How Important is Global Structure for Characters?

Abstract

Abstract-This paper studies the importance of the features that represent the global structure of character strokes to character recognition. Most existing character recognition methods based on character stroke features utilize a set or a sequence of local features such as xy-coordinates and local direction of strokes. This is natural from the viewpoint that each stroke is a trajectory and thus can be represented as a sequence of local features. This viewpoint, however, has a clear limitation in that local features cannot deal with global structure directly. For example, the sequence of local features cannot deal with the fact that the two end points of character "0" should be close to each other. In this paper we propose a simple and novel global feature that describes the global structure of the character shape of each class. We prove the importance of the global feature through a feature selection experiment. Specifically, we show that the global features are more often selected than local features to enhance classification accuracy under the AdaBoostbased machine learning framework. Recognition experiments using online numeral data show also that the use of global features improves recognition accuracy.

Dynamic Programming Matching with Global Features for Online Character Recognition

“…They used online patterns, each of which was re-sampled to just 6 points, and all 6 C 2 pairs were used for extracting 6 C 2 global features. Our previous research [4] has proposed the relative vector between an arbitrary point pair on the stroke and revealed that each character class has its own important global structure through the feature selection experiments.…”

“…As global features, we exploit the feature proposed in [4]. This global feature is defined as a simple relative vector between an arbitrary point pair on the stroke.…”

“…Mori et al [4] have shown that a global feature is useful for improving the online character recognition accuracy. Their global feature is defined as the relative vector between two temporarily distant points.…”

“…For a pattern of length I, it is possible to consider I(I − 1)/2 global features. In [4], the discriminative global features are selected automatically from those I(I − 1)/2 features by using the AdaBoost. Surprisingly, the conventional DP matching methods have not dealt with such a global feature.…”