Image Colorization Models And Their Algorithms Based On Coupled Total Variation

Image colorization refers to the process of colorizing grayscale images.Image colorization technology has been used in many fields such as medical and aerospace industries.Therefore,the study of image colorization is a significant topic.The research in this thesis is based on the coupled total variation model of image colorization.The main research contents and innovations are as follows:1.Briefly introducing the coupled total variation image colorization model.On this basis,a new image colorization model is proposed by adding control functions.The fast algorithm of the model is given by the alternating direction multiplier method(ADMM),and the convergence of the algorithm is proved.The numerical experiments show that the colorization quality of this model is good and the simulation operation time is short.2.Based on the coupled total variation image colorization model,combined with the fidelity term of Le denoising model,a new image colorization model is proposed.The model is quickly solved using the ADMM algorithm.The numerical experiment results show that the model has shorter colorization time and better colorization effect.3.Based on the coupled total variation image colorization model,combined with the fidelity term of the denoising model proposed by Krissian et al.,a new image colorization model is proposed.The model is solved using the Split Bregman algorithm.The experimental results show that the colorization effect of the model is better.