Eric Trupin

Building Synthetic Graphical Documents for Performance Evaluation

In this paper we present a system that allows to build synthetic graphical documents for the performance evaluation of symbol recognition systems. The key contribution of this work is the building of whole documents like drawings or maps. We exploit the layer property of graphical documents by positioning symbol sets in different ways from a same background using positioning constraints. Experiments are presented to build two kinds of test document databases : bags of symbol and architectural drawings.

Classification method study for automatic form class identification

We present three classifiers used in automatic forms class identification. The first category of classifier includes the k-nearest neighbours (kNN) and the multilayer perceptron (MLP) classifiers. The second category corresponds to a new structural classifier based on tree comparison. The low level information based on a pyramidal decomposition of the document image is used by the kNN and the MLP classifiers, while the high level information represents the form content with a hierarchical structure used by the new structural classifier. Experimental results are presented. Some strategies of classifier co-operation are proposed.

A general framework for the evaluation of symbol recognition methods

Performance evaluation is receiving increasing interest in graphics recognition. In this paper, we discuss some questions regarding the definition of a general framework for evaluation of symbol recognition methods. The discussion is centered on three key elements in performance evaluation: test data, evaluation metrics and protocols of evaluation. As a result of this discussion we state some general principles to be taken into account for the definition of such a framework. Finally, we describe the application of this framework to the organization of the first contest on symbol recognition in GREC’03, along with the results obtained by the participants.

Automatic Ground-truth Generation for Document Image Analysis and Understanding

Performance evaluation for document image analysis and understanding is a recurring problem. Many ground- truthed document image databases are now used to evaluate algorithms, but these databases are less useful for the design of a complete system in a precise context. This paper proposes an approach for the automatic generation of synthesised document images and associated ground-truth information based on a derivation of publishing tools. An implementation of this approach illustrates the richness of the produced information.

Local structural analysis: A primer

The structural analysis is a processing step during which graphs are extracted from binary images. We can decompose the structural analysis into local and global approaches. The local approach decomposes the connected components, and the global approach groups them together. This paper deals especially with the local structural analysis. The local structural analysis is employed for different applications like symbol recognition, line drawing interpretation, and character recognition. We propose here a primer on the local structural analysis. First, we propose a general decomposition of the local structural analysis into four steps: object graph extraction, mathematical approximation, high-level object construction, and object graph correction. Then, we present some considerations on the method comparison and combination.

Using bags of symbols for automatic indexing of graphical document image databases

A database is only usefull if it is associated a set of procedures allowing to retrieve relevant elements for the users’ needs. A lot of IR techniques have been developed for automatic indexing and retrieval in document databases. Most of these use indexes depending on the textual content of documents, and very few are able to handle graphical or image content without human annotation. This paper describes an approach similar to the bag of words technique for automatic indexing of graphical document image databases and different ways to consequently query these databases. In an unsupervised manner, this approach proposes a set of automatically discovered symbols that can be combined with logical operators to build queries.

A graph classification approach using a multi-objective genetic algorithm application to symbol recognition

In this paper, a graph classification approach based on a multi-objective genetic algorithm is presented. The method consists in the learning of sets composed of synthetic graph prototypes which are used for a classification step. These learning graphs are generated by simultaneously maximizing the recognition rate while minimizing the confusion rate. Using such an approach the algorithm provides a range of solutions, the couples (confusion, recognition) which suit to the needs of the system. Experiments are performed on real data sets, representing 10 symbols. These tests demonstrate the interest to produce prototypes instead of finding representatives which simply belong to the data set.

Knowledge Representation and Acquisition for Engineering Document Analysis

This paper tackles the problem of bootstrapping engineering documents recognition systems. A user-friendly interface is presented. Its aim is to acquire knowledge concerning the graphical appearance of objects, but also to learn the best approach to use among our tools in order to recognise the learned objects.

Structural Classification for Retrospective Conversion of Documents

This paper describes the structural classification method used in a strategy for retrospective conversion of documents. This strategy consists in an cycle in which document analysis and document understanding interact. This cycle is initialized by the extraction of the outline of the layout and logical structures of the document. Then, each iteration of the cycle consists in the detection and the processing of inconsistencies in the document modeling. The cycle ends when no more inconsistency occurs. A structural representation is used to describe documents. This representation is detailed. Retrospective conversion consists in identifying each entity of the document and its structures as well. The structural classification method based on graph comparison is used at several levels of this process. Graph comparison is also used in the learning of generic entities.

Adaptable Vectorisation System Based on Strategic Knowledge and XML Representation Use

This paper presents a vectorisation system based on the use of strategic knowledge. This one is composed of two parts: a processing library and a graphic user interface. Our processing library is composed of image pre-processing and vectorisation tools. Our graphic user interface is used for the strategic knowledge acquisition and operationalisation. It allows to construct and to execute scenarios, exploiting any processing of our library, according to documents’ contexts and users’ adopted strategies. A XML data representation is used, allowing an easy data manipulation. A scenario example is presented for graphics recognition on utility maps.