| Summary: | 碩士 === 元智大學 === 資訊工程學系 === 92 === Due to a large amount of float-point operations are utilized, the performance achieved by adopting the traditional image enlarging methods is quite low. The problem becomes more serious for the bilinear interpolation method, which is one of the most commonly used image techniques for enlarging pictures. In the paper, we propose a novel method to reduce significantly the processing time for the enlarging techniques. The concept is quite simple, and the method is easy to implement. The float-point operations are replaced by integer operations, so the processing speed is increased. Experimental results show the superiority of the proposed method.
|
|---|
