Multi-view hac for Semi-supervised Document Image Classification

Abstract

Abstract. This paper presents a semi-supervised document image classification system that aims to be integrated into a commercial document reading software. This system is asserted like an annotation help. From a set of unknown document images given by a human operator, the system computes regrouping hypothesis of same physical layout images and proposes them to the operator. Then he can correct them, validate them, keeping in mind that his objective is to have homogeneous groups of images. These groups will be used for the training of the supervised document image classifier. Our system contains N feature spaces and a metric function for each of them. These allow to compute the similarity between two points of the same space. After projecting each image in these N feature spaces, the system builds N hierarchical agglomerative classification trees (hac) corresponding to each feature space. The proposals for regroupings formulated by the various hac are confronted and merged. Results, evaluated by the number of corrections done by the operator are presented on different image sets.

FOL Learning for Knowledge Discovery in Documents

“…The authors proposed an algorithm based on a logistic regression classifier working on a set of manually segmented and labelled page images, followed by statistical classifiers for the logical layout analysis. A similar approach can be found in (Carmagnac, 2004a) in which a semisupervised document image classification system is presented. Given a set of unknown document images, the system uses unsupervised clustering to obtain grouping hypotheses for the same physical layout images.…”