Bulletin Autumn‧Winter 1998

It is a content-based i ma ge database system s u p p o r t i ng content-based retrieval of d i g i t al images b y colour, texture, sketch, a n d shape w h i c h runs o n the W i n d o ws N T / 9 5 p l a t f o rm. It is also an image management system w h i c h allows users to catalogue, query, a nd manage their i ma ge assets efficiently a n d effectively. Its usefulness lies i n its c a p a b i l i ty to search the database for images w i t h a p p r o x i ma t e ly similar user-specified features. F i g u r e 2 i l l u s t r a t es s o me o f t h e se functions: (a) retrieving the p a i n t i ng i n Figure 1 u s i n g the c o l o ur s k e t ch p a d p l u g - i n, (b) r e t r i e v i ng f a s h i on i ma g es u s i n g the c o l o ur s i m i l a r i ty p l u g - i n, a n d (c) r e t r i e v i ng fabric patterns u s i ng the classification tree tool. The d e s i gn of M o n t a ge is based o n an o p en system architecture, w h i c h ma k es the system flexible, m o d u l a r, a nd extensible. It includes a p l u g - in f r a m e w o rk that makes the c u s t o m i z a t i on of a p l u g - i n m o d u l e f o r a specific a p p l i c a t i on easy. T h is a l l o ws users to extend the f u n c t i on of the Mo n t a ge system to suit their o w n needs. How Does Montage Work? F i g u r e 3 s h o w s t h e t h r e e m a j o r c o m p o n e n ts of M o n t a g e 's c o n t e n t - b a s ed search engine. T he first c o m p o n e nt deals w i t h i ma ge p r o c e s s i ng a n d p a t t e rn r e c o g n i t i o n. T h is i n c l u d es i m a g e a n a l y s i s, i m a g e manipulation, a n d feature extraction. Feature extraction of an i ma ge comes i n different forms. Currently, Mo n t a ge contains four general purpose p l u g - in mo d u l es w h i c h process colour similarity, textile texture, colour s k e t c h, a n d p o l y g o n s h a pe i n f o r m a t i o n. Research is b e i n g d o n e to d e v e l op b e t t er o p t i m i z ed basic p l u g - i ns as w e l l as m o r e specialized p l u g - i ns to be u s ed i n d i f f e r e nt applications, for example, for face recognition. T h e s e c o nd c o m p o n e n t d e a ls w i t h advanced database technology w h i c h tries to o p t i m i z e the storage a n d o r g a n i z a t i on of Figure 2: Montage snapshots (a)Colour Sketch It allows users to draw a pattern with specific colours. Montage will then search similar matches from the database. (b) Colour Similarity Users can search similar images from a database based on the overall impression of colours. (c) Classification Tree A tree-like organization tool which allows users to easily manage and organize their database. Users can search an image or build a complicated query by simple clicks. This reduces search time and avoids the typing mistakes made by traditional keyword search. Montage 31