Morphological systems for character image processing and recognition

Authors:
Ping-Fai Yang;Petros Maragos
Affiliations:
Division of Applied Sciences, Harvard University, Cambridge, MA;Division of Applied Sciences, Harvard University, Cambridge, MA
Venue:
ICASSP'93 Proceedings of the 1993 IEEE international conference on Acoustics, speech, and signal processing: image and multidimensional signal processing - Volume V
Year:
1993

Citing 6
Cited 1

Adaptive signal processing

Adaptive signal processing
Pattern Spectrum and Multiscale Shape Representation

IEEE Transactions on Pattern Analysis and Machine Intelligence
Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations

Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations
Image Analysis and Mathematical Morphology

Image Analysis and Mathematical Morphology
Fourier Descriptors for Plane Closed Curves

IEEE Transactions on Computers
Structuring element adaptation for morphological filters

Journal of Visual Communication and Image Representation

Granulometries and Opening Trees

Fundamenta Informaticae

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we applied min/max signal operations, common in morphological image analysis, to both feature extraction and classification of character images. We propose a system that computes an improved version of the morphological shape-size histogram, which reduces sensitivity to stroke thickness, size and rotation. For pattern classification we introduce the class of min-mas classifier, which generalizes Boolean DNF functions for real-valued inputs. A Least Mean Square (LMS) algorithm was used for practical training of min-max classifiers. Experimental results show that min-max classifiers were able to achieve error rates that are comparable to neural networks trained using back propagation. The main advantage of the min-max/LMS algorithm is its simplicity and faster speed of convergence.