Efficient Color Transfer Method Based on Colormap Clustering for Night Vision Applications

Recent night-vision cameras provide multiband images with complementary information which is useful to enable operations at night and in adverse weather conditions. The grayscale fused image is unnatural in appearance and therefore it is difficult to design a reliable intelligent system based on this results. In this paper, efficient natural color transfer method based on color map clustering is proposed for Night Vision applications. The proposed algorithm is a novel and efficient framework to colorize the night vision imagery utilizing color map clustering and cluster recognition based on color similarity. The target color look up table is derived from the set of natural color image database for a specific environment. Proposed method is applied on datasets of different environment and compared with standard color transfer method using objective evaluation parameters to evaluate efficacy of color transfer algorithm. The simulation results show that proposed method enhances the natural color appearance in the resultant image and provides consistent results for various datasets.