No abstract

Abstract:Field spectroscopy has been suggested to be an efficient method for predicting soil properties using quantitative mathematical models in a rapid and non-destructive manner. Traditional multivariate regression algorithms usually regard the modeling of each soil property as a single task, which means only one response variable is considered as the output during modeling. Therefore, these algorithms are less suitable for the prediction of several key soil properties with low concentrations or unobvious spectral absorption signals. In the current study, we investigated the performance of a linear multi-task learning (LMTL) algorithm based on a regularized dirty model for modeling and predicting several key soil properties using field spectroscopy (350-2500 nm) as an integrated approach. We tested seven key soil properties including available nitrogen (N), phosphorus (P) and potassium (K), pH, water content (WC), organic matter (OM), and electrical conductivity (EC) in drylands. The model performances of LMTL models were compared with the commonly used single-task algorithm of the partial least squares regression (PLS-R). Our results show that the LMTL models outperformed the PLS-R models with the advantage of shared features; the ratio of performance to deviation (RPD) values in the validation set improved by 10.24%, 4.93%, 25.77%, 11.76%, 6.74%, 53.13%, and 3.15% for N, P, K, pH, WC, OM, and EC, respectively. The best prediction was obtained for OM with RPD = 2.29, indicating high accuracy (RPD > 2). The prediction results of N, P, WC, and pH were categorized as of moderate accuracy (1.4 < RPD < 2), while K and EC were categorized as of poor accuracy (RPD < 1.4). However, the explanatory power of the LMTL models was moderate due to fewer features being selected by the regularization algorithm of the LMTL approach, which should be further studied in the soil spectral analysis. Our results highlight the use of LMTL in field spectroscopy analysis that can improve the generalization performance of regression models for predicting soil properties.

In this study, a rapid and non-invasive technology for predicting soil moisture content (SMC) was presented based on hyperspectral imaging (HSI). Firstly, a set of HSI system was developed to collect both spectral (400-1000 nm) and spatial (1620×841 pixels) information from sandy soil samples with variable SMC levels in the laboratory. Principal component analysis (PCA) transformation, K-means clustering, and several other image processing methods were performed to obtain a region of interest (ROI) of soil sample from the original HSI data. Then, 256 optimal spectral wavelengths were selected from the average reflectance of the ROI, and 28 textural features were extracted using a gray-level co-occurrence matrix (GLCM). Data dimensionality reduction was conducted on both the spectral information and textural information by using a partial least square algorithm. Six latent variables (LVs) extracted from the spectral information, four LVs extracted from the textural information and fused data were used to build regression models with a three-layer BPNN, respectively. The results showed that all of the three calibration models achieved high prediction accuracy, particularly when using spectral information with R 2 C =0.9532 and RMSEC=0.0086. However, validation models demonstrate that predicting SMC using fused data is more effective than using spectral reflectance and textural features separately, with a R 2 P =0.9350 and RMSEP=0.0141, thus proving that the HSI technique is capable of detecting SMC.