Extraction Of Paddy Rice Planting Area Based On Environmental Satellite Image

Paddy rice is China’s first major grain crops, and rice planting area in China accounts for about18.5%that of the world, and China’s rice output occupies the first in the world, which accounts for about29.1%of the world’s total rice production. So mastering the rice planting area promptly and accurately is of great importance to help guide China’s grain production and ensure the food security. Remote sensing technology with the characteristics of timely, accurately, objectively provides an effective means to monitor the rice planting area.The research of rice spectrum characteristic of Jianghan Plain, uses the field spectrometer by measuring the canopy spectral reflectance data, calculating the reflectivity of the different breeds during the different period, to analyze the growth and spectrum characteristics during different period and determine the optimum waveband and temporal for identification. The spectrum shows more difference in the fist reflection peak and reflection dip, which means the reflectivity increases successively in the rice breeding period, sprouting period, milky ripeness period and ripeness period. Rice seedling is the best time for early planting acreage monitoring while the breeding and sprouting period is the best time for identifying the growth situation.Environmental satellite is a kind of potential data sources for extracting the planting area of rice and other crops, because it can revisit and monitor shortly and widely. This paper takes Jianli Country and its adjacent area, located on the hinterland of Jianghan Plain, as the study area. According to the rice phonological calendar, it selects multi-temporal HJ-1A/1B images in critical period of rice growth including transplanting stage and heading and seeding stage, choosing Land Short Water Indices (LSWI) is sensitivity to the increased surface moisture, so it uses LSWI to distinguish cotton plants and construction land in paddy rice’s transplanting period; There are large differences between Normalized Vegetation Indices (NDVI) of water and plant, Then chooses NDVI in paddy rice’s heading and seeding stage to distinguish water;The time variation of NDVI of forest with paddy rice and lotus root is different, and paddy rice’s Shortwave infrared reflectance(RIRS-B2) is smaller than lotus root’s, so the study uses DNDVI[Difference Normalized vegetation index and RIRS-B2to differentiate paddy rice from forest and lotus root.It obtains a method of extracting the planting area of rice based on the HJ satellite data. Then taking SPOT images as reference, the area accuracy and sample point accuracy are not less than90%, and Kappa is not less than0.80, so this proposed method can extract the rice planting cultivation exactly.The research of the rice planting area extraction based on CART decision tree. It choosing the HJ-CCD image of the sprouting rice, analyses the index of RVI, PVI, NDVI, MSAVI2and NDWI, combing the textural features to extract the rice planting area. And after examination the area accuracy and sample point accuracy are not less than80%.