Block-Based Sparse Integral Histogram with Interpolation

Integral Histogram is widely used among various real-time algorithms due to its computationally efficient facility to compute histograms from an image. However, its advantage is greatly diminished for algorithms requiring only a few histograms since required initialization operation takes up majority of total running time. In this paper we propose a block-based integral histogram that drastically reduces initialization time, a major weakness of integral histograms, and suggest interpolation methods to generate more reliable histogram for the location whose histogram is not available. Experimental results show that the proposed method is fast and accurate to apply real-time applications.