Face Image Super-Resolution Technology

Research in face image always is one of the hot spot research questions which in the image process, the patter recognition and computer vision domain. Because of the affection of device and other disturbance, the face resolution obtained is often very slow, and it brings the human face recognition and other application great difficulty. Image super resolution technique offers effective solution for this puzzle. By using a number of noisy, blurred low resolution images, it can produce a high resolution image through the signal-processing technology. And it is widely applied in many important areas, such as video, imaging, remote sensing, medical, surveillance and military etc.We mainly study the single face image super resolution techniques, proposing a non-linear face image super resolution algorithm based on semantic constraint patch-based Markov Network. The specific content of the thesis as follows:Firstly, this paper reviews the basic concept and methods of image super resolution technology, and describes learning-based image super resolution technology in detail. Then, By Using Markov Network as learning Model, this paper studies the special kind of face image. Considering the local feature of face image, we combine patch position constraint operation based on aligned face image with the semantic characteristics of patch position, and proposes a non-linear searching algorithm which decreases the complexity of the searching operation and increases the effect of matching and speeds up the convergence of the Markov network. After the matched high-resolution patches are collected, the algorithm uses these patches to dispose in vision compatibility, and directly output relative patches as restituted face image. The experimental result shows that the proposed algorithm has a good output quality, high efficiency and some practical value.