Cloud detection for FY-3D MERSI <bold>Ⅱ</bold> images combine radiative transfer simulation and shallow neural network

نویسندگان

چکیده

æœ¬æ–‡ç»“åˆè¾å°„ä¼ è¾“æ¨¡åž‹å’Œæœºå™¨å­¦ä¹ æå‡ºäº†ä¸€ç§ä»ŽFY-3D卫星MERSI â ¡ä¼ æ„Ÿå™¨å ‰å­¦å½±åƒä¸­è¯†åˆ«äº‘åƒå ƒçš„æ–¹æ³•CRMC(Combine Reflectance simulation and Machine learning for Cloud detectionï¼‰ã€‚è¯¥æ–¹æ³•é€šè¿‡è®¾ç½®å˜åŒ–çš„åœ°ç‰©å’Œå¤§æ°”å† åœ¨å ‰å­¦ç‰¹æ€§IOPs(Inherent Optical Properties),达到考虑多种下垫面的二项反射特征和不同大气条件下气溶胶和云参数的目的。CRMCæ–¹æ³•ä¸»è¦åŒ å«3个步骤:(1)通过聚类分析从MODIS二项反射参数产品中分离出11ç§å ¸åž‹ä¸‹åž«é¢åœ°è¡¨åå°„å‚æ•°ï¼›ï¼ˆ2)将随机设置的气溶胶和云参数以及地表反射率参数(即IOPsï¼‰è¾“å ¥SBDARTè¾å°„ä¼ è¾“æ¨¡åž‹ï¼Œå¾—åˆ°æ¨¡æ‹Ÿçš„åå°„çŽ‡å€¼æ•°æ®é›†ï¼Œå¹¶ä»¥æ­¤è®­ç»ƒæµ å±‚ç¥žç»ç½‘ç»œæ¨¡åž‹ï¼›ï¼ˆ3ï¼‰åˆ©ç”¨æµ å±‚ç¥žç»ç½‘ç»œæ¨¡åž‹é€åƒå ƒé¢„æµ‹äº‘æ¦‚çŽ‡ï¼Œå¹¶æ ¹æ®å®žé™ éœ€è¦ç¡®å®šåŒºåˆ†äº‘åƒå ƒå’Œéžäº‘åƒå ƒçš„äº‘æ¦‚çŽ‡é˜ˆå€¼ã€‚é€šè¿‡ä¸ŽCALIPSO垂直特性掩膜产品(VFMï¼‰é€åƒå ƒå¯¹æ¯”éªŒè¯å‘çŽ°ï¼ŒCRMC方法总正确率为79.6%,且在陆地和海面上分别为78.5%和81.2%。通过与MODIS云掩膜产品横向对比(MYD35)发现,当云阈值设定为0.2时,CRMCæ–¹æ³•åœ¨é™†åœ°ï¼Œä¸»è¦æ˜¯é˜”å¶æž—ã€å†œç”°ã€åŸŽå¸‚å’Œè£¸åœŸç­‰åœ°è¡¨ä¸Šçš„äº‘è¯†åˆ«æ•ˆæžœè¾ƒå¥½ï¼Œä½†åœ¨æµ·é¢çš„äº‘è¯†åˆ«æ•ˆæžœä»å¾ è¿›ä¸€æ­¥æé«˜ã€‚

برای دانلود باید عضویت طلایی داشته باشید

برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

A multi-scale convolutional neural network for automatic cloud and cloud shadow detection from Gaofen-1 images

The reconstruction of the information contaminated by cloud and cloud shadow is an important step in pre-processing of high-resolution satellite images. The cloud and cloud shadow automatic segmentation could be the first step in the process of reconstructing the information contaminated by cloud and cloud shadow. This stage is a remarkable challenge due to the relatively inefficient performanc...

متن کامل

An Improved Cloud Classification Algorithm for China’s FY-2C Multi-Channel Images Using Artificial Neural Network

The crowning objective of this research was to identify a better cloud classification method to upgrade the current window-based clustering algorithm used operationally for China's first operational geostationary meteorological satellite FengYun-2C (FY-2C) data. First, the capabilities of six widely-used Artificial Neural Network (ANN) methods are analyzed, together with the comparison of two o...

متن کامل

3D Continuum Radiative Transfer Images of a Molecular Cloud Core Evolution

We analyze a 3D Smoothed Particle Hydrodynamics simulation of an evolving and later collapsing pre-stellar core. Using a 3D Continuum Radiative Transfer program, we generate images at 7, 15, 175 μm, and 1.3 mm for different evolutionary times and viewing angles. We discuss the observability of the properties of pre-stellar cores for the different wavelengths. For examples of non-symmetric fragm...

متن کامل

Radiative Transfer in 3d

The high resolution and sensitivity provided by ALMA will reveal new small scale structures in many sources, e.g. star forming regions. Such inhomogeneities may not have been considered in the analysis of past observations but they will be essential to the understanding of future data. Radiative transfer methods are needed to interpret the observations and the presence of complicated source str...

متن کامل

A Cloud-Based Neural Network Simulation Environment

We present N2Sky, a novel Cloud-based neural network simulation environment. The system implements a transparent environment aiming to enable arbitrary and experienced users to do neural network simulations easily and comfortably. The necessary resources, as CPUcycles, storage space, etc., are provided by using Cloud infrastructure. N2Sky also fosters the exchange of neural network specific kno...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

ژورنال

عنوان ژورنال: Journal of remote sensing

سال: 2022

ISSN: ['1007-4619', '2095-9494']

DOI: https://doi.org/10.11834/jrs.20210406