Digital filtering for image processing.
Abstract
The use of filtering for correctly distinguishing between the subjects, background and noise of a digital image is examined. The image improvement techniques of gray scale modification, geometrical distortion correction and smoothing are surveyed. The principle features of Wiener filters and filters which make use of constrained matrix deconvolutions and circulant matrices are presented, and it is pointed out that the use of circulant matrices results in considerable time and memory savings in the restoration of a one dimensional image. As an example, the improvement techniques described are applied to the recognition and classification of photographic images of vegetable chromosomes.
- Publication:
-
Memorie della Societa Astronomica Italiana
- Pub Date:
- September 1979
- Bibcode:
- 1979MmSAI..50..483A
- Keywords:
-
- Digital Filters;
- Image Enhancement;
- Image Processing;
- Data Smoothing;
- Histograms;
- Image Resolution;
- Mode (Statistics);
- Random Variables;
- Thresholds;
- Instrumentation and Photography;
- Image Processing