A method of scale conversion for binary image including dithered image

✍ Scribed by Yasuo Kurosu

• Publisher: John Wiley and Sons
• Year: 1998
• Tongue: English
• Weight: 208 KB
• Volume: 29
• Category: Article
• ISSN: 0882-1666
• DOI: 10.1002/(sici)1520-684x(19980615)29:6<28::aid-scj3>3.0.co;2-n

✦ Synopsis

To process a character/picture combined document (including dithered images) in a filing system, high quality scale conversion of documents is required. If a binary document-image including a dithered image is reduced based on its gray scale, its character region blurs. If a dithered image is reduced by coordinate conversion alone, moiré patterns occur in the dithered regions. This paper proposes a method which reduces these two region types separately by discriminating them based on the dispersion of black pixels and the shape of the picture/photograph. Subjective evaluation of the method was carried out by six male markers, by scoring the reduced images on a scale of five grades, the highest being 5. The images processed by the proposed method were evaluated as higher in quality than those by a conventional method by 1.5 points (maximum) and by 0.5 point (mean). The accuracy of area discrimination was also evaluated by using different character sizes. When characters are larger than 5.6 points (font size), the image can be recognized with an accuracy of 100%. Using the proposed method, therefore, binary images including dithered images can be reduced with little deterioration.