Feature extraction and evaluation using edge histogram descriptor in MPEG-7

Abstract

According to the definition of the edge histogram descriptor (EHD) in MPEG-7, one can easily generate an extra histogram bin from the 5-bin local edge histogram of each 4 x 4 sub-image. This extra histogram bin defines the ratio of the non-edge area (i.e., monotonous region) in the sub-image. Forming a feature vector with 6 edge/non-edge types, we can generate 33 different feature vectors (or 33 x 6 = 198 feature elements) including 16 vectors from 4x4 sub-images, 1 vector from a global histogram, 13 vectors from semi-global histograms, 1 vector from entropy, and 2 vectors from centers of gravity. A statistical hypothesis testing is employed to see which feature vectors/elements are most informative to differentiate different image classes. Experimental results show that non-edge and entropy features are the most informative features among all 33/198 feature vectors/elements. © Springer-Verlag 2004.