Enhancing Image Thresholding with Masi Entropy: An Empirical Approach to Parameter Selection

dc.contributor.authorHegazy, Eslam
dc.contributor.authorGabr, Mohamed
dc.contributor.editorSkala, Václav
dc.date.accessioned2025-07-30T09:41:00Z
dc.date.available2025-07-30T09:41:00Z
dc.date.issued2025
dc.description.abstract-translatedImage multilevel thresholding is used as a preprocessing step in computer vision or image processing applications. In recent years, various methods and criteria have been tested to improve thresholding performance and efficiency. State-of-the-art methods combine a metaheuristic optimization algorithm to optimize some objective function. Commonly used objective functions include Otsu criterion and entropy-based methods. Masi entropy is one of the objective functions that recently showed significant potential in image thresholding. It can deal with additivity and non expandability of information. However, in order for applying Masi entropy, there is a parameter r that needs to be tuned. Different works proposed different values and methods to determine the value of r. Nevertheless, there is a lack of validation and comparisons between these methods. This work is an experimental study validating a previously introduced method for selecting r. We compare the performance over two datasets including 700 various images while varying the number of thresholds from 1 to 15. Structural similarity index (SSIM) and peak signal-to-noise ratio (PSNR) metrics are used to evaluate the performance. The collective and individual performance improvements are elaborated. It is shown that the tested method achieves a significant improvement in segmentation quality over all tested numbers of thresholds. Higher numbers of thresholds showed greater improvement than smaller numbers. These results demonstrate the practical utility of the tested method in entropy-based image multilevel thresholding.en
dc.format8 s.cs
dc.format.mimetypeapplication/pdf
dc.identifier.doihttp://www.doi.org/10.24132/CSRN.2025-16
dc.identifier.issn2464-4617 (Print)
dc.identifier.issn2464-4625 (online)
dc.identifier.urihttp://hdl.handle.net/11025/62222
dc.language.isoenen
dc.publisherVaclav Skala - UNION Agencyen
dc.rights© Vaclav Skala - UNION Agencyen
dc.rights.accessopenAccessen
dc.subjectvíceúrovňové prahovánícs
dc.subjectsegmentace obrazucs
dc.subjectmasi entropiecs
dc.subjectvýběr parametrůcs
dc.subjectneřízená klasifikacecs
dc.subject.translatedmultilevel thresholdingen
dc.subject.translatedimage segmentationen
dc.subject.translatedmasi entropyen
dc.subject.translatedparameter selectionen
dc.subject.translatedunsupervised classificationen
dc.titleEnhancing Image Thresholding with Masi Entropy: An Empirical Approach to Parameter Selectionen
dc.typekonferenční příspěvekcs
dc.typeconferenceObjecten
dc.type.statusPeer revieweden
dc.type.versionpublishedVersionen
local.files.count1*
local.files.size2079074*
local.has.filesyes*

Files

Original bundle
Showing 1 - 1 out of 1 results
No Thumbnail Available
Name:
B97.pdf
Size:
1.98 MB
Format:
Adobe Portable Document Format
License bundle
Showing 1 - 1 out of 1 results
No Thumbnail Available
Name:
license.txt
Size:
1.71 KB
Format:
Item-specific license agreed upon to submission
Description: