Resolution enhancement algorithm based on wavelet and edge extraction techniques in noise presence

The images and video sequences registering in optical, radar, medical applications, presented in digital photographs, on HD TV, in electron microscopy, etc. are obtained from electronic devices that use different sensors [1-3]. The visual quality of the images and frames in the video sequences depend on spatial resolution, and because of the physical limitations this suffers of precision needed that can be improved developing better sensors via manufacturing process that seems as a difficult and high cost task. That is why, in many applications of the image/video processing, the additional methods and algorithms are developed where the goal is to restore the resolution degraded in a sensor, permitting better observations of the fine details, edges, etc. [1, 4, 5]. This can be performed using the super resolution (SR) procedures generating a high-resolution (HR) images from one or several low-resolution (LR) images/video frames [1-6]. The goal of the developed algorithm is to provide better resolution than those obtained by other state-of-the-art filter. A number of approaches have been proposed designing the SR algorithms [1-7]. Among them there are: the nearest neighbor algorithms, the bilinear interpolation, the bi-cubic technique, the fuzzy logic methods and techniques based on the spline technique. Image resolution enhancement using wavelet transform (WT) domain is a relatively new subject, and recently many novel algorithms have been designed [6-9].