eduzhai > Applied Sciences > Engineering >

Cloud detection in Landsat-8 imagery in Google Earth Engine based on a deep neural network

  • Save

... pages left unread,continue reading

Document pages: 17 pages

Abstract: Google Earth Engine (GEE) provides a convenient platform for applicationsbased on optical satellite imagery of large areas. With such data sets, thedetection of cloud is often a necessary prerequisite step. Recently, deeplearning-based cloud detection methods have shown their potential for clouddetection but they can only be applied locally, leading to inefficient datadownloading time and storage problems. This letter proposes a method todirectly perform cloud detection in Landsat-8 imagery in GEE based on deeplearning (DeepGEE-CD). A deep neural network (DNN) was first trained locally,and then the trained DNN was deployed in the JavaScript client of GEE. Anexperiment was undertaken to validate the proposed method with a set ofLandsat-8 images and the results show that DeepGEE-CD outperformed the widelyused function of mask (Fmask) algorithm. The proposed DeepGEE-CD approach canaccurately detect cloud in Landsat-8 imagery without downloading it, making ita promising method for routine cloud detection of Landsat-8 imagery in GEE.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×