Research On The Algorithm Of Super Resolution Reconstruction Based On Image Sequence

Since CCD and CMOS image sensors have been widely used, in many fields, the spatial resolution imaging equipment is one of the major technical indicators. However, due to hardware cost and fabrication complexity limitations, how to increase the current resolution level by applying low cost software tools has become a topic of very great interest. Serial number or the use of super-resolution image reconstruction method has now become a hot research.In this paper, for sequences of movement between the images is estimated that the lower quality imaging system model and super-resolution image reconstruction algorithm of the three aspects of research, the campaign is estimated that the current parameters of the more popular optical flow method and thinking of the combination of multi-resolution To improve the accuracy of a Movement estimated parameters. These include a low-resolution images can be super-resolution use of the alias information as well as sub-pixel movement; reconstruction algorithm based on Bayesian methods to select the maximum a posteriori probability(MAP) as a model to achieve super-resolution image reconstruction framework, Markov random field through the introduction of the a priori model of effective control to understand the convergence space. At the same time, Gibbs random field into the model, in addition to the restrictions on Markov Random Field, the full use of pixel point between the relevant information. In enhancing the resolution on the basis of the images to retain more of the details of the message.