Near-duplicate keyframe detection based on gray-scale pyramid

The exponential growth of online videos imposes urgent demands on near-duplicate video detection. Near-duplicate keyframe detection is at the base of near-duplicate video detection. In this paper, based on the analysis of the state-of-the-art of near-duplicate keyframe detection, GrayScale Pyramid (GSP) is processed for improving the global feature of color histogram. By constructing the spatial pyramid of luminance and cutting scale, the algorithm enhances illumination and cutting scale transform robustness of global feature. The experiments show that the GSP is more robust than color histogram in detection of illumination transformed and scale transformed near-duplicate keyframe, while for the other transformation, the validity of them is almost the same.