CLASSIFIER COMBINATION BY BAYESIAN NETWORKS FOR HANDWRITING RECOGNITION

In the field of handwriting recognition, classifier combination received much more interest than the study of powerful individual classifiers. This is mainly due to the enormous variability among the patterns to be classified, that typically requires the definition of complex high dimensional feature spaces: as the overall complexity increases, the risk of inconsistency in the decision of the classifier increases as well. In this framework, we propose a new combining method based on the use of a Bayesian Network. In particular, we suggest to reformulate the classifier combination problem as a pattern recognition one in which each input pattern is associated to a feature vector composed by the output of the classifiers to be combined. A Bayesian Network is then used to automatically infer the probability distribution for each class and eventually to perform the final classification. Experiments have been performed by using two different pools of classifiers, namely an ensemble of Learning Vector Quantization neural networks and an ensemble of Back Propagation neural networks, and handwritten specimen from the UCI Machine Learning Repository. The obtained performance has been compared with those exhibited by multi-classifier systems adopting the classifiers, but three of the most effective and widely used combining rules: the Majority Vote, the Weighted Majority Vote and the Borda Count.

Ensemble segmentation using AdaBoost with application to liver lesion extraction from a CT volume

This paper describes an ensemble segmentation trained by the AdaBoost algorithm, which finds a sequence of weak hypotheses, each of which is appropriate for the distribution on training example, and combines the weak hypotheses by a weighted majority vote. In our study, a weak hypothesis corresponds to a weak segmentation process. This paper shows a procedure for generating an ensemble segmentation algorithm using AdaBoost, and applies it to a liver lesion extraction problem from a contrast enhanced abdominal CT volume. A leave-one-patient-out validation test using 16 CT volumes demonstrated the effectiveness of the generated ensemble segmentation algorithm. In addition, we evaluated the performance by applying the algorithm to unknown test data provided by the �3D Liver Tumor Segmentation Challenge 2008�.