Image restoration based on structure and fruit fly optimization algorithm
Yibo Wang, Yanping Bai, Hao Yan · 2016
Image restoration is one of the hot spots in the image vision research. Criminisi algorithm is a commonly used algorithm. In order to avoid the problem of inaccurate priority calculation caused by confidence dropping rapidly, and considering the variance difference between patches can reflect the image edge structure, we redefine the priority calculation. The fruit fly optimization algorithm is applied to the search process of the best matching patch, which avoids the heavy workload and inaccuracy caused by the global search, and it improves the efficiency and accuracy of the algorithm. By restoring the different damaged images, it can be seen that not only the PSNR values are improved, but also we can get a better restoration result.