Enhancing Alzheimer Recognition Through Super-Resolution Images
Pejman Rasti, Nicolas Heudron and Salma Samiei (CERADE, ESAIP, France)
The majority of Alzheimer's Disease (AD) recognition systems rely on images of adequate resolution and quality. In fact, even the most challenging benchmark databases for AD recognition hardly contain low-resolution images. An upsampling method such as bicubic interpolation fails to retrieve all the detailed information that can assist in recognition. As a solution to this problem, we proposed a dictionary-based single image super-resolution algorithm in this paper. Experimental results obtained on the down-sampled version of a subset of the OASIS benchmark database indicate that the proposed system can enhance the performance of a state-of-the-art AD recognition system for handling low-resolution images.
Journal: International Journal of Simulation- Systems, Science and Technology- IJSSST V23
Published: no date/time given