Super-resolution For Multi-view Images Of Plenoptic Camera

Plenoptic camera, a new photography equipment based on light field imaging is constructed with internal microlens arrays between main camera lens and sensor to capture both spatial and angular information, i.e. the full four-dimensional radiance, of a scene. It is better than conventional camera. Since the additional information of light field information, and post flexible, the light field data has become a hot topic in image processing field. For the problem of low spatial and angular resolution of the plenoptic camera, this thesis is focus on super-resolution algorithm research for multi-view image of the plenoptic camera.The main work of this thesis is listed as follows:(1) We summarize the background and the research status for plenoptic camera and super-resolution reconstruction technology, and introduce the basic theory of light field imaging, the principles of plenoptic camera, and the relative theory of the super-resolution reconstruction technique.(2) According to the rendering of multi-view image and obtaining of depth information, we introduce and implement the maximum a posteriori light field image super-resolution reconstruction algorithm based on total-variation model. The experimental results show that this algorithm can improve the spatial and angular resolution of multi-view image of plenoptic camera.(3) This thesis designs and implements a edge preserved light field image super-resolution method based on weighted beyond-digital-total-variation model. We use edge residual error function of reconstructed images and different low-resolution multi-view images to weight the data term, and apply beyond-digital-total-variation prior model for denoising and edge preserve. The experimental results show that our approach not only improves the spatial and angular resolution of multi-view image of plenoptic, but also preserves the edge information of reconstruction.