Certificate: View Certificate
Published Paper PDF: PDF
Hema Khanna
Independent Researcher
India
ABSTRACT
This manuscript investigates the application of classical image‐processing techniques to optical character recognition (OCR) systems as of 2014. We evaluate preprocessing methods—noise removal, binarization, skew correction—feature extraction approaches such as zoning, projection profiles, and transform‐based descriptors, and classification algorithms including template matching, statistical pattern recognition, and early neural networks. We implement a simulation framework in MATLAB 2013b to process scanned handwritten and printed text samples from publicly available datasets. Statistical analysis of recognition accuracy, precision, recall, and processing time is presented. Simulation results demonstrate that combining adaptive thresholding with zoning‐based features and a multilayer perceptron yields average character recognition rates above 94% for printed text and 87% for handwritten text under moderate noise conditions. Limitations, implications for embedded OCR devices, and directions for research prior to 2015 are discussed.
KEYWORDS
Image processing, OCR, feature extraction, classification, MATLAB, zoning, thresholding, simulation
REFERENCES
- Otsu, N. “A threshold selection method from gray‐level histograms.” IEEE Trans. Syst. Man Cybern., vol. 9, no. 1, pp. 62–66, 1979.
- Sauvola, J. and Pietikäinen, M. “Adaptive document image binarization.” Pattern Recognit., vol. 33, no. 2, pp. 225–236, 2000.
- Serra, J. “Image Analysis and Mathematical Morphology.” Academic Press, 1983.
- Postl, W. “Detection of linear features and skew angle of printed text.” Comput. Graph. Image Process., vol. 13, no. 3, pp. 375–386, 1980.
- Shapiro, L. G. and Stockman, G. C. “Computer Vision.” Prentice-Hall, 2001.
- Hull, J. J. “Document image skew detection: survey and annotated bibliography.” Pattern Anal. Appl., vol. 3, pp. 311–325, 2000.
- Rumelhart, D. E., Hinton, G. E., and Williams, R. J. “Learning internal representations by error propagation.” In Parallel Distributed Processing: Explorations in the Microstructure of Cognition, vol. 1, pp. 318–362, MIT Press, 1986.
- Vapnik, V. and Cortes, C. “Support‐vector networks.” Mach. Learn., vol. 20, no. 3, pp. 273–297, 1995.
- Rabiner, L. R. and Juang, B.-H. “An introduction to hidden Markov models.” IEEE ASSP Mag., vol. 3, no. 1, pp. 4–16, 1986.
