Cloud and Snow Segmentation in Satellite Images Using an Encoder–Decoder Deep Convolutional Neural Networks

The segmentation of cloud and snow in satellite images is a key step for subsequent image analysis, interpretation, and other applications. In this paper, a cloud and snow segmentation method based on a deep convolutional neural network (DCNN) with enhanced encoder–decoder architecture—ED-CNN—is pro...

Full description

Saved in:
Bibliographic Details
Published in:ISPRS international journal of geo-information Vol. 10; no. 7; p. 462
Main Authors: Zheng, Kai, Li, Jiansheng, Ding, Lei, Yang, Jianfeng, Zhang, Xucheng, Zhang, Xun
Format: Journal Article
Language:English
Published: Basel MDPI AG 01.07.2021
Subjects:
Accuracy
Artificial neural networks
Atmospheric aerosols
Cameras
cloud and snow detection
Clouds
CNN
Coders
data collection
Datasets
Deep learning
encoder–decoder
Experiments
Image analysis
Image processing
Image segmentation
Labels
Machine learning
Methods
Neural networks
Principal components analysis
satellite image
Satellite imagery
Satellites
semantic segmentation
Semantics
Snow
Spaceborne remote sensing
spatial data
TH-1
ISSN:2220-9964, 2220-9964
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The segmentation of cloud and snow in satellite images is a key step for subsequent image analysis, interpretation, and other applications. In this paper, a cloud and snow segmentation method based on a deep convolutional neural network (DCNN) with enhanced encoder–decoder architecture—ED-CNN—is proposed. In this method, the atrous spatial pyramid pooling (ASPP) module is used to enhance the encoder, while the decoder is enhanced with the fusion of features from different stages of the encoder, which improves the segmentation accuracy. Comparative experiments show that the proposed method is superior to DeepLabV3+ with Xception and ResNet50. Additionally, a rough-labeled dataset containing 23,520 images and fine-labeled data consisting of 310 images from the TH-1 satellite are created, where we studied the relationship between the quality and quantity of labels and the performance of cloud and snow segmentation. Through experiments on the same network with different datasets, we found that the cloud and snow segmentation performance is related more closely to the quantity of labels rather than their quality. Namely, under the same labeling consumption, using rough-labeled images only performs better than rough-labeled images plus 10% fine-labeled images.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
content type line 23
ISSN:2220-9964
2220-9964
DOI:10.3390/ijgi10070462