A Rapid Model (COV_PSDI) for Winter Wheat Mapping in Fallow Rotation Area Using MODIS NDVI Time-Series Satellite Observations: The Case of the Heilonggang Region

Abstract

Rapid and accurate monitoring of spatial distribution patterns of winter wheat over a long period is of great significance for crop yield prediction and farmland water consumption estimation. However, weather conditions and relatively long revisit cycles often result in an insufficient number of continuous medium-high resolution images over large areas for many years. In addition, the cropland pattern changes frequently in the fallow rotation area. A novel rapid mapping model for winter wheat based on the normalized difference vegetation index (NDVI) time-series coefficient of variation (NDVI_COVfp) and peak-slope difference index (PSDI) is proposed in this study. NDVI_COVfp uses the time-series index volatility to distinguish cultivated land from background land-cover types. PSDI combines the key growth stages of winter wheat phenology and special bimodal characteristics, substantially reducing the impact of abandoned land and other crops. Taking the Heilonggang as an example, this study carried out a rapid mapping of winter wheat for four consecutive years (2014–2017), and compared the proposed COV_PSDI with two state-of-the-art methods and traditional methods (the Spectral Angle Mapping (SAM) and the Iterative Self-Organizing Data Analysis Technique Algorithm (ISODATA)). The verification results revealed that the COV_PSDI model improved the overall accuracy (94.10%) by 4% compared with the two state-of-art methods (90.80%, 89.00%) and two traditional methods (90.70%, 87.70%). User accuracy was the highest, which was 93.74%. Compared with the other four methods, the percentage error (PE) of COV_PSDI for four years was the lowest in the same year, with the minimum variation range of PE being 1.6–3.6%. The other methods resulted in serious overestimation. This demonstrated the effectiveness and stability of the method proposed in the rapid and accurate extraction of winter wheat in a large area of fallow crop rotation region. Our study provides insight for remote sensing monitoring of spatiotemporal patterns of winter wheat and evaluation of “fallow rotation” policy implementation.

Wheat crop genotype and age prediction using machine learning with multispectral radiometer sensor data

“…Raoufi et al (2018) showed that the model could be used to predict rice yields. Zhang et al (2021) used MODIS NDVI time-series satellite data to distinguish winter wheat from other crops. The Heilongjiang region was chosen for winter wheat mapping over four consecutive years (2014)(2015)(2016)(2017).…”

Machine Learning-Based Fine Classification of Agricultural Crops in the Cross-Border Basin of the Heilongjiang River between China and Russia

“…Using the minimum distance method, the Classification and Regression Tree (CART) classifier, and the random forest (RF) classifier, he extracted spatial information for soybeans, maize, and rice in Jilin Province. Zhang et al [13], employing the NDVI, the time-series coefficient of variation (NDVI_COVfp), and the Peak Slope Difference Index (PSDI), proposed a new rapid mapping model for winter wheat in Heilonggang, Hebei Province.…”

Walnut Acreage Extraction and Growth Monitoring Based on the NDVI Time Series and Google Earth Engine

“…(2) The twi-difference algorithm is used to classify and extract different crops by identifying crop pixels with obvious crests [7]. Zhang X et al [8] proposed identifying and extracting the winter wheat planting area in the Heilongjiang region from 2014 to 2017. Their work was based on MODIS remote sensing imagery, a technique based on the normalized difference vegetation index (NDVI) and time series coefficient of variation (NDVI-CV) combined with the NDVI curve features of different objects on the ground and second-order differencing (3) The supervised classification method: in this method, after classifying different ground objects in remote sensing images, the spatial distribution information of different crops is extracted with the use of decision trees [9], support vector machines [10,11], random forests [12][13][14], etc.…”