Data mining approach for estimating cloud-covered areas in MODIS satellite images

Mahdi Saleh; Sami Serbey; Bouchra Fahs

doi:10.1109/iscc.2017.8024682

Data mining approach for estimating cloud-covered areas in MODIS satellite images

Mahdi Saleh, Sami Serbey, Bouchra Fahs

EEE - Academic & Research

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

Abstract

One of the main remote sensors used for monitoring snow covered areas is the Moderate Resolution Imaging Spectroradiometer (MODIS) employed by NASA's Terra and Aqua satellites. Using MODIS-derived snow cover images is limited under cloud-covered regions due to the sensor's capabilities. This paper presents an automated process based on K-Nearest Neighbor (KNN) algorithm using spatiotemporal features, to estimate the pixel cover for cloud-covered regions. The algorithm was tested using MODIS's daily snow-cover datasets obtained from Terra (MOD10A1) and Aqua (MYD10A1) satellite for the Lebanese territories as a study region. Several experiments were implemented to test the accuracy of the proposed algorithm, and the results were highly acceptable (>90%).

Original language	English
Title of host publication	2017 IEEE Symposium on Computers and Communications (ISCC)
Place of Publication	Heraklion, Greece
Publisher	IEEE
Pages	1164-1167
Number of pages	4
ISBN (Electronic)	9781538616291, 9781538616284
ISBN (Print)	9781538616307
DOIs	https://doi.org/10.1109/iscc.2017.8024682
Publication status	Published - Jul 2017

Publication series

Name	IEEE Symposium on Computers and Communications (ISCC)
Publisher	Institute of Electrical and Electronics Engineers

Access to Document

10.1109/iscc.2017.8024682

Cite this

@inproceedings{966500fb1c904fe8bc3aae2fae2d1dde,

title = "Data mining approach for estimating cloud-covered areas in MODIS satellite images",

abstract = "One of the main remote sensors used for monitoring snow covered areas is the Moderate Resolution Imaging Spectroradiometer (MODIS) employed by NASA's Terra and Aqua satellites. Using MODIS-derived snow cover images is limited under cloud-covered regions due to the sensor's capabilities. This paper presents an automated process based on K-Nearest Neighbor (KNN) algorithm using spatiotemporal features, to estimate the pixel cover for cloud-covered regions. The algorithm was tested using MODIS's daily snow-cover datasets obtained from Terra (MOD10A1) and Aqua (MYD10A1) satellite for the Lebanese territories as a study region. Several experiments were implemented to test the accuracy of the proposed algorithm, and the results were highly acceptable (>90%).",

author = "Mahdi Saleh and Sami Serbey and Bouchra Fahs",

year = "2017",

month = jul,

doi = "10.1109/iscc.2017.8024682",

language = "English",

isbn = "9781538616307",

series = "IEEE Symposium on Computers and Communications (ISCC)",

publisher = "IEEE",

pages = "1164--1167",

booktitle = "2017 IEEE Symposium on Computers and Communications (ISCC)",

address = "United States",

}

Data mining approach for estimating cloud-covered areas in MODIS satellite images. / Saleh, Mahdi; Serbey, Sami; Fahs, Bouchra.
2017 IEEE Symposium on Computers and Communications (ISCC). Heraklion, Greece: IEEE, 2017. p. 1164-1167 (IEEE Symposium on Computers and Communications (ISCC)).

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Data mining approach for estimating cloud-covered areas in MODIS satellite images

AU - Saleh, Mahdi

AU - Serbey, Sami

AU - Fahs, Bouchra

PY - 2017/7

Y1 - 2017/7

N2 - One of the main remote sensors used for monitoring snow covered areas is the Moderate Resolution Imaging Spectroradiometer (MODIS) employed by NASA's Terra and Aqua satellites. Using MODIS-derived snow cover images is limited under cloud-covered regions due to the sensor's capabilities. This paper presents an automated process based on K-Nearest Neighbor (KNN) algorithm using spatiotemporal features, to estimate the pixel cover for cloud-covered regions. The algorithm was tested using MODIS's daily snow-cover datasets obtained from Terra (MOD10A1) and Aqua (MYD10A1) satellite for the Lebanese territories as a study region. Several experiments were implemented to test the accuracy of the proposed algorithm, and the results were highly acceptable (>90%).

AB - One of the main remote sensors used for monitoring snow covered areas is the Moderate Resolution Imaging Spectroradiometer (MODIS) employed by NASA's Terra and Aqua satellites. Using MODIS-derived snow cover images is limited under cloud-covered regions due to the sensor's capabilities. This paper presents an automated process based on K-Nearest Neighbor (KNN) algorithm using spatiotemporal features, to estimate the pixel cover for cloud-covered regions. The algorithm was tested using MODIS's daily snow-cover datasets obtained from Terra (MOD10A1) and Aqua (MYD10A1) satellite for the Lebanese territories as a study region. Several experiments were implemented to test the accuracy of the proposed algorithm, and the results were highly acceptable (>90%).

U2 - 10.1109/iscc.2017.8024682

DO - 10.1109/iscc.2017.8024682

M3 - Conference contribution

SN - 9781538616307

T3 - IEEE Symposium on Computers and Communications (ISCC)

SP - 1164

EP - 1167

BT - 2017 IEEE Symposium on Computers and Communications (ISCC)

PB - IEEE

CY - Heraklion, Greece

ER -

Data mining approach for estimating cloud-covered areas in MODIS satellite images

Abstract

Publication series

Access to Document

Fingerprint

Cite this