期刊名称：International Journal of Electronics and Computer Science Engineering
电子版ISSN：2277-1956
出版年度：2014
卷号：3
期号：2
页码：86-90
出版社：Buldanshahr : IJECSE
摘要：Orthodontics image retrieval is a very important job for aiding doctors to diagnose orthodontic diseases by providing successful treatment method in their history. Patients, ordinarily having nervous sense and no grasp about their current ill situation and therefore being afraid to get dealt with by doctors, can get confidence on their curing if they learn how their diseases were treated and what the result is after treatment. The paper uses three different methods to retrieve target image in image library and to compare their effects in order to decide which is best. The first method is to extract color feature by combining the cumulative histograms achieving image segmentation by combining Canny operator and Ostu threshold method. The second is to extract seven rectangular-shape features from spatial information. The third one is to apply multi-features by combining color component and shape component. It finds out that image retrieval result of the third one is the bets for orthodontics images by experiments
关键词：Image retrieval; Cumulative histograms; Image segmentation; Moment invariant
