAR) firstly promoted and then inhibited plant growth, while pH 3.0 NAR showed mild inhibitory effects during the experiment; and heavy acid rain (pH 2.0) significantly inhibited plant growth. Compared with NAR, SAR induced more serious damages to Q. glauca. These results could help monitor acid rain stress on plants on a regional scale using remote sensing techniques.

Early detection of crop injury from herbicide glyphosate by leaf biochemical parameter inversion

Early detection of crop injury from herbicide glyphosate is of significant importance in crop management. In this paper, we attempt to detect glyphosate-induced crop injury by PROSPECT (leaf optical PROperty SPECTra model) inversion through leaf hyperspectral reflectance measurements for non-Glyphosate-Resistant (non-GR) soybean and non-GR cotton leaves. The PROSPECT model was inverted to retrieve chlorophyll content (Ca+b), equivalent water thickness (Cw), and leaf mass per area (Cm) from leaf hyperspectral reflectance spectra. The leaf stress conditions were then evaluated by examining the temporal variations of these biochemical constituents after glyphosate treatment. The approach was validated with greenhouse-measured datasets. Results indicated that the leaf injury caused by glyphosate treatments could be detected shortly after the spraying for both soybean and cotton by PROSPECT inversion, with Ca+b of the leaves treated with high dose solution decreasing more rapidly compared with leaves left untreated, whereas the Cw and Cm showed no obvious difference between treated and untreated leaves. For both non-GR soybean and non-GR cotton, the retrieved Ca+b values of the glyphosate treated plants from leaf hyperspectral data could be distinguished from that of the untreated plants within 48h after the treatment, which could be employed as a useful indicator for glyphosate injury detection. These findings demonstrate the feasibility of applying the PROSPECT inversion technique for the early detection of leaf injury from glyphosate and its potential for agricultural plant status monitoring.

Early Detection of Crop Injury from Glyphosate on Soybean and Cotton Using Plant Leaf Hyperspectral Data

In this paper, we aim to detect crop injury from glyphosate, a herbicide, by both traditionally used spectral indices and newly extracted features with leaf hyperspectral reflectance data for non-Glyphosate-Resistant (non-GR) soybean and non-GR cotton. The new features were extracted by canonical analysis technique, which could provide the largest separability to distinguish the injured leaves from the healthy ones. Spectral bands used for constructing these new features were selected based on the sensitivity analysis results of a physically-based leaf radiation transfer model (leaf optical PROperty SPECTra model, PROSPECT), which could help extend the effectiveness of these features to a wide range of leaf structures and growing conditions. This approach has been validated with greenhouse measured data acquired in glyphosate treatment experiments. Results indicated that glyphosate injury could be detected by NDVI (Normalized Difference Vegetation Index), RVI (Ratio Vegetation Index), SAVI (Soil Adjusted Vegetation Index), and DVI (Difference Vegetation Index) in 48 h After the Treatment (HAT) for soybean and in 72 HAT for cotton, but the other spectral indices either showed little use for separation, or did not show consistent separation for healthy and injured soybean and cotton. Compared with the traditional spectral indices, the new features were more feasible for the early detection of glyphosate injury, with leaves sprayed with a higher rate of glyphosate solution having larger feature values. This trend became more and more pronounced with time. Leaves sprayed with different glyphosate rates showed some separability 24 HAT using the new features and could be totally distinguished at and beyond 48 HAT for both soybean and cotton. These findings demonstrated the feasibility of applying leaf hyperspectral reflectance measurements for the early detection of glyphosate injury using these newly proposed features.

Plant Species Identification Based on Independent Component Analysis for Hyperspectral Data

By investigating the possibility of plant species classification based on independent component analysis (ICA) for hyperspectral data with minor difference, the framework of a general plant species classification model that consists of ICA based data reduction, classifier training and verification is proposed in this paper. Five different types of discriminant analysis classifiers including Linear, Quadratic, DiagLinear, DiagQuatic and Mahalanobis, with data reduction that based on principal components analysis (PCA) and ICA, are implemented and compared. Accuracy assessment of classification for real leaf hyperspectral data is demonstrated, indicating that data reduction based on ICA performs better than that of PCA. Moreover, the proposed classification model with ICA based data reduction and Quadratic Discriminant Analysis works best, and its accuracy is about 98.35% with dimension 25 reduced from 2500.

Discrimination of Switchgrass Cultivars and Nitrogen Treatments Using Pigment Profiles and Hyperspectral Leaf Reflectance Data

The objective of this study was to compare the use of hyperspectral narrowbands, hyperspectral narrowband indices and pigment measurements collected from switchgrass leaf as potential tools for discriminating among twelve switchgrass cultivars and five N treatments in one cultivar (Alamo). Hyperspectral reflectance, UV-B absorbing compounds, photosynthetic pigments (chlorophyll a, chlorophyll b and carotenoids) of the uppermost fully expanded leaves were determined at monthly intervals from May to September. Leaf hyperspectral data was collected using ASD FieldSpec FR spectroradiometer (350–2,500 nm). Discrimination of the cultivars and N treatments were determined based on Principal Component Analysis (PCA) and linear discriminant analysis (DA). The stepwise discriminant analysis was used to determine the best indices that differentiate switchgrass cultivars and nitrogen treatments. Results of PCA showed 62% of the variability could be explained in PC1 dominated by middle infrared wavebands, over 20% in PC2 dominated by near infrared wavebands and just over 10% in PC3 dominated by green wavebands for separating both cultivars and N treatments. Discriminating among the cultivars resulted in an overall accuracy of 81% with the first five PCs in the month of September, but was less accurate (27%) in classifying N treatments using the spectral data. Discrimination based on pigment data using the first two PCs resulted in an overall accuracy of less than 10% for separating switchgrass cultivars , but was more accurate (47%) in grouping N treatments. The plant senescence ratio index (PSRI) was found to be the best index for separating the cultivars late in the season, while the transform chlorophyll absorption ration index (TCARI) was best for separating the N treatments. Leaf spectra data was found to be more useful than pigment data for the discrimination of switchgrass cultivars, particularly late in the growing season.