Colour Map Classification for Archive Documents
This paper presents our work on color map classification for archive documents. An approach is proposed which is very similar to the way that humans perceive document colors, by using the diversity quantization algorithm applied within the HSV colour space. Template color maps are manually registered for sample document images; images from batches to be classified are then mapped onto the closest template color map using a fuzzy color classification algorithm. The results of testing this approach on batches of archive index cards from the UK Natural History Museum were very encouraging. We tested over 400 biological specimen index cards, and achieved more than 98% correct color classification.
- 1.He, J., Downton, A.C.: User-as si steel archive document image analysis for digital library construction. In: Proc. 7th lnt.Conf.on Document Analysis and Recognition, lCDAR 2003, vol. 1, pp. 498–502 (2003)Google Scholar
- 2.Zadeh, L.A., Kacprayk, J.: Fuzzy Logic for the Management of Uncertainty. John-Wiley, New York (1992)Google Scholar
- 3.Dubois, D., Prade, H., Yager, R.Y.: Fuzzy Sets for Intelligent Systems. Morgan Kaufmann Publishers, San Mateo (1993)Google Scholar
- 4.Bradley, J.: XV,Interactive Image Display for the X Window System, Version 3.10a 1053 Floyd Terrace. Bryn Mawr, PA 19010, USA (1994)Google Scholar
- 5.Hildcbrand, L., Reusch, B.: Fuzzy Color Processing. In: Fuzzy Techniques in Image Processing, Physica-Verlag, Heidelberg (2000)Google Scholar
- 7.Foley, J.D., Dam, A.V., Feiner, S.K., Hughes, J.F.: Computer graphics: principles and pratice, 2nd edn. Addison-Wesley, Reading (1996)Google Scholar