Study On Remote Sensing Information Extraction Of Paddy Rice Planting Area

Extracting rice information with remote sensing is important contents that the remote sensing technology was used in the field of agriculture, it is the important component of the resource remote sensing too. Remote sensing technology as front technology of the geo-information science, it is the most effective observing technology and the means of obtaining information at present and has objective, prompt characteristics. It can obtain the ground information on large-scale in succession in a short time and realize collecting fast and analyzing quantitative for agricultural information. So remote sensing technology provides new opportunity for agricultural modernize management, and it is the basic means with scientific decision for agriculture China has carried on agricultural monitor work with remote sensing since 1980s, and has made great progress now.However, rice information was often extracted based on NOAA/AVHRR data in the past year. For restriction of AVHRR data, its monitoring accuracy needs further improvement in rice monitor. Terra MODIS is a new remote sensing sensor. It provides new opportunity for paddy rice planting area extraction on large-scale. This thesis sets out to research how to make full use of MODIS data for rice information monitor, and develops new methods and models for extracting paddy rice information according to MODIS images on large-scaleThe detailed study content and results were presented in the following sections.(1) Based on remote sensing knowledge and hyper-spectral characteristic of the vegetables on the land surface, combined with SPOT-5 satellite images data of the specific characteristics, analysis of the spectral and vegetation feature of rice and other background, according to layer decision tree identification rice information model, and it realized to extract paddy rice planting area information with remote sensing automatically.(2) According to the different paddy rice growth stages,multi-temporal MODIS images were selected by the flooding and transplanting period,growing period,maturation period,and the fallow period after harvest.There is a paddy rice mapping algorithm that uses time series of three vegetation indices (LSWI, EVI, and NDVI) derived from MODIS images to identify that initial period of flooding and transplanting in paddy rice field, base on the sensitivity of LSWI to the increased surface moisture during the period of flooding and rice transplanting. The algorithm was run to map paddy rice information in Heilongjiang and Zhejiang provinces, using 8-day composite MODIS Surface Reflectance produces.(3) The resultant MODIS-derived paddy rice map was compared to SPOT/TM data and land use map and statistical yearbook data at the provincial and county levels. The results showed that the overall pixel accuracy of paddy rice aera is more than 95.6% at Heilongjiang Province, and exceeded 85% in the plains area of Zhejiang. Therefore, it could be applied to monitor paddy rice information with MODIS image data at large spatial scales, and provide information services for the government departments.