摘要
以整个中国区域为例,采用3种检测方法对Spot vegetation S10影像上存在的云和雪盖的检测结果进行了对比分析。结果表明,BISE检测器得到的云层和实际最为接近,用Spot vegetation状态地图得到的云层过少,而用Spot vegetation检测器V2.0得到的云和雪盖像素有部分重叠。最后利用云、雪盖像素在时间序列上的相邻像素进行平滑处理方法达到去云和雪盖的目的,研究结果对于Spot vegetation S10影像的噪音消除和应用精度的提高具有借鉴意义。
Taking China as the test area, the three kinds of methods for detecting the cloud and snow on Spot Vegetation S10 images are compared. According to the comparison, BISE is the best for detecting cloud among the three; the cloud extracted by the status map （SM） is too little; cloud and the ice and snow extracted by the detector V2.0 overlap each other. The averaging of timely neighboring （before and after the cloud pixels） cloudy （or ice and snow）-free pixels is adopted to remove cloud, ice, and snow. The result is valuable for noise reduction and application accuracy improvement of Spot Vegetation S10.
出处
《水土保持通报》
CSCD
北大核心
2010年第2期236-238,共3页
Bulletin of Soil and Water Conservation
基金
黄土高原土壤侵蚀与旱地农业国家重点实验室基金(10501-201)
国家重点基础研究发展计划(973计划)项目"中国主要水蚀区土壤侵蚀过程与调控研究"(2007CB407203)
国家文物局文物保护科学和技术研究课题"基于多源遥感的古墓葬遗址信息提取研究:以汉五陵原遗址为例"(20070107)
