The GA-BPNN-Based Evaluation of Cultivated Land Quality in the PSR Framework Using Gaofen-1 Satellite Data

Abstract

Rapid and efficient assessment of cultivated land quality (CLQ) using remote sensing technology is of great significance for protecting cultivated land. However, it is difficult to obtain accurate CLQ estimates using the current satellite-driven approaches in the pressure-state-response (PSR) framework, owing to the limitations of linear models and CLQ spectral indices. In order to improve the estimation accuracy of CLQ, this study used four evaluation models (the traditional linear model; partial least squares regression, PLSR; back propagation neural network, BPNN; and BPNN with genetic algorithm optimization, GA-BPNN) to evaluate CLQ for determining the accurate evaluation model. In addition, the optimal satellite-derived indicator in the land state index was selected among five vegetation indices (the normalized vegetation index, NDVI; enhanced vegetation index, EVI; modified soil-adjusted vegetation index, MSAVI; perpendicular vegetation index, PVI; and soil-adjusted vegetation index, SAVI) to improve the prediction accuracy of CLQ. This study was conducted in Conghua District of Guangzhou, Guangdong Province, China, based on Gaofen-1 (GF-1) data. The prediction accuracies from the traditional linear model, PLSR, BPNN, and GA-BPNN were compared using observations. The results demonstrated that (1) compared with other models (the traditional linear model: R2 = 0.14 and RMSE = 91.53; PLSR: R2 = 0.33 and RMSE = 74.58; BPNN: R2 = 0.50 and RMSE = 61.75), the GA-BPNN model based on EVI in the land state index provided the most accurate estimates of CLQ, with the R2 of 0.59 and root mean square error (RMSE) of 56.87, indicating a nonlinear relationship between CLQ and the prediction indicator; and (2) the GA-BPNN-based evaluation approach of CLQ in the PSR framework was driven to map CLQ of the study area using the GF-1 data, leading to an RMSE of 61.44 at the regional scale, implying that the GA-BPNN-based evaluation approach has the potential to map CLQ over large areas. This study provides an important reference for the high-accuracy prediction of CLQ based on remote sensing technology.

Crop Growth Stage GPP-Driven Spectral Model for Evaluation of Cultivated Land Quality Using GA-BPNN

“…Based on the comparison of the evaluation results from three kinds of models, SVR (average R 2 = 0.64 and NRMSE = 9.78%) and GA-BPNN (average R 2 = 0.69 and NRMSE = 8.59%) models performed better than the PLSR model (average R 2 = 0.38 and NRMSE = 11.55%), implying that there is obvious non-linear correlation of CLQ with GPP spectral indicator. This conclusion is consistent with the findings of previous studies [16], indicating that the non-linear models are appropriate. It was also found that the GA-BPNN models provided more accurate predictions of CLQ than the SVR and PLSR models, which was mainly attributed to the integration of BPNN with GA which has the ability of optimizing the BPNN weights and thresholds.…”

“…Thus, it is necessary to realize real-time monitoring and evaluation of CLQ in agricultural regions, especially vulnerable or urban fringe areas [48,49]. Previous studies [4,15,16] on remote sensing-based evaluation of CLQ mainly focused on retrieving spectral indicators in both the traditional evaluation system and the PSR framework system. However, it is impossible to acquire accurate CLQ data using the previous evaluation methods due to ignoring spectral relationships of spectral indicators from crop growth stages with CLQ.…”

“…For example, Liu et al (2010) developed a linear model for evaluating CLQ based on predictors, including slope, sandy area ratio in a pixel, and modified soil-adjusted vegetation index [4,15]. Liu et al (2019) generated the spatial distribution of CLQ estimates based on the Genetic Algorithm-Based Back Propagation Neural Network (GA-BPNN) model. The authors utilized five remote sensing data derived predictors, including Slope, Vegetation Index, Temperature Vegetation Dryness Index, Road Accessibility, and Patch Fractal Dimension, and found that CLQ was significantly and nonlinearly correlated with the spectral predictors [16].…”

“…Liu et al (2019) generated the spatial distribution of CLQ estimates based on the Genetic Algorithm-Based Back Propagation Neural Network (GA-BPNN) model. The authors utilized five remote sensing data derived predictors, including Slope, Vegetation Index, Temperature Vegetation Dryness Index, Road Accessibility, and Patch Fractal Dimension, and found that CLQ was significantly and nonlinearly correlated with the spectral predictors [16]. Xie et al (2018) developed a frequent pattern-growth algorithm for improving the evaluation efficiency of CLQ [17].…”

Spatial Structure Characteristics of Slope Farmland Quality in Plateau Mountain Area: A Case Study of Yunnan Province, China

“…At present, the frequently-used farmland quality evaluation model includes the weighted sum method, land productivity evaluation method, Pressure-state-response (P-S-R) frame model method, analytic hierarchy process, fuzzy comprehensive evaluation method, farmland potential evaluation method, suitability evaluation method, soil environment quality evaluation method, etc. Among them, the weighing-sum method, land productivity evaluation method, and P-S-R frame model method are the most widely used [31]. In addition, the introduction of GIS, remote sensing(RS), etc.…”