2026-06-01T17:33:57Z https://meral.edu.mm/oai

oai:meral.edu.mm:recid/4284 2021-12-13T06:17:43Z 1582963302567:1597824273898 user-ucsy

Automatic Image Annotation and Retrieval Yu, May The Sein, Myint Myint In this paper, an automatic imageannotation and retrieval model is developed baseon the intensity invariant approach. The givenuncaptioned image is divided into backgroundand foreground images and segmented intoregions, which are classified into region typesusing a variety of features. Firstly, preprocessingstages such as gray-scale converting, noisefiltering for image enhancing is processed. Aftersegmentation, calculate the eigenvectors ofimages and examined the associated word byusing database. The various types of images areapplied for training and testing. The top wordsare described for annotated image in result.Manual image annotation is time-consuming,laborious and expensive; so, there has been alarge amount of research done on automaticimage annotation and retrieval technologies arecombined to improve the performance. 2011-05-05 http://hdl.handle.net/20.500.12678/0000004284 https://meral.edu.mm/records/4284