A. Pujol and L. Chen (France)
Image Processing, Low level features, Hough Transform
While the problem of Content Based Image Retrieval
(CBIR) and automated image indexing systems has been
widely studied in the past years they still represent a chal
lenging research ﬁeld. Indeed capturing high level seman
tics from digital images basing on low level basic descrip
tors remains an issue. A review of existing systems shows
that edge descriptors are among the most popular features.
While color features have led to extensive work, edge fea
tures haven’t produced such active research and most cur
rent systems rather rely on completing basic edge informa
tion with other, more computationally expensive features
such as texture. In this paper we propose to work on a
more accurate edge feature while keeping a relatively low
computation cost. We will begin with a review of common
edge features used in CBIR and automated indexing sys
tems, we will then explain our Enhanced Fast Hough Trans
form algorithm and the edge descriptor we derived from it.
Through a study of computational complexity, we will ex
plain that the computational burden is kept minimal and ex
perimental results using a sample automated indexing sys
tem will show that our new edge feature signiﬁcantly im
proves over more traditional descriptors.