Classification of Rice Heavy Metal Stress Levels Based on Phenological Characteristics Using Remote Sensing Time-Series Images and Data Mining Algorithms.

Tianjiao Liu,Xiangnan Liu,Ling Wu,Meiling Liu

doi:10.3390/s18124425

Tianjiao Liu, Xiangnan Liu + Show 2 more

Open Access

https://doi.org/10.3390/s18124425

Copy DOI

Abstract

Heavy metal pollution in crops leads to phenological changes, which can be monitored by remote sensing technology. The present study aims to develop a method for accurately evaluating heavy metal stress in rice based on remote sensing phenology. First, the enhanced spatial and temporal adaptive reflectance fusion model (ESTARFM) was applied to blend Moderate Resolution Imaging Spectroradiometer (MODIS) and Landsat to generate a time series of fusion images at 30 m resolution, and then the vegetation indices (VIs) related to greenness and moisture content of the rice canopy were calculated to create the time-series of VIs. Second, phenological metrics were extracted from the time-series data of VIs, and a feature selection scheme was designed to acquire an optimal phenological metric subset. Finally, an ensemble model with optimal phenological metrics as classification features was built using random forest (RF) and gradient boosting (GB) classifiers, and the classification of stress levels was implemented. The results demonstrated that the overall accuracy of discrimination for different stress levels is greater than 98%. This study suggests that fusion images can be utilized to detect heavy metal stress in rice, and the proposed method may be applicable to classify stress levels.

Highlights

The present situation of heavy metal pollution in China is severe, which poses a serious threat to agricultural product quality, and harms human health, and leads to social instability [1]
A method for accurately monitoring rice under heavy metal stress at the region scale was proposed (Figure 2). It focuses on the following procedures: (i) calculate vegetation indices and create time series; (ii) design phenological parameters based on changes in rice phenology under heavy metal stress using Matlab and Tsfresh package; (iii) select an optimal phenological feature subset from the original feature set, and establish an ensemble model based on machine learning algorithms to classify heavy metal stress levels in rice; and, (iv) evaluate the accuracy of classification results
We aasatisfactory performance in in classifying heavy metal stress levels in rice, rice, which can Weobtained obtained satisfactory performance classifying heavy metal stress levels in rice, which can be partly attributed to the effectiveness of spatio-temporal fusion images for detecting rice stress be partly attributed to the effectiveness of spatio-temporal fusion images for detecting rice stress can be partly attributed to the effectiveness of spatio-temporal fusion images for detecting rice stress andselection the selection of high-quality ricepixels

Summary

Introduction

The present situation of heavy metal pollution in China is severe, which poses a serious threat to agricultural product quality, and harms human health, and leads to social instability [1]. The national soil survey jointly released by the Ministry of Land and Resources and the Ministry of Environmental Protection showed that the over-standard rate of heavy metals in cultivated land was 19.4%. More than 10 Mt of grain are lost every year in China due to heavy metal pollution [2,3]. China has identified the prevention and control of heavy metal pollution in cultivated land as Sensors 2018, 18, 4425; doi:10.3390/s18124425 www.mdpi.com/journal/sensors. National Quality Standard Pollution Level None Moderate Severe. The quality standard of soil environment is used to evaluate pollution levels.

Methods

Results

Discussion

Conclusion