Optimization of Sample Construction Based on NDVI for Cultivated Land Quality Prediction

Abstract

The integrated use of remote sensing technology and machine learning models to evaluate cultivated land quality (CLQ) quickly and efficiently is vital for protecting these lands. The effectiveness of machine-learning methods can be profoundly influenced by training samples. However, in the existing research, samples have mainly been constructed by random point (RPO). Little attention has been devoted to the optimization of sample construction, which may affect the accuracy of evaluation results. In this study, we present two optimization methods for sample construction of random patch (RPA) and area sequence patch (ASP). Differing from RPO samples, it aims to include cultivated land area and its size into sample construction. Based on landsat-8 Operational Land Manager images and agricultural land grading data, the proposed sample construction methods were applied to the machine learning model to predict the CLQ in Dongtai City, Jiangsu Province, China. Four machine learning models (the backpropagation neural network, decision tree, random forest (RF), and support vector machine) were compared based on RPO samples to determine the accurate evaluation model. The best machine learning model was selected to compare RPA and ASP samples with RPO samples. Results determined that the RF model generated the highest accuracy. Meanwhile, a high correlation was noted between the cultivated land area and CLQ. Thus, incorporating cultivated land area in the sample construction attributes can improve the prediction accuracy of the model. Among the three sample construction methods, the ASP yielded the highest prediction accuracy, indicating that the use of a large, cultivated land patch as the sample unit can further elevate the model performance. This study provides a new sample construction method for predicting CLQ using a machine learning model, as well as providing a reference for related research.

Research on the Inversion Model of Cultivated Land Quality Using High-Resolution Remote Sensing Data

“…In this paper, the importance of model variables is determined with the help of principal component analysis, and machine learning algorithms are used for the remote sensing inversion of cultivated land quality. The selection of model input variables is a prerequisite for accurate cultivated land quality inversion, and most of the previous studies on remote sensing inversion of cultivated land quality selected the vegetation index as the input variable of the model and inverted the cultivated land quality with the vegetation covered on the surface and the vegetation growth [37], which makes it difficult to accurately reflect the cultivated land quality.…”

Can remote sensing enable a Biomass Climate Adaptation Index for agricultural systems?

“…Htitiou et al (2021) use NDVI values extracted from spatial transects created across the study site to compare the performance of Very Deep Super-Resolution (VDSR) against the enhanced spatial and temporal adaptive reflectance fusion model (ESTARFM) and the flexible spatiotemporal data fusion (FSDAF) method in producing high resolution NDVI time series datasets. Based on the random point (RPO) sample construction method, Li et al (2022) have investigated the prediction capacity of four machine learning approaches, (backpropagation neural network, decision tree, RF, and support vector machine), to predict the quality of cultivated land, where RF was found to be the most accurate.…”

Multiple-model machine learning identifies potential functional genes in dilated cardiomyopathy

“…For example, least absolute shrinkage and selection operator (LASSO) processed a precision matrix of Gaussian variables using an ℓ1-penalty ( 13 ) until small values to zero but eliminated too many variables. For SVM, separated hyperplanes allow for correct partitioning and maximize geometric spacing but may be worse in a small sample size ( 14 ) compared with other MLs ( 15 ). Different ML algorithms possess both characteristics and limitations which cannot be ignored, especially in the choice of variables.…”