基于ENVI ZOOM面向对象的FY3A云提取方法研究

Study on the Extraction Method of the FY3A Cloud Based on ENVI ZOOM Object-Oriented

利用风云三号A星MERSI数据,基于ENVI ZOOM软件平台采用面向对象的多尺度影像分割技术,并结合监督分类技术提取云层边缘线,继而对影像进行反演处理时剔除云层覆盖区域的干扰。研究表明,与同时段彩色卫星云图进行视觉对比,分类结果较理想,该分类方法易于操作,可有效提高解译精度。

Using the MERSI data of the FY 3A meteorological satellite , based on the ENVI ZOOM software platform and object -orien-ted multi-scale image segmentation technology , combine with supervised classification techniques to extract the clouds edge line , then processes the image inversion cloud coverage area excluding the interference .Studies show that：with the same period for visual contrast color satellite imagery , the classification results have more satisfactory , and this classification method is easy to operate , ef-fectively improving the interpretation accuracy .