Automated pseudo-coloring of grayscale images based on contourlet transform

Colorizing of grayscale images is a frequent and a widely used phenomenon in image processing. The approach presented in the paper is an automated one, consisting of a pair of sequential steps; reference image selection from the database followed by pixel coloring using luminance and neighbourhood statistics matching algorithm. The efficiency of the proposed technique is highlighted by the use of an effective retrieval technique, Content Based Image Retrieval (CBIR), in which the inherent directional features of the images are well captured. The proposed technique has been compared with other three relevant works using a subjective measure called Mean Opinion Score (MOS). The simulation results along with the comparison performed; given in results and discussion section, show the superiority of the proposed method over the existing ones.