DCT Sign Index for Example-Based Super-Resolution

Abstract

The purpose of this study is to propose a new index generation method of the database for the example-based super-resolution. In the example-based super-resolution, the high frequency components of the local blocks are registered in the database. The conventional method had a problem that it takes a long time to search for optimal high-frequency component of the local block from the database. So, the proposed method aims to accelerate the database search by generating an index expressing gray level pattern from DCT signs of the local block. First, the low-resolution input image is expanded to the target resolution using bicubic interpolation. This enlarged image lacks high frequency components. Then, local blocks are extracted from the enlarged image, and the unique index is generated from these local blocks. The optimal high frequency components for the local block are then searched for from the database using the generated index. Finally, the enlarged image is converted into a high definition image by synthesizing the searched optimal high frequency component to the local blocks. As a result, a high-definition enlarged image can be obtained at faster speeds using the proposed method than with the conventional method. In conclusion, we have demonstrated that the DCT sign index can adequately express gray level patterns in local blocks and is effective in speeding up example-based super-resolution.