Title :
Multiple keywords assignment to images using SVM
Author :
Ji, Ye ; Chen, Yan
Author_Institution :
Sch. of Econ. & Manage., Dalian Maritime Univ., Dalian
Abstract :
Use of semantic content is one of the important tasks in image analysis, which needs to be addressed for improving image retrieval effectiveness. We present a method to assign multiple keywords to image using SVMs. Images are divided into three-level regions called global image, semi-global images and sub-images. For each of them, color, texture and edge features are extracted. Then, the trained SVMs are employed and the results of classification are added based on the weight of the levels. The keywords are assigned according to the total of the results. Experiment results show the method is helpful to represent main contents of images.
Keywords :
feature extraction; image retrieval; support vector machines; SVM; edge feature extraction; global image; image analysis; image retrieval; multiple keywords assignment; semantic content; semi-global images; sub-images; three-level regions; Content based retrieval; Cybernetics; Image color analysis; Image edge detection; Image generation; Image retrieval; Image texture analysis; Machine learning; Support vector machine classification; Support vector machines; Classification; Content-based image retrieval; Support vector machines;
Conference_Titel :
Machine Learning and Cybernetics, 2008 International Conference on
Conference_Location :
Kunming
Print_ISBN :
978-1-4244-2095-7
Electronic_ISBN :
978-1-4244-2096-4
DOI :
10.1109/ICMLC.2008.4620841
