A Study on Image Categorization Techniques

Simple item page
dc.creatorRenu
dc.creatorPrincy
dc.creatorBhatia, Kirti
dc.creatorSharma, Rohini
dc.date2023-05-31
dc.date.accessioned2023-08-21T08:03:04Z
dc.date.available2023-08-21T08:03:04Z
dc.descriptionImage segmentation is the act of splitting a picture into meaningful and non-overlapping parts. It is a crucial step in comprehending natural scenes and has become a hotbed of research in the fields of image processing and computer vision. Even after decades of work and several successes, feature extraction and model design remain difficult. In this article, we carefully review the development in image segmentation techniques. Three crucial stages of image segmentation—classical segmentation, collaborative segmentation, and semantic segmentation based on deep learning—are primarily examined in accordance with segmentation principles and image data characteristics. We compare, contrast, and briefly discuss the benefits and drawbacks of segmentation models as well as their applicability. We also elaborate on the primary algorithms and critical strategies in each stage.en-US
dc.formatapplication/pdf
dc.identifierhttps://journals.researchparks.org/index.php/IJOT/article/view/4438
dc.identifier.urihttp://dspace.umsida.ac.id/handle/123456789/16706
dc.languageeng
dc.publisherResearch Parks Publishing LLCen-US
dc.relationhttps://journals.researchparks.org/index.php/IJOT/article/view/4438/4159
dc.sourceInternational Journal on Orange Technologies; Vol. 5 No. 5 (2023): International Journal on Orange Technologies; 164-170en-US
dc.source2615-8140
dc.source2615-7071
dc.subjectImage Segmentationen-US
dc.subjectClusteringen-US
dc.subjectNeural networken-US
dc.subjectEdge Detectionen-US
dc.titleA Study on Image Categorization Techniquesen-US
dc.typeinfo:eu-repo/semantics/article
dc.typeinfo:eu-repo/semantics/publishedVersion
dc.typePeer-reviewed Articleen-US
Files
Collections
International Journal on Orange Technologies