Extraction Of Wheat Area By GF-1and Landsat-8 In Changji

Remote sensing technology has long been a core part of space technology, it has advantage of high amount of information, macroscopic view, objectivity and effectiveness, makes it become the irreplaceable monitoring and management technical means in agricultural, forestry, land and resources sectors. Accurate estimates of crop planting area related to the production forecast, the market price and other important practical livelihood issues. The traditional area estimation method is time-consuming, labor-intensive, consumption funds, and the result does not has high accuracy because of affected by human factors, can't meet the management needs. At present, for most of the researches of recognition wheat area are based on the Landsat series of satellites, ERS-1 satellite and on the SPOT series satellite data. However, the researches of wheat area extraction based on GF-1 remote sensing satellite, which is developed independently by China, are to be exploration of.This research is based on GF-1 and Landsat-8 remote sensing image data, and field collected wheat, corn, cotton and other crop sample data of Changji City, where is set as the study area. At the same time, we comparative analyzed the optimal data source, the optimal identification phase and the best solutions to extract Changji City wheat area by using single-phase, multi-temporal NDVI sequence remote sensing data and five common supervised classifications. The results showed that:1) The wheat production of precision accuracy and user interpretation accuracy of GF-1 satellite are higher than Landsat-8 Satellite, however, Landsat-8 overall classification accuracy is slightly better than the Gf-1 satellite. Therefore, if the purpose is to study the interpretation of the spatial distribution of crops, it is recommended to use Lansat-8 satellite data which has better data quality and stability. In the situation of identifying the wheat acreage in Changji City, it is recommended to use GF-1 data which has higher interpretation accuracy.2) Single-phase image interpretation, May is the best phase to identify wheat area. Multi-temporal NDVI time series, which has higher interpretation of the overall precision, wheat production accuracy and user accuracy than single-phase data, is the preferred method of interpretation of wheat area.3) If the Changji City wheat area extraction is based on single-phase data, maximum likelihood supervised classification method is the best method. The maximum likelihood method and clustering method are the minimum best practices if the Changji City wheat area extraction is based on multi-temporal NDVI series data. Overall, the maximum likelihood method is the most accurate method of supervised classifications.4) In order to achieve the highest wheat area recognition accuracy, the best model of Changji City wheat area extraction is based on the GF-1 16m multispectral data, NDVI time series utilize temporal, and select the maximum likelihood or minimum distance classifier Crop Monitoring. The extraction area of wheat Changji City of 2015 is 12486 ha. Compared with the reported statistics of the wheat area, wheat area extraction accuracy is 91.8%.