Abstract:
The conventional histogram intersection (HI) algorithm
computes the intersected section of the corresponding
color histograms in order to measure the
matching rate between two color images. Since this
algorithm is strictly based on the matching between
bins of identical colors, the final matching rate can be
easily affected by color variation caused by various
environment changes. In this paper, a Gaussian
weighted histogram intersection (GWHI) algorithm is
proposed to facilitate the histogram matching via taking
into account matching of both identical and similar
colors. The weight is determined by the distance between
two colors. The algorithm is applied to license
plate classification. Experimental results show that the
proposed algorithm produces a much lower intra-class
distance and a much higher inter-class distance than
previous HI algorithms for tested images which are
captured under various illumination conditions.
