| Popis: |
Degraded Character Recognition (DCR) is an important area of research in the field of Document Image Analysis and Recognition (DIAR). The degradation of characters poses lot of challenges like broken characters, characters mixed with noise etc. Kannada language script has curves and complex patterns, which makes recognizing these characters very difficult. The degradation in the document can introduce gaps in these patterns, which complicates the recognition problem. In this paper, the importance of Kannada DCR system for printed scripts is addressed and also it proposes a framework consisting of four stages namely preprocessing, rebuilding, feature extraction and classification. This framework is also supported with an efficient implementation that achieved a recognition accuracy of 99% in characters with degradation. |
