Panoramic image generation with image mosaicing

An image mosaic is a panoramic image generated by assembling sequential images belong to the same scene. This method is mainly used in mapping applications, the evaluation of the images acquired from unmanned aerial vehicles, and various computer vision applications. In this study, a novel blending method is proposed for automatic and seamless mosaicing of photos taken by an unprofessional digital camera. For the correction of consistent seams occurred during the mosaicing, the color differences around the overlapping regions are measured. For this purpose, first, the intensity values of image pixels inside and outside the overlapping region boundary in a 1 pixel wide zone are determined. Secondly, the intensity values are averaged and finally, the average values of color differences are applied to overlapping regions. Besides, motion blur filtering is performed for the corresponding boundary regions in order to reduce the seams. In order to test the accuracy of the proposed method a subjective evaluation metric is used. According to the results obtained, the seamless image mosaics are generated with a subjective accuracy of 0.66 in average where the full scale is between 0 (the worst) and 1 (the best). The proposed method is appeared to be more successful (by 0.20 points) than the 'Feathering' method found in the literature. The results of study indicate that the seams, which occurs after mosaicing, can be substantially reduced or completely eliminated. Besides, it is observed that the illumination differences are minimized.