Study On Extraction Methods Of Wheat Straws Burning Using HJ-1B IRS Data

Recently, the problem of straw burning is increasingly prominent in China, which leads to serious air pollution, therefore, more emphasis should put on the monitoring and management of straw burning. Moreover, the monitoring of straw burning mainly on manpower and hasn’t completely achieved the informatization management level. Based on the analysis of the related theory and methods of ground fire monitoring by remote sensing, taking Huaian city of Jiangsu Province as the research area, study the methods of monitoring wheat straw burning point using the HJ-1data, the following conclusions are obtained:Analysis of HJ-1IRS image for typical forest fire shows that the fire burning surface brightness temperature is about370K in general, while in the wheat harvest seasons the normal surface brightness temperature is about320K, at the same time the surface brightness temperature changing with the seasons has certain fluctuation.Based on the sampling analysis of typical fire point on the IRS images, brightness temperature growth factor N (3-4), which grows more obvious on the fire point pixel level in the process of the fire point decision is proposed on the basis of the3rd (3.7um) and4th (11.5um) band of IRS data. At the same time, the remote sensing extraction algorithm model of wheat straw incineration, which is used to the extraction of ground fire in Huaian city in June12th,2012, based on HJ-1satellite IRS data is established by the combination of N (3-4) and the existing fire point extraction of remote sensing algorithm. Combined the wheat variation in growth stage and in maturity stage with classification techniques of remote sensing image, the space of wheat can be extracted. Then determine the ground fire point using crop spatial scope, to exclude non straw burning point (industrial and mining hot, grassland and forest fire point, farmland fire point, cloud reflection and solar flares, etc), which is extracted because of the lack of ground properties.The result, obtained by verifying the extraction of ground straw burning point using the data of that at the same time, shows that this method is feasible in the straw burning point extraction. In addition, referring to the comparison result of MODIS Fire Point Product data and extraction results of this paper at the same time, using HJ-1satellite IRS data with the method raised in this paper can get more fire points than that extracted from MODIS data, but they reflect same distribution of ground fire points on the whole.