Search

High levels of chromium (Cr) in soil pose a significant threat to both humans and the environment. Laboratory-based chemical analysis methods for Cr are time consuming and expensive; thus, there is an urgent need for a more efficient method for detecting Cr in soil. In this study, a deep neural network (DNN) approach was applied to the Land Use and Cover Area frame Survey (LUCAS) dataset to develop a hyperspectral soil Cr content prediction model with good generalizability and accuracy. The optimal DNN model was constructed by optimizing the spectral preprocessing methods and DNN hyperparameters, which achieved good predictive performance for Cr detection, with a correlation coefficient value of 0.79 on the testing set. Four important hyperspectral bands with strong Cr sensitivity (400–439, 1364–1422, 1862–1934, and 2158–2499 nm) were identified by permutation importance and local interpretable model-agnostic explanations. Soil iron oxide and clay mineral content were found to be important factors influencing soil Cr content. The findings of this study provide a feasible method for rapidly determining soil Cr content from hyperspectral data, which can be further refined and applied to large-scale Cr detection in the future. [ABSTRACT FROM AUTHOR]

High levels of chromium (Cr) in soil pose a significant threat to both humans and the environment. Laboratory-based chemical analysis methods for Cr are time consuming and expensive; thus, there is an urgent need for a more efficient method for detecting Cr in soil. In this study, a deep neural network (DNN) approach was applied to the Land Use and Cover Area frame Survey (LUCAS) dataset to develop a hyperspectral soil Cr content prediction model with good generalizability and accuracy. The optimal DNN model was constructed by optimizing the spectral preprocessing methods and DNN hyperparameters, which achieved good predictive performance for Cr detection, with a correlation coefficient value of 0.79 on the testing set. Four important hyperspectral bands with strong Cr sensitivity (400–439, 1364–1422, 1862–1934, and 2158–2499 nm) were identified by permutation importance and local interpretable model-agnostic explanations. Soil iron oxide and clay mineral content were found to be important factors influencing soil Cr content. The findings of this study provide a feasible method for rapidly determining soil Cr content from hyperspectral data, which can be further refined and applied to large-scale Cr detection in the future.

Estimation of the amino acid content in maize leaves is helpful for improving maize yield estimation and nitrogen use efficiency. Hyperspectral imaging can be used to obtain the physiological and biochemical parameters of maize leaves with the advantages of being rapid, non-destructive, and high throughput. This study aims to estimate the multiple amino acid contents in maize leaves using hyperspectral imaging data. Two nitrogen (N) fertilizer experiments were carried out to obtain the hyperspectral images of fresh maize leaves. The partial least squares regression (PLSR) method was used to build the estimation models of various amino acid contents by using the reflectance of all bands, sensitive band range, and sensitive bands. The models were then validated with the independent dataset. The results showed that (1) the spectral reflectance of most amino acids was more sensitive in the range of 400–717.08 nm than other bands. The estimation accuracy was better by using the reflectance of the sensitive band range than that of all bands; (2) the sensitive bands of most amino acids were in the ranges of 505.39–605 nm and 651–714 nm; and (3) among the 24 amino acids, the estimation models of the b-aminobutyric acid, ornithine, citrulline, methionine, and histidine achieved higher accuracy than those of other amino acids, with the R2, relative root mean square error (RE), and relative percent deviation (RPD) of the measured and estimated value of testing samples in the range of 0.84– 0.96, 8.79%–19.77%, and 2.58–5.18, respectively. This study can provide a non-destructive and rapid diagnostic method for genetic sensitive analysis and variety improvement of maize. [ABSTRACT FROM AUTHOR]

This study aims to identify different types of stress on maize leaves using feature selection and spectral index methods. Spectral data were collected from leaves under heavy metal, water, fertilizer stress, as well as under normal healthy conditions. Preprocessing steps such as continuum removal (CR), standard normal variable (SNV) transformation, multiple scattering correction (MSC), detrend correction (DT), and first-order derivative (FOD) were applied to the raw spectra. Various feature selection methods including ReliefF, chi-square test, recursive feature elimination (FRE), mutual information (MI), random forest (RF), and gradient boosting tree (GBT) were employed to determine the importance scores of different spectral bands, thus identifying sensitive spectral features capable of distinguishing various stress types. Spectral indices for stress type differentiation were constructed using label correlation method. Classification models were built using support vector machine (SVM), K-nearest neighbors (KNN), Gaussian naive Bayes (GNB), extreme gradient boosting (XGBoost), RF, and adaptive boosting (AdaBoost) algorithms. Results indicate that the characteristic spectral bands for differentiating stress types are primarily distributed around the red edge (near 700–800 nm) and water absorption valley (near 1900 nm). Spectral indices constructed using combinations of spectral bands around the near-infrared plateau absorption valley (near 1185 nm) and water absorption valley (near 1460 nm) effectively differentiate maize stress types. Among the modeling classification algorithms, RF and AdaBoost algorithms exhibited optimal performance, demonstrating high classification accuracy on both training and validation sets. These findings hold promise for providing new technical support for maize stress monitoring and diagnosis in agricultural production. [ABSTRACT FROM AUTHOR]

Soil nitrogen (N) content plays a vital role in agriculture and biogeochemical processes, ranging from the N fertilization management for intensive agricultural production to the patterns of N cycling in agroecological systems. While proximal sensing in laboratory settings can achieve ideal soil N estimation accuracy, the estimation and mapping by using remote sensing methods in a large spatial scale diplays low ability. A new hyperspectral imager with 166 spectral channels, the ZY1-02D, makes possible the detection of subtle but important spectral features of soil. This study aimed at exploring the capability of the ZY1-02D to estimate and map the topsoil N content of the black soil-covered farmlands in northeast China. To this aim, 646 soil samples from study sites were collected, processed, spectrally and geochemically measured for the soil N sensitive bands detection and partial least squares regression (PLSR) calibration and validation. The sensitive bands detection results showed an appealing regularity of the variability and stable tendency of the soil N sensitive spectral bands with the change of the sample size. Based on this, we compared the estimation capacity of the models developed with the full wavelength spectra and the models developed with the sensitive bands. The estimation based on ZY1-02D full wavelength spectral reflectance were robust, with R2 of 0.64 in validation. Further, the results of model developed with the sensitive bands showed better validation accuracy with R2 of 0.66 and were applied to create a map of topsoil N content of farmlands in the northeast China black soil area. The results demonstrated that sensitive bands modelling could enhance the accuracy of the estimation and simplify model, and what is more, showed the ideal capability of ZY1-02D for soil N content estimation at the regional scale.

Crop nitrogen monitoring techniques, particularly choosing sensitive monitoring bands and suitable monitoring models, have great significance both in theory and in practice for achieving non-destructive monitoring of nitrogen concentration and accurate management of water and fertilizer in large-scale areas. In this study, a lysimeter experiment was carried out to examine the characteristics of canopy spectral reflectance variation of summer corn under different fertilization levels. The relationship between canopy spectral reflectance and nitrogen concentration was investigated, based on which sensitive bands for the corn canopy nitrogen monitoring were selected and a suitable spectral index model was determined. The results suggest that under different fertilization levels, the canopy spectral reflectance of summer corn decreases with the increase of the canopy nitrogen concentration in the visible light band, but varies in the opposite direction in the near-infrared band, with a premium put on a higher correlation between the spectral reflectance of the characteristic bands and their first derivatives and the canopy nitrogen concentration. The most sensitive bands for monitoring the canopy nitrogen concentration using spectral reflectance and its first derivative are found to be 762 nm and 726 nm and the correlation coefficients are 0.550 and 0.795, respectively. The optimal band combination, generated by multivariate stepwise regression analysis, is composed of 762 nm, 944 nm and 957 nm bands. From the 55 reported spectral index models of crop nitrogen concentration monitoring, the most suitable index model, NDRE, is chosen such that this index model has the highest correlation with the canopy nitrogen concentration in summer corn. This model has a significant positive correlation with the canopy nitrogen concentration at each growth period, and the correlation coefficient is up to 0.738 during the whole growth period. Spectral monitoring models of canopy nitrogen concentration are constructed using sensitive bands, and a combination of bands and the spectral index, suggesting that these models perform well in monitoring. The models arranged in descending order of simulation accuracy are as follows: the suitable spectral index model, the optimal band combination model, the sensitive band reflectance first derivative model, the sensitive band reflectance model. The determination coefficients are 0.754, 0.711, 0.639 and 0.306, respectively.

• An inversion model based on thickness correction using modified Kubelka-Munk (K-M) theory. • Soil thickness observation experiment based on K-M theory was designed. • The high-precision soil organic matter inversion model was constructed by the scattering coefficients. • The sensitive band of soil organic matter is at 2.197 μm. • Recommended soil thickness for soil samples should be more than 7 mm. The existing research into soil component inversion based on spectroscopic techniques has mainly focused on traditional statistical learning. However, the most prominent drawback of this approach is the difficulty in obtaining the soil components' sensitive bands with explanatory inversion mechanisms. Whether for soil organic matter inversion or soil heavy metal inversion, there is still a lack of inversion models based on the physical theory of remote sensing. Hence, in this paper, an inversion model based on thickness correction using Kubelka-Munk (K-M) theory is proposed. Firstly, in this study, a soil thickness observation experiment based on K-M theory was undertaken. The impact of the soil thickness and the material of the container on the spectra was explored by selecting different experimental samples with different background container materials. A modified K-M thickness model was then developed by combining indoor spectral data. This allows the corresponding scattering coefficients and absorption coefficients for soil samples with different organic matter contents to be calculated. The optimal organic matter inversion model can then be constructed by the scattering coefficients, with the sensitive band at 2.197 μm. The results obtained in this study demonstrate the feasibility and superiority of the proposed method and further explain the sensitive bands of soil organic matter in hyperspectral data, with a determination coefficient accuracy of up to 0.97. The experimental results also demonstrate that the recommended soil thickness for soil samples should be more than 7 mm. In addition, when selecting background container materials, materials with obvious reflectance peak and valley characteristics should be avoided. [ABSTRACT FROM AUTHOR]

Estimation of the amino acid content in maize leaves is helpful for improving maize yield estimation and nitrogen use efficiency. Hyperspectral imaging can be used to obtain the physiological and biochemical parameters of maize leaves with the advantages of being rapid, non-destructive, and high throughput. This study aims to estimate the multiple amino acid contents in maize leaves using hyperspectral imaging data. Two nitrogen (N) fertilizer experiments were carried out to obtain the hyperspectral images of fresh maize leaves. The partial least squares regression (PLSR) method was used to build the estimation models of various amino acid contents by using the reflectance of all bands, sensitive band range, and sensitive bands. The models were then validated with the independent dataset. The results showed that (1) the spectral reflectance of most amino acids was more sensitive in the range of 400-717.08 nm than other bands. The estimation accuracy was better by using the reflectance of the sensitive band range than that of all bands; (2) the sensitive bands of most amino acids were in the ranges of 505.39-605 nm and 651-714 nm; and (3) among the 24 amino acids, the estimation models of the β-aminobutyric acid, ornithine, citrulline, methionine, and histidine achieved higher accuracy than those of other amino acids, with the R 2 , relative root mean square error (RE), and relative percent deviation (RPD) of the measured and estimated value of testing samples in the range of 0.84-0.96, 8.79%-19.77%, and 2.58-5.18, respectively. This study can provide a non-destructive and rapid diagnostic method for genetic sensitive analysis and variety improvement of maize., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Shu, Zhou, Chen, Wang, Meng and Ma.)

Based on the field hyperspectral data from the analytical spectral devices (ASD) spectrometer, we characterized the spectral properties of rice canopies infested with brown spot disease and selected spectral regions and bands sensitive to four severity degrees (severe, moderate, light, and healthy). The results show that the curves' variation on the original and the first- and second-order derivative curves are greatly different, but the spectral difference in the near-infrared region is the most obvious for each level. Specifically, the peaks are located at 822, 738, and 793 nm, while the valleys are located at 402, 570, and 753 nm, respectively. The sensitive regions are between 430-520, 530-550, and 650-710 nm, and the bands are 498, 539, and 673 nm in the sensitivity analysis, while they are in the ranges of 401-530, 550-730 as well as at 498 nm and 678 nm in the continuum removal. [ABSTRACT FROM AUTHOR]

Abstract: Soil salinization is an important desertification process that threatens the stability of ecosystems, especially in arid lands. Quantifying and mapping soil salinity to monitor soil salinization is difficult because of its large spatial and temporal variability. There has been a growing interest in the use of hyperspectral reflectance as a rapid and inexpensive tool for soil salinity characterization in the recent past. However, as soil moisture often jointly affects soil reflectance, a moisture-insensitive reflectance model is needed to provide the base for soil salinity monitoring from soil reflectance. In this paper, we developed an exponent reflectance model to estimate soil salt contents inversely under various soil moisture conditions, based on a control laboratory experiment on the two factors (soil salinity and soil moisture) to soil reflectance. Main soil salt types (Na2SO4, NaCl, Na2CO3) with wide soil salinity (0% to 20%) and soil moisture (1.75% to 20%) levels (in weight base) from Western China were examined for their effects on soil reflectance through a model based approach. Moisture resistant but salt sensitive bands of reflected spectra have been identified for the model before being applied to inversely estimate soil salt content. Sensitive bands for Na2SO4 type of salt affected soils were identified as from 1920 to 2230nm, and 1970 to 2450nm for NaCl, 350 to 400nm for Na2CO3 type of salt affected soils, respectively. The sensitive bands focused on ranged from 1950 to 2450nm when all data were considered when ignoring salt types. The model was then applied to inversely estimate soil salt contents. High R2 of 0.87, 0.79, and 0.66, and low mean relative error (MRE) of 16.42%, 21.17%, and 27.16%; have been obtained for NaCl, Na2SO4 and Na2CO3, respectively. Performance of the inverse model dropped but remained significant when ignoring salt types with an R2 of 0.56 and a MRE of 33.25%. The approach proposed in this study should thus provide a new direction for estimating salinity from reflectance under various soil moisture conditions and should have wide applications in future monitoring of soil salinization. [Copyright &y& Elsevier]

Soil nitrogen (N) content plays a vital role in agriculture and biogeochemical processes, ranging from the N fertilization management for intensive agricultural production to the patterns of N cycling in agroecological systems. While proximal sensing in laboratory settings can achieve ideal soil N estimation accuracy, the estimation and mapping by using remote sensing methods in a large spatial scale diplays low ability. A new hyperspectral imager with 166 spectral channels, the ZY1-02D, makes possible the detection of subtle but important spectral features of soil. This study aimed at exploring the capability of the ZY1-02D to estimate and map the topsoil N content of the black soil-covered farmlands in northeast China. To this aim, 646 soil samples from study sites were collected, processed, spectrally and geochemically measured for the soil N sensitive bands detection and partial least squares regression (PLSR) calibration and validation. The sensitive bands detection results showed an appealing regularity of the variability and stable tendency of the soil N sensitive spectral bands with the change of the sample size. Based on this, we compared the estimation capacity of the models developed with the full wavelength spectra and the models developed with the sensitive bands. The estimation based on ZY1-02D full wavelength spectral reflectance were robust, with R2 of 0.64 in validation. Further, the results of model developed with the sensitive bands showed better validation accuracy with R2 of 0.66 and were applied to create a map of topsoil N content of farmlands in the northeast China black soil area. The results demonstrated that sensitive bands modelling could enhance the accuracy of the estimation and simplify model, and what is more, showed the ideal capability of ZY1-02D for soil N content estimation at the regional scale. [ABSTRACT FROM AUTHOR]

Crop nitrogen monitoring techniques, particularly choosing sensitive monitoring bands and suitable monitoring models, have great significance both in theory and in practice for achieving non-destructive monitoring of nitrogen concentration and accurate management of water and fertilizer in large-scale areas. In this study, a lysimeter experiment was carried out to examine the characteristics of canopy spectral reflectance variation of summer corn under different fertilization levels. The relationship between canopy spectral reflectance and nitrogen concentration was investigated, based on which sensitive bands for the corn canopy nitrogen monitoring were selected and a suitable spectral index model was determined. The results suggest that under different fertilization levels, the canopy spectral reflectance of summer corn decreases with the increase of the canopy nitrogen concentration in the visible light band, but varies in the opposite direction in the near-infrared band, with a premium put on a higher correlation between the spectral reflectance of the characteristic bands and their first derivatives and the canopy nitrogen concentration. The most sensitive bands for monitoring the canopy nitrogen concentration using spectral reflectance and its first derivative are found to be 762 nm and 726 nm and the correlation coefficients are 0.550 and 0.795, respectively. The optimal band combination, generated by multivariate stepwise regression analysis, is composed of 762 nm, 944 nm and 957 nm bands. From the 55 reported spectral index models of crop nitrogen concentration monitoring, the most suitable index model, NDRE, is chosen such that this index model has the highest correlation with the canopy nitrogen concentration in summer corn. This model has a significant positive correlation with the canopy nitrogen concentration at each growth period, and the correlation coefficient is up to 0.738 during the whole growth period. Spectral monitoring models of canopy nitrogen concentration are constructed using sensitive bands, and a combination of bands and the spectral index, suggesting that these models perform well in monitoring. The models arranged in descending order of simulation accuracy are as follows: the suitable spectral index model, the optimal band combination model, the sensitive band reflectance first derivative model, the sensitive band reflectance model. The determination coefficients are 0.754, 0.711, 0.639 and 0.306, respectively. [ABSTRACT FROM AUTHOR]