2024-03-29T07:39:36Z
https://meral.edu.mm/oai
oai:meral.edu.mm:recid/4381
2021-12-13T04:41:57Z
1582963302567:1597824304333
user-ucsy
Content Based Image Classification And Retrieval Using Support Vector Machine
Thinzar, Chu
Tin, Hlaing Htake Kaung
Content-based image retrieval (CBIR) systemsare a special type of Information Retrieval (IR) systemwhere the elements in the repository are pictures. IRworks with finding digital resources in large databases.CBIR is developed to retrieve the desired target imagefrom the large collection of images based on thecontents of the given query image. The Contents ofimage can be extracted from any images which arespecified color, shape, texture or any othercharacteristics of images. The system derived two typesof different color features - color moment and colorauto-correlogram. The moments of image can be usedto indicate color distribution of an image which can bedescribed as a probability distribution of colors. Tocalculate the spatial correlation of pairs of colordifferent with distance, color correlogram is introducedin the system. Gabor wavelets are used to expresstexture of natural image. The characteristics of Gaborwavelets are similar to those of human visual features.The system firstly created the features vector with thesethree types of features and used to get higher retrievalresults of system. To gives the better result in retrievalof system and classify the query image, Support VectorMachine classifier is used with the combination ofthese visual features.
2019-03
http://hdl.handle.net/20.500.12678/0000004381
https://meral.edu.mm/records/4381