ترمیم تصاویر رنگی با نواحی مخدوش بزرگ براساس تجزیه تصویر به مولفه های بافت و ساختار

In this paper a new inpainting scheme applicable to images with wide missed regions is presented. The proposed method is based on  a new modification of the well known geodesic path image inpainting algorithm. Furthermore, to achieve a high-quality inpainting preserving details and textures, the image is decomposed firstly into texture and structure sub-images. Then, each sub-image is concealed based on a suitable technique and the final inpainted image is created. To evaluate the effectiveness of the proposed method, a comparative study with two basic methods over 30 images of different contents is performed and the results are reported. The experimental results confirm the effectiveness of the proposed method comparing with other methods, especially in the cases of images with wide missed regions.