A Comparative Study of Digital Interpolation Techniques.
Abstract
Interpolation has come into use in digital picture processing for many reasons. This paper is a report on the implementation and evaluation of four methods - replication, linear, cubic spline and sinc function. The evaluation is based on mean square error calculations, visual rating and computer usage costs. The results are limited but show that the best method depends upon the application for which it is used. Two side investigations were also conducted. The mean square error of an interpolated picture can be reduced if the picture is optimally sampled before interpolation. The results of an experiment bear this out. An interpolation method was extended to color pictures. The method proved feasible. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Nov 01, 1977
- Accession Number
- ADA050321
Entities
People
- Edward A. Haser
Organizations
- University of Maryland