Jörg Dahmen

Experiments with an extended tangent distance

Invariance is an important aspect in image object recognition. We present results obtained with an extended tangent distance incorporated in a kernel density based Bayesian classifier to compensate for affine image variations. An image distortion model for local variations is introduced and its relationship to tangent distance is considered. The proposed classification algorithms are evaluated on databases of different domains. An excellent result of 2.2% error rate on the original USPS handwritten digits recognition task is obtained. On a database of radiographs from daily routine, best results are obtained by combining the tangent distance and the proposed distortion model.

Adaptation in statistical pattern recognition using tangent vectors

We integrate the tangent method into a statistical framework for classification analytically and practically. The resulting consistent framework for adaptation allows us to efficiently estimate the tangent vectors representing the variability. The framework improves classification results on two real-world pattern recognition tasks from the domains handwritten character recognition and automatic speech recognition.

Statistical Image Object Recognition using Mixture Densities

In this paper, we present a mixture density based approach to invariant image object recognition. To allow for a reliable estimation of the mixture parameters, the dimensionality of the feature space is optionally reduced by applying a robust variant of linear discriminant analysis. Invariance to affine transformations is achieved by incorporating invariant distance measures such as tangent distance. We propose an approach to estimating covariance matrices with respect to image variabilities as well as a new approach to combined classification, called the virtual test sample method. Application of the proposed classifier to the well known US Postal Service handwritten digits recognition task (USPS) yields an excellent error rate of 2.2%. We also propose a simple, but effective approach to compensate for local image transformations, which significantly increases the performance of tangent distance on a database of 1,617 medical radiographs taken from clinical daily routine.

Learning of Variability for Invariant Statistical Pattern Recognition

In many applications, modelling techniques are necessary which take into account the inherent variability of given data. In this paper, we present an approach to model class specific pattern variation based on tangent distance within a statistical framework for classification. The model is an effective means to explicitly incorporate invariance with respect to transformations that do not change class-membership like e.g. small affine transformations in the case of image objects. If no prior knowledge about the type of variability is available, it is desirable to learn the model parameters from the data. The probabilistic interpretation presented here allows us to view learning of the variational derivatives in terms of a maximum likelihood estimation problem. We present experimental results from two different real-world pattern recognition tasks, namely image object recognition and automatic speech recognition. On the US Postal Service handwritten digit recognition task, learning of variability achieves results well comparable to those obtained using specific domain knowledge. On the SieTill corpus for continuously spoken telephone line recorded German digit strings the method shows a significant improvement in comparison with a common mixture density approach using a comparable amount of parameters. The probabilistic model is well-suited to be used in the field of statistical pattern recognition and can be extended to other domains like cluster analysis.

Discriminative Training of Gaussian Mixtures for Image Object Recognition

In this paper we present a discriminative training procedure for Gaussian mixture densities. Conventional maximum likelihood (ML) training of such mixtures proved to be very efficient for object recognition, even though each class is treated separately in training. Discriminative criteria offer the advantage that they also use out-of-class data, that is they aim at optimizing class separability. We present results on the US Postal Service (USPS) handwritten digits database and compare the discriminative results to those obtained by ML training. We also compare our best results with those reported by other groups, proving them to be state-of-the-art.

Automatic Classification of Red Blood Cells using Gaussian Mixture Densities

In this paper we present an invariant statistical approach to classifying red blood cells (RBC). Given a database of 5062 grayscale images, we model the distribution of the observations by using Gaussian mixture densities within a Bayesian framework. As invariance is of great importance when classifying RBC, we use a Fourier-Mellin based approach to extract features which are invariant with respect to 2D rotation, shift and scale. To prove the efficiency of our approach, we also apply it to the widely used US Postal Service handwritten digits recognition task, obtaining state-of-the-art results.

A Probabilistic View on Tangent Distance

In this paper we present a new probabilistic interpretation of tangent distance, which proved to be very effective in modeling image transformations in object recognition. Descriptions of the resulting distributions in pattern space are given for different possible models of Variation, leading to a natural derivation of tangent distance. Furthermore, a possible generalization is presented and experimental results on the well known US Postal Service database are presented.

Structured Covariance Matrices for Statistical Image Object Recognition

In this paper we present different approaches to structuring covariance matrices within Statistical classifiers. This is motivated by the fact that the use of full covariance matrices is infeasible in many applications. On the one hand, this is due to the high number of model Parameters that have to be estimated, on the other hand the computational complexity of a classifier based on full covariance matrices is very high. We propose the use of diagonal and band-matrices to replace full covariance matrices and we also show that computation of tangent distance is equivalent to using a structured covariance matrix within a Statistical classifier.

An Automatic Approach to Invariant Radiograph Classification

In this paper we present an invariant statistical approach to classifying medical radiographs, being an important step in the RWTH Aachen — University of Technology IRMA system (Image Retrieval in Medical Applications). We propose a Bayesian classifier based on Gaussian kernel densities, where invariance is incorporated by using invariant distance measures. The performance of the classifier is evaluated on a dataset of 1,617 radiographs coming from daily routine. The obtained error rate of 7.6% is significantly better than the results reported in other works, using the same dataset. Furthermore, the presented probabilistic framework is also applicable to other (multi-) object recognition tasks.

Invariant image object recognition using mixture densities

We present a mixture density based approach to invariant image object recognition. We start our experiments using Gaussian mixture densities within a Bayesian classifier. Invariance to affine transformations is achieved by replacing the Euclidean distance with SIMARDs tangent distance. We propose an approach to estimating covariance matrices with respect to image invariances as well as a new classifier combination scheme, called the virtual test sample method. On the US Postal Service handwritten digits recognition task (USPS), we obtain an excellent classification error rate of 2.7%, using the original USPS training and test sets.