Luciano da Fontoura Costa

Robust Skeletonization through Exact Euclidean Distance Transform and its Application to Neuromorphometry

This paper presents how robust 1-pixel-wide and 8-connected skeletons can be obtained simultaneously with exact distance transform calculation. The proposed approach is based on the new concept of exact dilation. Two alternative algorithms for exact Euclidean distance transform calculation allowing exact dilation are described: a simpler approach based on the SEDR (sorted exact distance representation) data structure, which allows exact distance transform calculation; and a more effective strategy based on border propagation. In both techniques the distances are assigned strictly according to sequences of increasing exact distances in the orthogonal lattice. Because of the high accuracy allowed by such procedures, progressive dilations, high quality and accurate 1-pixel-wide and 8-connected skeletons can be obtained corresponding to the frontiers between previously labelled distinct connected objects. Although this method can be useful for determining generalized Dirichlet tessellations, which is also illustrated in this article, its full potential is harnessed by previously segmenting the contours of connected objects by removing their points corresponding to curvature peaks, which are obtained by using an effective multi-scale curvature estimation technique. In such a way, not only high-quality 1-pixel-wide and 8-connected skeletons are obtained for any shape, but also the whole approach becomes considerably robust to small distortions in the object contours, thus avoiding one of the great shortcomings in traditional skeletonization methods. The application of such methods to an important problem in computational neuroscience and neuromorphometry, namely the automated extraction of tapered dendrograms, is described and illustrated. Considerations regarding distances in orthogonal lattices, typical problems in skeletonization, and the practical implementation of the proposed techniques are also included.

A percolation approach to neural morphometry and connectivity

This article addresses the issues of neural shape characterization and analysis from the perspective of one of the main roles played by neural shapes, namely, connectivity. This study is oriented toward the geometry at the individual cell level and involves the use of the percolation concept from statistical mechanics, which is reviewed in an accessible fashion. The characterization of the neural cell geometry with respect to connectivity is performed in terms of critical percolation probability obtained experimentally while considering several types of geometrical interactions between cells, therefore directly expressing the potential for connections defined by each situation. Two basic situations are considered: dendrite-dendrite and dendrite-axon interactions. The obtained results corroborate the potential of the critical percolation probability as a valuable resource for characterizing, classifying, and analyzing the morphology of neural cells.

Modeling and evaluating summaries using complex networks

This paper presents a summary evaluation method based on a complex network measure. We show how to model summaries as complex networks and establish a possible correlation between summary quality and the measure known as dynamics of the network growth. It is a generic and language independent method that enables easy and fast comparative evaluation of summaries. We evaluate our approach using manually produced summaries and automatic summaries produced by three automatic text summarizers for the Brazilian Portuguese language. The results are in agreement with human intuition and showed to be statistically significant.

Straight Line Detection as an Optimization Problem: An Approach Motivated by the Jumping Spider Visual System

Straight lines are important features in images, and their detection plays a major role in the compression, representation and analysis of visual information. The visual system of spiders from the Salticidae family is especially effective for straight line detection, due to their elongated and moveable retinae, which are used to scan the visual field. This paper presents a method for straight line motivated by the visual system of the Salticidae, which uses an optimization strategy (namely Nelder and Meads amoeba with simulated annealing) to find maxima on the continuous ρ-θ parameter space that correspond to straight lines in the image. The method considers the spatially quantized nature of the image spaces and allows unlimited parametric resolution without the need to sample large regions of the parameter space.

Parallel implementation of exact dilations and multi-scale skeletonization

Skeletonization has become an important pre-processing step in shape analysis and pattern recognition as it expresses the hierarchical organization of data to be analysed and simplifies the computational procedures. More recently, architectures for image processing have greatly benefitted from the use of reconfigurable devices such as field programmable gate arrays and complex programmable logic devices. We present in this work the implementation of a versatile method for Exact Dilation using reconfigurable devices. The developed method allows us to extract skeletons and other important features from an image.