Detecting the Optimal Seam Line Automatically in Image Mosaic with Twin Snakes Model
Weili Jiao · Science Technology and Engineering · 2007
A new approach based on twin snakes model was presented to detect an optimal seam line automatically in image mosaic. The color similarity and texture similarity between the images are formulated in the energy function of snakes. With this method an optimal seam line can be found automatically which is less differences in color and texture between the images. In the process of selection and iteration of the snakes the difference information of the images in color and texture is used. The seam line detected by the twin snakes model is more superior to that detected by normal method. Finally, a good result with this method was demonstrated in the experiment.