Based On Multi-source Satellite Remote Sensing Extraction Of Crop Cultivation Area In Hubei Province And Its Dynamic Monitoring

As a national basic data,the planting area of crop is not only the necessary parameters of crop yield estimation,but also the significant index of national food security. The main approach to obtain the crop area is by a filed measurement of a large area or investigation escalation of statistics department,but these two methods are time-consuming and in lack of spatial information.The fast development of "3S" Spatial Information Technology provides a good high-tech way to monitor crop acreage. Getting real-time and accurate range of crop cultivation is an important guarantee for food production and food security.Our country is the largest producer of rapeseed, canola crops are located in five major advantages of rice, wheat, corn and soybeans after, has become a major source of China’s major cash crops and edible oil. Remote sensing technology has many advantages including objectivity, spatiality, timeliness, rich in content, and low cost; Compared with other methods using remote sensing technology can obtain large-scale surface information under different temporal resolution and spatial resolution. Its advantage for monitoring crop acreage is incomparable.In this paper, the study area is Hubei province which is a major agricultural province in China.Its arable land accounts for about 20% of China’s land area, and 70% cultivated land of the entire province is mainly distributed in Jianghan plain, the east along the jianghan plain and hilly region.It contains both paddy field and dry land,and paddy field is a bit more than dry land.In the southern Hubei,about 70 percent of arable land is paddy field,double cropping is cultivated widely;but in northern and northwest Hube,68.3 percent of cultivated land is dry land, the main crops are canola, wheat, corn, rice etc. Timely grasping the crop planting area and output information has an important significance to Strengthen the management of agricultural production, adjust agricultural structure,and to assist the relevant government departments to formulate scientific and rational agricultural policyThis paper presents detailed introduction about the background and significance of this study,it also instroduces both domestic and international research in detail.By summarizing the results and the problems of previous research,the necessity and feasibility of subject research has been proposed. The key to extract crop acreage is to identify the type of crop accurately, But because the high resolution multispectral remote sensing image revisit cycle is long, and data acquisition is easily affected by the weather,it’s impossible to get the actual acreage crops. The use of high resolution data is difficult due to high cost. So this article chooses the resolution imager (MODIS), environment and disaster mitigation small satellite (HJ-1-A/B), such as high score 1 remote sensing satellite data and combining with the study of crop phenological and other auxiliary method USES A set of data to extract and research crop planting area (mainly rapeseed) on a large scale. and analyze dynamic variation of rapeseed acreage in this 10-year period from 2005 to 2014. Study describes the concrete contents, methods and results in detail(1)MODIS vegetation index (MODIS-NDVI) time series of products that reflect the continuous coverage of vegetation, is an important source of crop sensing measurements. Extraction techniques for the study of the distribution of information based on rapeseed planting MODIS data, select Hubei Province as the study area, the use of the 2009-2010 phase of the 15 MODIS-NDVI time series data, combined with crop phenology and ground survey sample and other auxiliary materials, through the establishment of oilseed rape acreage extraction model, using multiple threshold comparison method to extract the rapeseed planting distribution information Hubei Province in 2010, the overall accuracy of about 85% extraction. Finally, the environment of small satellites HJ-1A CCD data accuracy have proved the MODIS-NDVI time series data and the reliability of this method in the area of cultivation of rapeseed extraction, to master the rapeseed acreage and production information to enhance agricultural production and management, the adjustment of agricultural structure, support government departments to develop a scientific and rational agricultural policies is important.(2) Study on the extraction of the period 2005 to 2014 the Decade of Hubei rapeseed acreage, and its rate of change analysis and dynamic degrees. The results show that:in 2005-2007 in Hubei Province this year canola acreage gradually decline, from 2008 to date rapeseed acreage steadily increasing trend, from a whole decade of rapeseed area is increased.(3) The use of MODIS-EVI data extracted 2010 winter wheat acreage in Hubei Province. By MODIS data products from MODIS-EVI MOD13Q1 extracted data, established in Hubei MODIS-EVI time series data of winter wheat growing season; the author satellite images use high resolution 16m Tianmen been planted winter wheat distribution information via supervised classification images, video reference area with winter wheat was in Hubei MODIS-EVI time series data masking process to get winter wheat in Hubei reference MODIS-EVI data; Finally, using MODIS-EVI data unmixing and spectral analysis method to extract the entire Hubei winter wheat-growing province of distribution information. Correlation Analysis of Hubei Province,10 counties in the area of remote sensing and statistical winter wheat winter wheat area were, R2=0.85.(4) The use of MODIS reflectance data further calculations with vegetation, water, soil associated with a variety of remote sensing indices, such as the Normalized Difference Vegetation Index (NDVI), normalized water index (NDWI), normalized soil index (NDSI). Timing data will be different after the growing season (113 days to 273 days) of each index superposition principle component analysis, obtained by the first principal component of each planting index, and the first principal component of the superposition of three indices, the Figure superimposed use SVM classifier to get dry and paddy fields of information, and finally the use of corn and soybean NDVI special period on the basis of differences distinguish upland corn and soybeans, resulting in a fine crop of Hubei Province in 2014 classification status.