Comparative Study of Neural Networks and Non Parametric Statistical Methods for Off-Line Handwritten Character Recognition

Idan, Yizhak; Auger, J.-M.; Darbel, Nicolas; Sales, María Á.; Chevallier, Raymond; Dorizzi, Bernadette; Cazuguel, Guy

doi:10.1016/b978-0-444-89488-5.50167-6

Cited by 10 publications

(2 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The clustering effect is presented in figure 3. The supervision, the "on-line" labeling and the addition of an output layer, permit LASSO to present significantly better performances for optical character recognition compared to equivalent topological maps (-.20% improvement, Idan et al 1992). Detailed description of the adaptation of LASSO for OCR application may be found in Idan&Chevallier (1991 The left part depicis the WEK weights arriving to each neuron on an 1 1 by 11 map from a 16 by 16 pixels input.…”

Section: Lassomentioning

confidence: 98%

<title>Pattern recognition by cooperating neural networks</title>

Idan

Auger²

1992

SPIE Proceedings

View full text Add to dashboard Cite

The study of connectionist models for pattern recognition is mainly motivated by their presumed simultaneous feature selection and classification. Character recognition is a common test case to illustrate the feature extraction and classification characteristics of neural networks. Most of the variability concerning size and rotation can be handled easily, while acquisition conditions are usually controlled. Many examples of neural character recognition applications were presented where the most successful results for optical character recognition (OCR) with image inputs were reported on a layered network (LeCun et al. 1990) integrating feature selection and invariance notions introduced earlier in Neocognitron networks. Previously, we have presented a supervised learning algorithm, based on Kohonen's self-organizing feature maps, and its applications to image and speech processing . From pattern recognition point of view, the first network performs local feature extraction while the second does a global statistical template matching. We describe these models and their comparative results when applied to a common French handwritten zip-code database. We discuss possible cooperation schemes and show that the performances obtained by these networks working in parallel exceed those of the networks working separately. We conclude by the possible extensions of this work for automatic document processing systems.

show abstract

Section: Lassomentioning

confidence: 98%

<title>Pattern recognition by cooperating neural networks</title>

Idan

Auger²

1992

SPIE Proceedings

View full text Add to dashboard Cite

show abstract

“…There is no guarantee that a particular number of S and C-planes will be sufficient to absorb the feature information necessary for a particular recognition task. An attempt to circumvent this problem, using constructive techniques, has been reported [169] but that approach has not proved to be a great success [88]. While such redundancy may be vital in a biological vision system (to cope with the inevitable damage or destruction of neurons), it is unnecessary in an artificial setting and does not enhance the network's ability to recognize images.…”

Section: Hea 3nmentioning

confidence: 99%

The neocognitron as a system for handwritten character recognition : limitations and improvements

Lovell¹

View full text Add to dashboard Cite

This thesis is about the neocognitron, a neural network that was proposed by Fukushima in 1979. Inspired by Hubel and Wiesel's serial model of processing in the visual cortex, the neocognitron was initially intended as a self-organizing model of vision, however, we are concerned with the supervised version of the network, put forward by Fukushima in 1983. Through "training with a teacher", Fukushima hoped to obtain a character recognition system that was tolerant of shifts and deformations in input images. Until now though, it has not been clear whether Fukushima's approach has resulted in a network that can rival the performance of other recognition systems.In the first three chapters of this thesis, the biological basis, operational principles SHOP serves as a method for probing the behaviour of the neocognitron and is used to investigate the effect of cell masks, skeletonization of input data and choice of training patterns on the network*s performance. Even though SHOP is the best selectivity adjustment algorithm to be described to date, the system's peak correct recognition rate (for isolated ZIP code digits from the CEDAR database) is around 75% (with 75% reliability) after SHOP training. It is clear that the neocognitron, as originally described by Fukushima, is unable to match the performance of today's most accurate digit recognition systems which typically achieve 90% correct recognition with near 100% reliability.After observing the neocognitron's failure to exploit the distinguishing features of different kinds of digits in its classification of images, Chapter 6 proposes modifications to enhance the networks ability in this regard. Using this new architecture, a correct clcissification rate of 84.62% (with 96.36% reliability) was obtained on CEDAR ZIP codes, a substantial improvement but still a level of performance that is somewhat less than state-of-the-art recognition rates. Chapter 6 concludes with a critical review of the hierarchical feature extraction paradigm.The final chapter summarizes the material presented in this thesis and draws the significant findings together in a series of conclusions. In addition to the investigation of the neocognitron, this thesis also contains a derivation of statistical bounds on the errors that arise in multilayer feedforward networks as a result of weight perturbation (Appendix E).

show abstract