Single Image Resolution Enhancement

Executive Summary

Super-resolution is the process of increasing the resolution of a low-resolution image.  Manifold techniques have been used for super-resolution to enhance image, photo, or video quality, but they are often computationally demanding due to the complexity of the underlying algorithm.  MSU researchers have developed a novel method for single image super-resolution, requiring little computational complexity and meeting highest quality standards at a significantly lower cost.

Description of Technology

This technology is an algorithm for linear approximation of a lower dimension manifold using sparse subspace clustering for the purpose of super-resolution or the up-scaling of a single image.  Our technology yields comparable if not better image enhancement results than current technique with significantly less computational requirements, thus lowering computational cost.

Key Benefits

• Fast – less computational time processing images
• Low cost – very cost-effective due to less computational complexity and shorter processing time

Applications

• Image Super Resolution – cameras, etc.
• High definition TV – upscaling video, lower resolution content
• Image/Video quality enhancement
• Medical Imaging – tomography, MRI etc.
• Satellite imaging

IP Status: 

Software copyright

Licensing Rights Available

Full licensing available

Inventors: Chinh Dang, Hayder Radha

Tech ID: TEC2014-0006