Machine learning approaches for high-resolution urban land cover classification: [A comparative study]

Ranga Raju Vatsavai; Chandola Varun; Anil Cheriyadat; Eddie Bright; Bhaduri Budhendra; Jordan Grasser

doi:10.1145/1999320.1999331

Machine learning approaches for high-resolution urban land cover classification: [A comparative study]

Ranga Raju Vatsavai
, Chandola Varun
, Anil Cheriyadat
, Eddie Bright
, Bhaduri Budhendra
, Jordan Grasser

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

9 Scopus citations

Abstract

The proliferation of several machine learning approaches makes it difficult to identify a suitable classification technique for analyzing high-resolution remote sensing images. In this study, ten classification techniques were compared from five broad machine learning categories. Surprisingly, the performance of simple statistical classification schemes like maximum likelihood and Logistic regression over complex and recent techniques is very close. Given that these two classifiers require little input from the user, they should still be considered for most classification tasks. Multiple classifier systems is a good choice if the resources permit.

Original language	English
Title of host publication	COM.Geo 2011 - 2nd International Conference on Computing for Geospatial Research and Applications
DOIs	https://doi.org/10.1145/1999320.1999331
State	Published - 2011
Event	2nd International Conference on Computing for Geospatial Research and Applications, COM.Geo 2011 - Washington, DC, United States Duration: May 23 2011 → May 25 2011

Publication series

Name	ACM International Conference Proceeding Series

Conference

Conference	2nd International Conference on Computing for Geospatial Research and Applications, COM.Geo 2011
Country/Territory	United States
City	Washington, DC
Period	05/23/11 → 05/25/11

Keywords

Bayesian classification
MCS
Neural networks
Trees

Access to Document

10.1145/1999320.1999331

Cite this

Vatsavai, R. R., Varun, C., Cheriyadat, A., Bright, E., Budhendra, B., & Grasser, J. (2011). Machine learning approaches for high-resolution urban land cover classification: [A comparative study]. In COM.Geo 2011 - 2nd International Conference on Computing for Geospatial Research and Applications Article 11 (ACM International Conference Proceeding Series). https://doi.org/10.1145/1999320.1999331

@inproceedings{75a8a3f407b940c9a52a25c3db727b2e,

title = "Machine learning approaches for high-resolution urban land cover classification: [A comparative study]",

abstract = "The proliferation of several machine learning approaches makes it difficult to identify a suitable classification technique for analyzing high-resolution remote sensing images. In this study, ten classification techniques were compared from five broad machine learning categories. Surprisingly, the performance of simple statistical classification schemes like maximum likelihood and Logistic regression over complex and recent techniques is very close. Given that these two classifiers require little input from the user, they should still be considered for most classification tasks. Multiple classifier systems is a good choice if the resources permit.",

keywords = "Bayesian classification, MCS, Neural networks, Trees",

author = "Vatsavai, \{Ranga Raju\} and Chandola Varun and Anil Cheriyadat and Eddie Bright and Bhaduri Budhendra and Jordan Grasser",

year = "2011",

doi = "10.1145/1999320.1999331",

language = "English",

isbn = "9781450306812",

series = "ACM International Conference Proceeding Series",

booktitle = "COM.Geo 2011 - 2nd International Conference on Computing for Geospatial Research and Applications",

note = "2nd International Conference on Computing for Geospatial Research and Applications, COM.Geo 2011 ; Conference date: 23-05-2011 Through 25-05-2011",

}

Vatsavai, RR, Varun, C, Cheriyadat, A, Bright, E, Budhendra, B & Grasser, J 2011, Machine learning approaches for high-resolution urban land cover classification: [A comparative study]. in COM.Geo 2011 - 2nd International Conference on Computing for Geospatial Research and Applications., 11, ACM International Conference Proceeding Series, 2nd International Conference on Computing for Geospatial Research and Applications, COM.Geo 2011, Washington, DC, United States, 05/23/11. https://doi.org/10.1145/1999320.1999331

Machine learning approaches for high-resolution urban land cover classification: [A comparative study]. / Vatsavai, Ranga Raju; Varun, Chandola; Cheriyadat, Anil et al.
COM.Geo 2011 - 2nd International Conference on Computing for Geospatial Research and Applications. 2011. 11 (ACM International Conference Proceeding Series).

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

TY - GEN

T1 - Machine learning approaches for high-resolution urban land cover classification

T2 - 2nd International Conference on Computing for Geospatial Research and Applications, COM.Geo 2011

AU - Vatsavai, Ranga Raju

AU - Varun, Chandola

AU - Cheriyadat, Anil

AU - Bright, Eddie

AU - Budhendra, Bhaduri

AU - Grasser, Jordan

PY - 2011

Y1 - 2011

N2 - The proliferation of several machine learning approaches makes it difficult to identify a suitable classification technique for analyzing high-resolution remote sensing images. In this study, ten classification techniques were compared from five broad machine learning categories. Surprisingly, the performance of simple statistical classification schemes like maximum likelihood and Logistic regression over complex and recent techniques is very close. Given that these two classifiers require little input from the user, they should still be considered for most classification tasks. Multiple classifier systems is a good choice if the resources permit.

AB - The proliferation of several machine learning approaches makes it difficult to identify a suitable classification technique for analyzing high-resolution remote sensing images. In this study, ten classification techniques were compared from five broad machine learning categories. Surprisingly, the performance of simple statistical classification schemes like maximum likelihood and Logistic regression over complex and recent techniques is very close. Given that these two classifiers require little input from the user, they should still be considered for most classification tasks. Multiple classifier systems is a good choice if the resources permit.

KW - Bayesian classification

KW - MCS

KW - Neural networks

KW - Trees

UR - https://www.scopus.com/pages/publications/79960089324

U2 - 10.1145/1999320.1999331

DO - 10.1145/1999320.1999331

M3 - Conference contribution

AN - SCOPUS:79960089324

SN - 9781450306812

T3 - ACM International Conference Proceeding Series

BT - COM.Geo 2011 - 2nd International Conference on Computing for Geospatial Research and Applications

Y2 - 23 May 2011 through 25 May 2011

ER -

Machine learning approaches for high-resolution urban land cover classification: [A comparative study]

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this