Pseudo－Colorizing Scheme Using Vector Quantization of Luminance Mapping Techniques

In this paper, we present an effective algorithm for colorizing a grayscale video sequence.There are two phases to be achieved the goal; one is we use transferring colorizing technique to colorize the background scenes. The other is the pseudo-colorizing technique be used to colorize the object. In our approach, a color image, an RGB 2lαβ color transform ( l =luminance,αβ =chrominance), and a block-based Vector Quantization of Luminance Mapping (VQLM) technique are used to automatically colorize the grayscale background scenes. The VQLM technique compares the grayscale scene with the luminance of the color reference to obtain the information of theαβ planes of the grayscale scene. After the chrominance is padded, the inverse color transform, lαβ 2RGB , the grayscale scene was colorized. In addition, we pre-select several colors associated with the pseudo-colorizing (PC) technique to colorize the object of the grayscale video sequences, like the human hair, skin, shirt, and pants, respectively. Compared to the other colorizing schemes, our proposed method has two advantages: 1) The VQLM and reference color image can perfectly colorize the background scenes automatically; 2) The PC and image processing techniques achieve effective colorizing of the object.