Disjoint Camera Track Matching by an Illumination Effects Reduction and Major Colour Spectrum Histograms Representation Algorithms

In this paper we present a disjoint camera track matching algorithm based on a “cumulative colour histogram transformation”, the “major colour” histograms matching and the post-matching integration algorithms. In order to reduce the “effects of variable illuminations” in disjoint camera environment, a cumulative colour histogram transformation is applied to located moving object image first. Then, the Major Colour Spectrum Histogram Representation is introduced to represent a moving object in a single frame by its most frequent colours only. After that, a two-directional similarity measurement based on the MCSHR is proposed to measure the similarity of any two given moving objects in single frames. Finally, our track matching algorithm extends the singleframe matching along the objects’ tracks by a post-matching integration algorithm. Experimental results presented in this paper show that the unknown illumination effects on moving objects in disjoint camera environment have been reduced significantly by using the proposed cumulative colour histogram transformation algorithm, the proposed similarity measurement algorithm can measure the similarity of the two moving objects accurately, and the post-matching integration proves able to make track matching more robust and reliable.