Analysis of image informativeness measures

Vila, Marius; Bardera, Anton; Feixas, Miquel; Bekaert, Philippe; Sbert, Mateu

doi:10.1109/icip.2014.7025216

Cited by 6 publications

(3 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Let us define JS as JS = {JS( r 1 , q 1 ), JS( r 2 , q 2 ), ...JS( r j , q j ), ...JS( r J , q J )}, i.e., the vector of JS divergences for all channels for a single tile. JS divergence was previously reported in several publications ,− as a useful tool in image processing problems. To the author’s knowledge, JS divergence was not used in the difference assessment between two chromatographic data sets, which implies the innovative aspect of the method.…”

Section: Theorymentioning

confidence: 88%

“…The same holds for H 1 . Although the assumption of independence might be unsuited (especially if high fragmentation occurs), it has been demonstrated that it works in practice in many contexts in which this assumption is also strong (e.g., Bayesian e-mail spam filters). This assumption of independence is defined as constructing a naïve Bayes model, and it is broadly used.…”

Section: Theorymentioning

confidence: 99%

See 1 more Smart Citation

Use of Bayesian Statistics for Pairwise Comparison of Megavariate Data Sets: Extracting Meaningful Differences between GCxGC-MS Chromatograms Using Jensen–Shannon Divergence

Barcaru

Vivó-Truyols

2016

Anal. Chem.

View full text Add to dashboard Cite

A new method for comparison of GCxGC-MS is proposed. The method is aimed at spotting the differences between two GCxGC-MS injections, in order to highlight the differences between two samples, in order to flag differences in composition, or to spot compounds only present in one of the samples. The method is based on application of the Jensen-Shannon divergence (JS) analysis combined with Bayesian hypothesis testing. In order to make the method robust against misalignment in both time dimensions, a moving-window approach is proposed. Using a Bayesian framework, we provide a probabilistic visual map (i.e., log likelihood ratio map) of the significant differences between two data sets consequently excluding the deterministic (i.e., "yes" or "no") decision. We proved this approach to be a versatile tool in GCxGC-MS data analysis, especially when the differences are embedded inside a complex matrix. We tested the approach to spot contamination of diesel samples.

show abstract

Section: Theorymentioning

confidence: 88%

Section: Theorymentioning

confidence: 99%

Use of Bayesian Statistics for Pairwise Comparison of Megavariate Data Sets: Extracting Meaningful Differences between GCxGC-MS Chromatograms Using Jensen–Shannon Divergence

Barcaru

Vivó-Truyols

2016

Anal. Chem.

View full text Add to dashboard Cite

show abstract

“…The entropy rate of an image can measure the contrast, as proposed by Vila et. al [65]. Different from these two evaluations, mutual information reflects the consistency of the fused image with the source images.…”

Section: Objective Evaluationmentioning

confidence: 99%

Information-Theoretic Channel for Multi-exposure Image Fusion

Hao

Zhao

Sbert

et al. 2021

The Computer Journal

View full text Add to dashboard Cite

Multi-exposure image fusion has emerged as an increasingly important and interesting research topic in information fusion. It aims at producing an image with high quality by fusing a set of differently exposed images. In this article, we present a pixel-level method for multi-exposure image fusion based on an information-theoretic approach. In our scheme, an information channel between two source images is used to compute the Rényi entropy associated with each pixel in one image with respect to the other image and hence to produce the weight maps for the source images. Since direct weight-averaging of the source images introduce unpleasing artifacts, we employ Laplacian multi-scale fusion. Based on this pyramid scheme, images at every scale are fused by weight maps, and a final fused image is inversely reconstructed. Multi-exposure image fusion with the proposed method is easy to construct and implement and can deliver, in less than a second for a set of three input images of size 512$\times $340, competitive and compelling results versus state-of-art methods through visual comparison and objective evaluation.

show abstract