A Fuzzy ARTMAP Module for Graphics Symbol Recognition

Author(s): Bortolozzi, F. | Murshed, N. |

Year: 1998

Citation: In Proceedings of the International Joint Conference on Neural Networks, Vol. 3, IEEE Press, 1998, pp 1700-1705.

Abstract: This paper presents a method for recognizing graphics symbols of electronic components in a database of circuit layouts. The method is based on the one-class problem approach on our ability to recognize a 2D-objects without making an explicit decomposition. To satisfy these requirements, a fuzzy ARTMAP recognition module was developed with the objective of recognizing the graphics symbols of 19 electronic components. Each fuzzy ARTMAP was trained with 2D images of graphic symbols of one component only (positive patterns only). The recognition module was then used to search for a specific component in a database of 30 images of circuit layouts. The training and test sets contained respectively, 380 images (2D images/component), and 2051 images (an average of 108 images/component). Experimental results show an average percentage error of 3.49%

Topics: Image Analysis, Applications: Character Recognition, Models: Fuzzy ARTMAP,

PDF download

Cross References

  1. Fuzzy ARTMAP: A neural network architecture for incremental supervised learning of analog multidimensional maps
    A new neural network architecture is introduced for incremental supervised learning of recognition categories and multidimensional maps in response to arbitrary sequences of analog or binary input vectors, which may ... Article Details

  2. Off-line signature verification, without a priori knowledge of class. A new approach
    This work proposes a new approach to signature verification. It is inspired by the human learning and the approach adopted by the expert examiner of signatures, in which an a priori knowledge of the class of forgeries is not ... Article Details