摘要
根据农作物收割前后的HJ-1B卫星红外相机(IRS)遥感数据特征,提出农作物秸秆焚烧火点提取模式:利用IRS波段3(3.78μm)与波段4(11.36μm)结合两种经典火点探测算法首先提取出所有类型的地面火点;再利用收割前后的IRS波段2(1.72μm)与波段1(0.9μm)建立农作物秸秆地判定依据,从而从所有类型的地面火点中识别出秸秆火点。结合10月6日、10月21日的HJ-1B IRS影像,对10月21日江苏省农作物收割时期的秸秆火点进行提取实验,并与对应的MODIS探测结果进行相关分析,相关系数R2为0.685,验证了研究方法的合理性。
Gas and particulate matter which resulted from straw burning can seriously pollute the atmospheric environment and threaten human health and traffic safety.Based on the HJ-1B IRS data acquired in the non-harvest period and the harvest period,a straw-burning fire detection mode has been established.All types of fire spots during the harvest period were identified using a modified classic fire-detection algorithm and the IRS channel 3 and channel 4 measurements,and then a straw field identification criteria were established based on the channel 2 and channel 1 DN difference（B2-B1） using the two IRS data,in order to identify the straw-burning fires from the fires of all types.On the basis of the two HJ-1B IRS images acquired on Oct.6 and Oct.21,2009,a straw-burning detection experiment in Jiangsu Province on Oct.21,2009,when the crops have entered into the harvest period,was carried out here.Finally,the straw-burning fires detected by IRS were compared with that detected by MODIS,we found that their correlation coefficient R2 is 0.685 2,which validated the rationality and feasibility of the method in this study.
出处
《地理科学》
CSCD
北大核心
2011年第6期661-667,共7页
Scientia Geographica Sinica
基金
全球变化研究国家重大研究计划课题(2010CB951503)
973前期专项课题(2010CB434801)资助
