Abstract

In view of the slow speed of in painting, and the unsatisfactory restoration for large area shortcomings in traditional image in painting algorithm, this paper proposes a DWT-FMM image in painting method in wavelet domain. According to the features that the main information is mainly focused on the low-frequency component, and that the corresponding position information of each component is in consistent, the algorithm firstly decompose the damaged image via wavelet, and then to repair low-frequency component through fast marching method, meanwhile, the corresponding position Information in painting of other components is achieved, eventually to complete image in painting through wavelet reconstruction. Experiments show that the algorithm can well repair the damaged area and shorten the processing time greatly.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call