摘要
为快速高效地进行云检测和云相态识别,提出了基于BP神经网络的云检测和云相态识别方法,并针对中分辨率成像光谱仪(MODIS)的相关数据建立了BP神经网络。利用该算法对MODIS图像进行了云检测和云相态识别,并把识别结果与MOD06数据进行了对比,对比结果表明,该算法对冰云、水云以及混合云的识别准确率分别达到了100%、100%以及99.94%。该算法快速、准确,消除了未确定态,具有很强的自主学习能力。
In order to process cloud detection and cloud phase retrieval quickly and efficiently,the method based on the BP neural network is put forward.With the related moderate resolution imaging spectroradiometer（MODIS）data,the BP neural network is established.The cloud detection and cloud phase are retrieved based on the algorithm by using the MODIS images,and the recognition results are compared with the data of MOD06.The result indicates that the algorithm accuracy of water cloud,ice cloud,and mixed cloud recognition respectively reach 100%,100% and 99.94%.The algorithm is rapid and accurate and eliminates the uncertain cloud.It also has a strong ability of autonomous learning.
出处
《光学与光电技术》
2016年第5期74-77,共4页
Optics & Optoelectronic Technology
基金
国家公益性行业(气象)科研专项(GYHY201006049)资助项目
关键词
云检测
云相态
混合云
不确定态
神经网络
cloud detection
cloud phase retrieval
mixed cloud
uncertain state
neural network
