A global motion characterization method using one-dimensional characteristic curves for image processing applications

A very hardware and computation efficient method for frame motion characterization is developed. Instead of dealing a 2-dimensional m/spl times/n picture elements, this algorithm generates two 1-dimensional vectors, by mathematically operating the luminance values of vertical and horizontal lines as the characteristic valves of x and y direction, respectively. The efficiency of this algorithm is demonstrated by decomposition of frame changes into various factors: the motion of objects, panning and zooming. Applications in automatic motion tracking system and image stabilizer are discussed.