This document summarizes a presentation on single image super resolution using fuzzy deep convolutional networks. It introduces the problem of super resolution and conventional approaches like manifold learning and dictionary learning. It then presents a proposed approach using a fuzzy deep convolutional network that incorporates a fuzzy rule layer into a convolutional neural network structure. This allows for task-driven feature learning while preserving spatial coherence. Experimental results show the proposed approach achieves better quantitative measures of PSNR, SSIM, and FSIM compared to methods like bicubic interpolation and SRCNN for magnification factors of 3. The findings conclude the method can better preserve structural information in the high resolution image with better visual quality while avoiding additional overhead during learning.