QUT ePrints

A comparison of various emerging techniques for digital classification of urban environment

Pathak, Virendra (2008) A comparison of various emerging techniques for digital classification of urban environment. In Queensland Spatial Conference 2008 : Global Warning: What’s Happening In Paradise?, 17-19 July 2008, Gold Coast, Qld, Australia.

View at publisher

Abstract

A comparison of various emerging digital classification approaches for classification of urban environment has been carried out in this study. It has been done on two study sites in the state of U.P. in India. These sites represent good examples of rapidly changing urban environment in developing countries. The following alternative approaches were used for the digital classification: Back propagation Artificial Neural Network (BPANN), Classification with wavelet derived texture features and the Per-field classification approach. The satellite data from LISS-III sensors on board IRS-1C satellite was used for the study. Results from these approaches were compared with the conventional Gaussian Maximum classification (GML) classification approach. It was observed that the classifications results using BPANN approach were similar to or slightly better than GML classification. Resilient propagation (RPROP) method of BPANN was the best and robust method in comparison to other BPANN methods considered for the study. Investigations were also carried out to explore significance of spatial properties in the form of texture features. These features were derived using various techniques including wavelet-based approach. Results showed that classification accuracies using texture features show significant improvement over pure spectral classification. A novel global threshold based region growing segmentation method the 'Per-field classification' was also implemented for urban classification. This approach also showed significant improvement over the per-pixel GML classification approach.

Impact and interest:

Citation countsare sourced monthly from Scopus and Web of Science® citation databases.

These databases contain citations from different subsets of available publications and different time periods and thus the citation count from each is usually different. Some works are not in either database and no count is displayed. Scopus includes citations from articles published in 1996 onwards, and Web of Science® generally from 1980 onwards.

Citations counts from the Google Scholar™ indexing service can be viewed at the linked Google Scholar™ search.

Full-text downloads:

176 since deposited on 11 Feb 2009
28 in the past twelve months

Full-text downloadsdisplays the total number of times this work’s files (e.g., a PDF) have been downloaded from QUT ePrints as well as the number of downloads in the previous 365 days. The count includes downloads for all files if a work has more than one.

ID Code: 17692
Item Type: Conference Paper
Keywords: BPANN, RPROP, GML, Segmentation, Texture
ISBN: 9780958136679
Subjects: Australian and New Zealand Standard Research Classification > ENGINEERING (090000) > GEOMATIC ENGINEERING (090900) > Photogrammetry and Remote Sensing (090905)
Divisions: Past > QUT Faculties & Divisions > Faculty of Built Environment and Engineering
Past > Schools > School of Urban Development
Copyright Owner: Copyright 2008 (please consult author)
Copyright Statement: This paper is available under License Creative Commons Attribution Share Alike: http://creativecommons.org/licenses/by-sa/2.5/
Deposited On: 12 Feb 2009 08:53
Last Modified: 09 Jun 2010 23:20

Export: EndNote | Dublin Core | BibTeX

Repository Staff Only: item control page