Michel A. Westenberg

Contour detection based on nonclassical receptive field inhibition

We propose a biologically motivated method, called nonclassical receptive field (non-CRF) inhibition (more generally, surround inhibition or suppression), to improve contour detection in machine vision. Non-CRF inhibition is exhibited by 80% of the orientation-selective neurons in the primary visual cortex of monkeys and has been shown to influence human visual perception as well. Essentially, the response of an edge detector at a certain point is suppressed by the responses of the operator in the region outside the supported area. We combine classical edge detection with isotropic and anisotropic inhibition, both of which have counterparts in biology. We also use a biologically motivated method (the Gabor energy operator) for edge detection. The resulting operator responds strongly to isolated lines, edges, and contours, but exhibits weak or no response to edges that are part of texture. We use natural images with associated ground truth contour maps to assess the performance of the proposed operator for detecting contours while suppressing texture edges. Our method enhances contour detection in cluttered visual scenes more effectively than classical edge detectors used in machine vision (Canny edge detector). Therefore, the proposed operator is more useful for contour-based object recognition tasks, such as shape comparison, than traditional edge detectors, which do not distinguish between contour and texture edges. Traditional edge detection algorithms can, however, also be extended with surround suppression. This study contributes also to the understanding of inhibitory mechanisms in biology.

Contour and boundary detection improved by surround suppression of texture edges

We propose a computational step, called surround suppression, to improve detection of object contours and region boundaries in natural scenes. This step is inspired by the mechanism of non-classical receptive field inhibition that is exhibited by most orientation selective neurons in the primary visual cortex and that influences the perception of groups of edges or lines. We illustrate the principle and the effect of surround suppression by adding this step to the Canny edge detector. The resulting operator responds strongly to isolated lines and edges, region boundaries, and object contours, but exhibits a weaker or no response to texture edges. Additionally, we introduce a new post-processing method that further suppresses texture edges. We use natural images with associated subjectively defined desired output contour and boundary maps to evaluate the performance of the proposed additional steps. In a contour detection task, the Canny operator augmented with the proposed suppression and post-processing step achieves better results than the traditional Canny edge detector and the SUSAN edge detector. The performance gain is highest at scales for which these latter operators strongly react to texture in the input image.

Suppression of contour perception by band-limited noise and its relation to nonclassical receptive field inhibition.

Abstract. Band-spectrum noise has been shown to suppress the visual perception of printed letters. The suppression exhibits a specific dependence on the spatial frequency of the noise, and the frequency domain of most effective inhibition has been related to the size of the letters. In this paper, we address two important questions that were left open by previous studies: (1) Is the observed effect specific to text, and which parameters determine the domain of most effective suppression? (2) What is the origin of the effect in terms of underlying neural processes? We conduct a series of psychophysical experiments that demonstrate that the frequency domain of most effective inhibition depends on the stroke width of the letter rather than on the letter size. These experiments also demonstrate that the effect is not specific to the recognition of letters but also applies to other objects and even to single bars. We attribute the observed effect to nonclassical receptive field (non-CRF) inhibition in visual area V1. This mechanism has previously been suggested as the possible origin of various other perceptual effects. We introduce computational models of two types of cell that incorporate non-CRF inhibition, which are based on Gabor energy filters extended by surround suppression of two kinds: isotropic and anisotropic. The computational models confirm previous qualitative explanations of perceptual effects, such as orientation contrast pop-out, reduced saliency of lines embedded in gratings, and reduced saliency of contours surrounded by textures. We apply the computational models to the images used in the psychophysical experiments. The computational results show a dependence of the inhibition effect on the spatial frequency of the noise that is similar to the suppression effect measured in the psychophysical experiments. The experimental results and their explanation give further support to the idea of a possible functional role of non-CRF inhibition in the separation of contour from texture information and the mediation of object contours to higher cortical areas.

Shape Preserving Filament Enhancement Filtering

Morphological connected set filters for extraction of filamentous details from medical images are developed. The advantages of these filters are that they are shape preserving and do not amplify noise. Two approaches are compared: (i) multi-scale filtering (ii) single-step shape filtering using connected set (or attribute) thinnings. The latter method highlights all filamentous structure in a single filtering stage, regardless of the scale. The second approach is an order of magnitude faster than the first, filtering a 2563 volume in 41.65 s on a 400 MHz Pentium II.

Frequency domain volume rendering by the wavelet X-ray transform

We describe a wavelet based X-ray rendering method in the frequency domain with a smaller time complexity than wavelet splatting. Standard Fourier volume rendering is summarized and interpolation and accuracy issues are briefly discussed. We review the implementation of the fast wavelet transform in the frequency domain. The wavelet X-ray transform is derived, and the corresponding Fourier-wavelet volume rendering algorithm (FWVR) is introduced, FWVR uses Haar or B-spline wavelets and linear or cubic spline interpolation. Various combinations are tested and compared with wavelet splatting (WS). We use medical MR and CT scan data, as well as a 3-D analytical phantom to assess the accuracy, time complexity, and memory cost of both FWVR and WS. The differences between both methods are enumerated.

Kelp Diagrams: Point Set Membership Visualization

We present Kelp Diagrams, a novel method to depict set relations over points, i.e., elements with predefined positions. Our method creates schematic drawings and has been designed to take aesthetic quality, efficiency, and effectiveness into account. This is achieved by a routing algorithm, which links elements that are part of the same set by constructing minimum cost paths over a tangent visibility graph. There are two styles of Kelp Diagrams to depict overlapping sets, a nested and a striped style, each with its own strengths and weaknesses. We compare Kelp Diagrams with two existing methods and show that our approach provides a more consistent and clear depiction of both element locations and their set relations.

Compressed Adjacency Matrices: Untangling Gene Regulatory Networks

We present a novel technique-Compressed Adjacency Matrices-for visualizing gene regulatory networks. These directed networks have strong structural characteristics: out-degrees with a scale-free distribution, in-degrees bound by a low maximum, and few and small cycles. Standard visualization techniques, such as node-link diagrams and adjacency matrices, are impeded by these network characteristics. The scale-free distribution of out-degrees causes a high number of intersecting edges in node-link diagrams. Adjacency matrices become space-inefficient due to the low in-degrees and the resulting sparse network. Compressed adjacency matrices, however, exploit these structural characteristics. By cutting open and rearranging an adjacency matrix, we achieve a compact and neatly-arranged visualization. Compressed adjacency matrices allow for easy detection of subnetworks with a specific structure, so-called motifs, which provide important knowledge about gene regulatory networks to domain experts. We summarize motifs commonly referred to in the literature, and relate them to network analysis tasks common to the visualization domain. We show that a user can easily find the important motifs in compressed adjacency matrices, and that this is hard in standard adjacency matrix and node-link diagrams. We also demonstrate that interaction techniques for standard adjacency matrices can be used for our compressed variant. These techniques include rearrangement clustering, highlighting, and filtering.

Computer Analysis of Images and Patterns: 10th International Conference

The paper briefly overviews the design and applications of cylindrical panoramic cameras characterized by a rotating linear sensor capturing one image column at time. The camera provides very high image resolutions paid by motion distortions in dynamic scenes. The images are used for stereo reconstruction and visualization of static scenes when extremely high image resolution is of benefit.

Blood vessel segmentation using moving-window robust automatic threshold selection

Two moving-window methods, using either flat or Gaussian weighted windows, for local thresholding with robust automatic threshold selection are developed. The results show that fast segmentation of blood vessels against a varying background and noise is possible at modest computational cost. Volumes of 128 x 256/sup 2/ and 256/sup 3/ can be segmented in 3.1 s and 6.6 s, for flat, and 12.6 s and 30.8 s for Gaussian windows, respectively, on a 1.9 GHz Pentium 4.

eXamine: Exploring annotated modules in networks

BackgroundBiological networks have a growing importance for the interpretation of high-throughput “omics” data. Integrative network analysis makes use of statistical and combinatorial methods to extract smaller subnetwork modules, and performs enrichment analysis to annotate the modules with ontology terms or other available knowledge. This process results in an annotated module, which retains the original network structure and includes enrichment information as a set system. A major bottleneck is a lack of tools that allow exploring both network structure of extracted modules and its annotations.ResultsThis paper presents a visual analysis approach that targets small modules with many set-based annotations, and which displays the annotations as contours on top of a node-link diagram. We introduce an extension of self-organizing maps to lay out nodes, links, and contours in a unified way. An implementation of this approach is freely available as the Cytoscape app eXamineConclusionseXamine accurately conveys small and annotated modules consisting of several dozens of proteins and annotations. We demonstrate that eXamine facilitates the interpretation of integrative network analysis results in a guided case study. This study has resulted in a novel biological insight regarding the virally-encoded G-protein coupled receptor US28.