ADCIS - Spécialiste en traitement d'images et analyse d'images

Outils de reconnaissance de formes et classification

Télécharger la fiche produit

Le module Recognition Toolkit, developpé par la société Recognition Science, Inc. met à la disposition des utilisateurs de la version Developer d’Aphelion™ tout un ensemble d’outils de reconnaissance et de classification d’objets basés sur l’analyse par composantes principales et la mesure d’attributs. Des classificateurs peuvent être générés automatiquement évitant ainsi à l’utilisateur d’avoir à spécifier des règles complexes pour la reconnaissance des objets. Les outils proposés dans le module peuvent être appliqués aussi bien sur des ensembles d’objets Aphelion (ObjectSets) que sur des données scalaires standards. Des macros d’exemple sont fournies avec le module et peuvent soit être employées telles quelles, soit servir de modèle.

Ajoute à Aphelion Developer les fonctionnalités suivantes :
		Reconnaissance de formes			Classification automatique
		Classification statistique			Analyse d’attributs et évaluation

Recognition Toolkit fournit des outils de reconnaissance de formes capables de fonctionner sur des données continues, discrètes ou encore symboliques. Le module contient un ensemble d’opérateurs pour la classification, l’apprentissage, l’évaluation d’attributs, le codage et l’extraction de nuages de points. Une vingtaine de fonctions constituent l’ensemble des fonctions de reconnaissance de formes incluses dans le module et cinq fonctions supplémentaires servent aux extractions avancées et aussi comme opérateurs de démonstration lors du développement d’algorithmes complexes de reconnaissance. Le coeur du module inclut l’ensemble des classificateurs, les opérateurs d’apprentissage, une fonction de regroupement standard, une évaluation d’attributs, des fonctions d’analyse et des fonctions utilitaires pour le prétraitement et la constitution de bases d’apprentissage. La liste complète des opérateurs avec leur description en anglais est fournie ci-dessous.

Opérateurs du module Recognition Toolkit (description en anglais) :

AutoCode Utility Automatically codes features to discrete form. The user can select the maximum number of discrete states permitted. (description) Bayes_Cls Classifier This trainer sets up all of the reference data required by Bayes_Cls. This includes generation of class centroids, covariance, and base weights. Bayes_Trn Training Multivariate normal Bayesian classifier for normally distributed measurement data which can produce optimal results when its assumptions are met. Prior class weights can be accepted from the reference file, be set to default values, or adjusted by the user to favor specific classes. Boxes Extract Finds related opposite and adjacent corners for the extraction of rectangular objects. Centroid Utility Given data and labels this routine computes class centroids in the feature space. Corners Extract Finds corners under user specified tolerances. Uses basic line object input. CorrEval Evaluation Correlation based on feature evaluation. This operator can assist in evaluation of the utility of measurement data and reducing redundancy in a set of features. DIT_Cls Classifier The Discriminant Information Tree (DIT) classifier provides a capability to classify data which is discrete, coded or symbolic. It can function without any class centroid separability under complex data distributions. Continuous measurement data can be coded for use with this classifier using the AutoCode or Encode operators. DIT_Trn Training This trainer uncovers the optimal discriminant information tree and probability structures for DIT_Cls. Encode Utility Encodes measurement data using specified partitions. The partitions can be from AutoCode or user generated. FindROI Extract This routine detects regions of interest (ROI's) in an image using a K-nearest neighbor algorithm. It can be used for cueing or as the first stage in a multistage classifier. InfoEval Evaluation This tool determines, on a feature-by-feature basis, the best features for class separation. It can be used to evaluate symbolic features, discrete features, and coded measurement data. It provides a measure of the discriminant information in each feature, and automatically rank-orders an output feature set from most powerful to least powerful. KNN_Trn Training Generates a reference prototype file for the K-Nearest-Neighbor Classifiers of KNN_Cls and FindROI.

LinkEval Evaluation Finds tree linkage structure which maximizes second-order discriminant information. Can assist in feature discrimination or redundancy evaluation. MeanAsgn Utility This routine assigns data to closest centroid. Permits rapid evalua tion of class separability of user selected or modified centroid values. Can be used with Centroid operator or PFSCIust operator to develop alternative classifiers. MinD_Cls Classifier This Minimum Distance Classifier provides an easy to use classification capability with very minimal training data requirements and minimal computational complexity. It assigns samples based on Euclidean distance from class centroids in the feature space. MinD_Trn Training Sets up class centroids and weights for the minimum distance classifier. An excellent tool for the interpretation of measurement data from well separated classes. MRegres Evaluation This Multiple Regression analysis technique computes the optimal linear combination of features to estimate an outcome. It can be used in feature evaluation and redundancy analysis. Normal Utility Generates normalized features by extracting the mean and standardizing variances to one. Eliminates measurement unit scaling effects in data. PairEval Evaluation Computes pairwise discriminant information. It provides a measure of how well each pair of features can be used to predict the correct class label. It can also help identify redundant features in a selection of discriminant feature sets. Parallels Extract Finds parallel line pairs under user specified control of separation distance ranges and degree of overlap. Uses line object input. PFSClust Clustering Pseudo F-Statistic (PFS) Clustering finds the natural classes in a set of data. This unsupervised classification algorithm can be used to automatically find natural separations in the feature space. It can guide in feature selection and class definition for a later classifier implementation. WinFeats Extract Produces window based features useful for detection and classification. It supports FindROI in detection of regions of interest. Window size and overlaps are selected by the user. KNN_Cls Classifier K-Nearest-Neighbor classifier for utilization on measurement data with complex distributions. This classifier can operate when classes are not easily separable.