In my diploma thesis, I proposed a new way to enhance the resolution of thermal images by using compressed sensing methods and sparse representations.
MATLAB, Signal Processing Toolbox
This thesis deals with the problem of resolution enhancement (super resolution) of thermal images using compressed sensing methods. We solve the problem of resolution enhancement in four stages. First, we seek for a sparse representation of a low-resolution image with respect to two statistically-learned overcomplete dictionaries (for high and low resolution images respectively). Next, we use the coefficients of this representation to calculate the high resolution image. Finally, we calculate the high resolution image using methods that require multiple low resolution images aligned with subpixel accuracy (conventional approach). We compare the results of each method using well known metrics for reconstruction quality standards.