Sébastien Adam

Symbol and character recognition: application to engineering drawings

Abstract. In this paper, we consider the general problem of technical document interpretation, as applied to the documents of the French Telephonic Operator, France Télécom. More precisely, we focus the content of this paper on the computation of a new set of features allowing the classification of multioriented and multiscaled patterns. This set of invariants is based on the Fourier–Mellin Transform. The interests of this computation rely on the excellent classification rate obtained with this method and also on using this Fourier–Mellin transform within a “filtering mode”, with which we can solve the well known difficult problem of connected character recognition.

Dynamic Random Forests

In this paper, we introduce a new Random Forest (RF) induction algorithm called Dynamic Random Forest (DRF) which is based on an adaptative tree induction procedure. The main idea is to guide the tree induction so that each tree will complement as much as possible the existing trees in the ensemble. This is done here through a resampling of the training data, inspired by boosting algorithms, and combined with other randomization processes used in traditional RF methods. The DRF algorithm shows a significant improvement in terms of accuracy compared to the standard static RF induction algorithm.

On the selection of decision trees in Random Forests

In this paper we present a study on the Random Forest (RF) family of ensemble methods. In a “classical” RF induction process a fixed number of randomized decision trees are inducted to form an ensemble. This kind of algorithm presents two main drawbacks : (i) the number of trees has to be fixed a priori (ii) the interpretability and analysis capacities offered by decision tree classifiers are lost due to the randomization principle. This kind of process in which trees are independently added to the ensemble, offers no guarantee that all those trees will cooperate effectively in the same committee. This statement rises two questions : are there any decision trees in a RF that provide the deterioration of ensemble performance? If so, is it possible to form a more accurate committee via removal of decision trees with poor performance? The answer to these questions is tackled as a classifier selection problem. We thus show that better subsets of decision trees can be obtained even using a sub-optimal classifier selection method. This proves that “classical” RF induction process, for which randomized trees are arbitrary added to the ensemble, is not the best approach to produce accurate RF classifiers. We also show the interest in designing RF by adding trees in a more dependent way than it is traditionally done in “classical” RF induction algorithms.

Using Random Forests for Handwritten Digit Recognition

In the pattern recognition field, growing interest has been shown in recent years for multiple classifier systems and particularly for bagging, boosting and random sub-spaces. Those methods aim at inducing an ensemble of classifiers by producing diversity at different levels. Following this principle, Breiman has introduced in 2001 another family of methods called random forest. Our work aims at studying those methods in a strictly pragmatic approach, in order to provide rules on parameter settings for practitioners. For that purpose we have experimented the forest-RI algorithm, considered as the random forest reference method, on the MNIST handwritten digits database. In this paper, we describe random forest principles and review some methods proposed in the literature. We present next our experimental protocol and results. We finally draw some conclusions on random forest global behavior according to their parameter tuning.

Influence of Hyperparameters on Random Forest Accuracy

In this paper we present our work on the Random Forest (RF) family of classification methods. Our goal is to go one step further in the understanding of RF mechanisms by studying the parametrization of the reference algorithm Forest-RI. In this algorithm, a randomization principle is used during the tree induction process, that randomly selects K features at each node, among which the best split is chosen. The strength of randomization in the tree induction is thus led by the hyperparameter K which plays an important role for building accurate RF classifiers. We have decided to focus our experimental study on this hyperparameter and on its influence on classification accuracy. For that purpose, we have evaluated the Forest-RI algorithm on several machine learning problems and with different settings of K in order to understand the way it acts on RF performance. We show that default values of K traditionally used in the literature are globally near-optimal, except for some cases for which they are all significatively sub-optimal. Thus additional experiments have been led on those datasets, that highlight the crucial role played by feature relevancy in finding the optimal setting of K .

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.

Learning to Detect Tables in Scanned Document Images Using Line Information

This paper presents a method to detect table regions in document images by identifying the column and row line-separators and their properties. The method employs a run-length approach to identify the horizontal and vertical lines present in the input image. From each group of intersecting horizontal and vertical lines, a set of 26 low-level features are extracted and an SVM classifier is used to test if it belongs to a table or not. The performance of the method is evaluated on a heterogeneous corpus of French, English and Arabic documents that contain various types of table structures and compared with that of the Tesseract OCR system.

An integer linear program for substitution-tolerant subgraph isomorphism and its use for symbol spotting in technical drawings

This paper tackles the problem of substitution-tolerant subgraph isomorphism which is a specific class of error-tolerant isomorphism. This problem aims at finding a subgraph isomorphism of a pattern graph S in a target graph G. This isomorphism only considers label substitutions and forbids vertex and edge insertion in G. This kind of subgraph isomorphism is often needed in pattern recognition problems when graphs are attributed with real values and no exact matching can be found between attributes due to noise. Our proposal to solve the problem of substitution-tolerant subgraph isomorphism relies on its formulation in the Integer Linear Program (ILP) formalism. Using a general ILP solver, the approach is able to find, if one exists, a mapping of a pattern graph into a target graph such that the topology of the searched graph is kept and the editing operations between the labels have a minimal cost. This technique is evaluated on both a set of synthetic graphs and a problem of symbol detection in technical drawings. In the second case, document and symbol images are represented by vector-attributed Region Adjacency Graphs built from a segmentation process. Obtained results demonstrate the relevance of considering subgraph isomorphism as an optimization process.

Multi-Objective Optimization for SVM Model Selection

In this paper, we propose a multi-objective optimization method for SVM model selection using the well known NSGA-II algorithm. FA and FR rates are the two criteria used to find the optimal hyperparameters of a set of SVM classifiers. The proposed strategy is applied to a digit/outlier discrimination task embedded in a more global information extraction system that aims at locating and recognizing numerical fields in handwritten incoming mail documents. Experiments conducted on a large database of digits and outliers show clearly that our method compares favorably with the results obtained by a state-of-the- art mono-objective optimization technique using the classical Area Under ROC Curve criterion (AUC).

Learning graph prototypes for shape recognition

This paper presents some new approaches for computing graph prototypes in the context of the design of a structural nearest prototype classifier. Four kinds of prototypes are investigated and compared: set median graphs, generalized median graphs, set discriminative graphs and generalized discriminative graphs. They differ according to (i) the graph space where they are searched for and (ii) the objective function which is used for their computation. The first criterion allows to distinguish set prototypes which are selected in the initial graph training set from generalized prototypes which are generated in an infinite set of graphs. The second criterion allows to distinguish median graphs which minimize the sum of distances to all input graphs of a given class from discriminative graphs, which are computed using classification performance as criterion, taking into account the inter-class distribution. For each kind of prototype, the proposed approach allows to identify one or many prototypes per class, in order to manage the trade-off between the classification accuracy and the classification time. Each graph prototype generation/selection is performed through a genetic algorithm which can be specialized to each case by setting the appropriate encoding scheme, fitness and genetic operators. An experimental study performed on several graph databases shows the superiority of the generation approach over the selection one. On the other hand, discriminative prototypes outperform the generative ones. Moreover, we show that the classification rates are improved while the number of prototypes increases. Finally, we show that discriminative prototypes give better results than the median graph based classifier.