Landcover classification in MRF context using Dempster-Shafer fusion for multisensor imagery

Sarkar, Arnab; Banerjee, Amit; Banerjee, Nairit; Brahma, Sherine; Kartikeyan, B.; Chakraborty, Manab; Majumder, K. L.

doi:10.1109/tip.2005.846032

Cited by 36 publications

(15 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although DS fusion is now well known and widely used in various situations [5,9,11,19,[32][33][34]38], its use in the Markov field context is very rare; only a few papers deal with this kind of models [3,6,10,15,31,36]. To be more precise, we show how the use of TMF allows one to simultaneously integrate the possibly evidential aspects of Image and Vision Computing 24 (2006) the prior information and the possibly evidential aspects of the sensors.…”

Section: Introductionmentioning

confidence: 99%

Multisensor triplet Markov fields and theory of evidence

Pieczynski¹,

Benboudjema²

2006

Image and Vision Computing

View full text Add to dashboard Cite

Section: Introductionmentioning

confidence: 99%

Multisensor triplet Markov fields and theory of evidence

Pieczynski¹,

Benboudjema²

2006

Image and Vision Computing

View full text Add to dashboard Cite

“…[17] Dempster-Shafer data fusion has been utilized in applications such as land cover classification, machine vision, and medical diagnoses. [18][19][20] An example of Dempster-Shafer theory applied to sensors follows: consider a hypothetical sensor system that attempts to identify a sensed object as belonging to one of four object types: A, B, C, or D. The exhaustive and mutually exclusive hypothesis space, or frame of discernment, is represented by the set {A,B,C,D}. The evidence observed from a given sensor response provides support for one or more of these elements including, potentially, supersets combining two or more elements.…”

Section: Data Fusion Architecture Developmentmentioning

confidence: 99%

Intelligent data fusion for wide-area assessment of UXO contamination

Johnson

Minor

Guthrie

et al. 2008

Stoch Environ Res Risk Assess

View full text Add to dashboard Cite

Standard Form 298 (Rev. 8-98)Prescribed by ANSI Std. Z39.18Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS. 5a. CONTRACT NUMBER 5b. GRANT NUMBER 5c. PROGRAM ELEMENT NUMBER 5d. PROJECT NUMBER 5e. TASK NUMBER 5f. WORK UNIT NUMBER Intelligent data fusion techniques are being developed and optimized for use in enhancing wide-area assessment (WAA) for UXO remediation efforts. This report describes the investigations in the second year of project MM-1510 that focused on data fusion. A generalized method for processing input data feature streams from UXO WAA surveys was developed. The method requires the generation of a corresponding georeferenced feature intensity map and the specification of a functional relationship between the feature's intensity and the hypotheses supported by the presence or absence of that feature. The method accommodates diverse feature sets and widely varying evidential relationships as potential inputs for data fusion. Three data fusion theories were investigated: heuristic, Bayesian theoretic and Dempster-Shafer theoretic approaches. Two prototype data fusion frameworks were developed and evaluated with feature sets for the Pueblo and Kirtland sites. Preliminary results obtained with a prototype Dempster-Shafer based data fusion framework agreed well with the limited ground truth data available at the Pueblo site. Objective. The objective of this work is to develop a data fusion framework that will form the basis of a cohesive data management and decision making utility for processing information acquired in the course of performing wide-area surveys of potential UXO remediation sites. This framework will be capable of capturing UXO-related information from all available data and effectively combining this information to provide site-wide assessments of the likelihood of UXO contamination that are more accurate than any single information source on its own. The final data fusion framework is intended to allow site managers to more efficiently direct the expenditure of time, labor and resources in remediation efforts. REPORT TYPE 1. REPORT DATE (DD-MM-YYYY) TITLE AND SUBTITLE AUTHOR(S) PERFORMING ORGANIZATION REPORT NUMBER PERFORMING ORGA...

show abstract

“…Ghimire et al [15] applied random forests, and Stuckens et al [39] employed a linkage-based clustering algorithm for land cover classification using contextual information. Sarkar et al [37] presented a hybrid approach among MRF and Dempster-Shafer theory. It is also valuable to shed light over the seminal work of Wharton [42], which presented a contextual land cover classification approach based on the local distribution of nearby labels, as well as the work of Guo et al [16], which proposed a contextual approach based on cascaded classifiers for remote sensing imagery classification.…”

Section: Introductionmentioning

confidence: 99%

Improving land cover classification through contextual-based optimum-path forest

Osaku

Nakamura²,

Pereira

et al. 2015

Information Sciences

View full text Add to dashboard Cite

a b s t r a c tTraditional machine learning algorithms very often assume statistically independent data samples. However, this is clearly not the case in remote sensing image applications, in which pixels present spatial and/or temporal dependencies. In this work, it has been presented an approach to improve land cover image classification using a contextual approach based on optimum-path forest (OPF) and the well-known Markov random fields (MRFs), hereinafter called OPF-MRF. In addition, it is also introduced a framework to the optimization of the amount of contextual information used by OPF-MRF. Experiments over high-and mediumresolution satellite (CBERS-2B, Landsat 5 TM, Ikonos-2 MS and Geoeye) and radar (ALOS-PALSAR) images covering the area of two Brazilian cities have shown the proposed approach can overcome several shortcomings related to standard OPF classification. In some cases, the proposed approach outperformed traditional OPF in about 9% of recognition rate, which is crucial for land cover classification.

show abstract

Landcover classification in MRF context using Dempster-Shafer fusion for multisensor imagery

Cited by 36 publications

References 32 publications

Multisensor triplet Markov fields and theory of evidence

Multisensor triplet Markov fields and theory of evidence

Intelligent data fusion for wide-area assessment of UXO contamination

Improving land cover classification through contextual-based optimum-path forest

Contact Info

Product

Resources

About