【摘要】 合成孔径雷达(SAR)数据对于南方多云多雨天气的地表农作物类型的探测具有独特的优势。以江苏省海安县为例,基于多极化SAR数据,包括双极化ALOS PALSAR以及全极化Radarsat-2数据,采用面向对象的方法,针对当地水稻/旱田进行识别。针对双极化SAR数据,利用了其强度信息进行分类识别;而基于全极化数据,除强度信息外,还利用了其SAR信号统计分布概率进行分类规则建立。结果表明:L波段的ALOS PALSAR在识别旱地的桑树方面具有很大的优势,而基于两种分类方法的C波段Radarsat-2数据识别水稻的精度分别为85%和75%,略低于ALOSPALSAR的识别结果(87.5%)。更多还原

【Abstract】 Synthetic Aperture Radar(SAR) data has the advantage in detecting the crop types in South China where has the frequent cloudy and rainy days.Based on object-based method,this study used the multi-polarization satellite SAR data,including the dual-polarization ALOS PALSAR and polarimetric Radarsat-2 data,to discriminate the rice and dry-land crop in Haian county,Jiangsu Province.For dual-polarization SAR data,the intensity information was used.For polarimetric SAR data,besides the intensity-information-based classification,the statistica-properties-based method was also applied.The result shows that the L-band ALOS PALSAR outperforms the C-band Radatasat-2 data in discriminating the dry-land mulberry.Moreover,the classification accuracy from ALOS PALSAR based method(87.5%) is also higher than the other two results from Radarsat-2 data based methods(75% and 85% respectively).更多还原