J. Fdez-Valdivia

Information theoretic measure for visual target distinctness

It is of great benefit to have advance knowledge of human visual target acquisition performance for targets or other relevant objects. However, search performance inherently shows a large variance and depends strongly on prior knowledge of the perceived scene. A typical search experiment therefore requires a large number of observers to obtain statistically reliable data. Moreover, measuring target acquisition performance in field situations is usually impractical and often very costly or even dangerous. The paper presents a method for characterizing information of a target relative to its background. The resultant computational measures are then applied to quantify the visual distinctness of targets in complex natural backgrounds from digital imagery. A generalization of the Kullback-Leibler joint information gain of various random variables is shown to correlate strongly with visual target distinctness as estimated by human observers. Bootstrap methods for assessing statistical accuracy were used to produce this inference.

The RGFF representational model: a system for the automatically learned partitioning of "visual patterns" in digital images

This paper describes a system for the automatically learned partitioning of visual patterns in 2D images, based on sophisticated band-pass filtering with fixed scale and orientation sensitivity. The visual patterns are defined as the features which have the highest degree of alignment in the statistical structure across different frequency bands. The analysis reorganizes the image according to an invariance constraint in statistical structure and consists of three stages: pre-attentive stage, integration stage, and learning stage. The first stage takes the input image and performs filtering with log-Gabor filters. Based on their responses, activated filters which are selectively sensitive to patterns in the image are short listed. In the integration stage, common grounds between several activated sensors are explored. The filtered responses are analyzed through a family of statistics. For any given two activated filters, a distance between them is derived via distances between their statistics. The third stage performs cluster partitioning for learning the subspace of log-Gabor filters needed to partition the image data. The clustering is based on a dissimilarity measure intended to highlight scale and orientation invariance of the responses. The technique is illustrated on real and simulated data sets. Finally, this paper presents a computational visual distinctness measure computed from the image representational model based on visual patterns. Experiments are performed to investigate its relation to distinctness as measured by human observers.

The selection of natural scales in 2D images using adaptive Gabor filtering

This paper analyzes how the natural scales of the shapes in 2D images can be extracted. Spatial information is analyzed by multiple units sensitive to both spatial and spatial-frequency variables. Scale estimates of the relevant shapes are constructed only from strongly responding detectors. The meaningful structures in the response of a detector (computed through 2D Gabor filtering) are, at their natural level of resolution, relatively sharp and have well-defined boundaries. A natural scale is so defined as a level producing local minimum of a function that returns the relative sharpness of the detector response filtered over a range of scales. In a second stage, to improve a first crude estimate of the local scale, the criterion is also rewritten to directly select scales at locations of significant features of each activated detector.

A dynamic approach for clustering data

Abstract This paper introduces a new method for clustering data using a dynamic scheme. An appropriate partitioning is obtained based on both a dissimilarity measure between pairs of entities as well as a dynamic procedure of splitting. A dissimilarity function is defined by using the cost of the optimum path from a datum to each entity on a graph, with the cost of a path being defined as the greatest distance between two successive vertices on the path. The procedure of clustering is dynamic in the sense that the initial problem of determining a partition into an unknown number of natural groupings has been reduced to a sequence of only two class splitting stages. Having arisen from any particular application, the proposed approach could be effective for many domains, and it is especially successful to identify clusters if there is lack of prior knowledge about the data set. The usefulness of the dynamic algorithm to deal with elongated or non-piecewise linear separable clusters as well as sparse and dense groupings is demonstrated with several data sets.

Mapping citation patterns of book chapters in the Book Citation Index

In this paper we provide the reader with a visual representation of relationships among the impact of book chapters indexed in the Book Citation Index using information gain values and published by different academic publishers in specific disciplines. The impact of book chapters can be characterized statistically by citations histograms. For instance, we can compute the probability of occurrence of book chapters with a number of citations in different intervals for each academic publisher. We predict the similarity between two citation histograms based on the amount of relative information between such characterizations. We observe that the citation patterns of book chapters follow a Lotkaian distribution. This paper describes the structure of the Book Citation Index using ‘heliocentric clockwise maps’ which allow the reader not only to determine the grade of similarity of a given academic publisher indexed in the Book Citation Index with a specific discipline according to their citation distribution, but also to easily observe the general structure of a discipline, identifying the publishers with higher impact and output.

Using models of feature perception in distortion measure guidance

Abstract In this paper we present three error measures based on feature perception models, in which pixel errors are computed on locations at which humans might perceive features in the reference image. In the first part of this work, the three schemes of feature detection will be discussed and evaluated in terms of their performance for a simple visual signal-processing task. The first model is based on the use of local intensity gradients, the second based on the use of phase congruency in an image, and the third based on the use of local energy maxima for a few active sensors under a multichannel organization of the reference picture. In the second part of this paper, examples are provided of object detection and recognition applications that illustrate the ability of the induced error measures to predict the detectability of objects in natural backgrounds as well as their perceptual capabilities.

Computing visual target distinctness through selective filtering, statistical features, and visual patterns

This paper presents three computational visual distinctness measures, computed from image representational models based on selective filtering, statistical features, and visual patterns, respectively. They are applied to quantify the visual distinctness of targets in complex natural scenes. The measure that applies a simple decision rule to the distances between segregated visual patterns is shown (1) to predict human observer performance in search and detection tasks on complex natural imagery, and (2) to correlate strongly with visual target distinctness estimated by human observers.

Boundary simplification in cartography preserving the characteristics of the shape features

Abstract In this paper, we assume that cartographic boundaries have features at a variety of different degrees of simplification and therefore each line segment showing a different feature must be simplified at its proper degree. Our boundary simplification method preserves the characteristics of the shape features, and therefore avoids missing fine features and overlooking coarse features. Here simplification consists of the dominant points detected on the line which has been segmented previously into a number of parts, each one showing a different feature at an appropriate degree of smoothing. We propose an automatic method to segment the line into a number of nonoverlapping parts, each one revealing a different feature. To find the best degree of simplification for each segment, we select the simplification minimizing a normalized measure of the zeros of curvature of the segment.

Mapping academic institutions according to their journal publication profile: Spanish universities as a case study

We introduce a novel methodology for mapping academic institutions based on their journal publication profiles. We believe that journals in which researchers from academic institutions publish their works can be considered as useful identifiers for representing the relationships between these institutions and establishing comparisons. However, when academic journals are used for research output representation, distinctions must be introduced between them, based on their value as institution descriptors. This leads us to the use of journal weights attached to the institution identifiers. Since a journal in which researchers from a large proportion of institutions published their papers may be a bad indicator of similarity between two academic institutions, it seems reasonable to weight it in accordance with how frequently researchers from different institutions published their papers in this journal. Cluster analysis can then be applied to group the academic institutions, and dendrograms can be provided to illustrate groups of institutions following agglomerative hierarchical clustering. In order to test this methodology, we use a sample of Spanish universities as a case study. We first map the study sample according to an institutions overall research output, then we use it for two scientific fields (Information and Communication Technologies, as well as Medicine and Pharmacology) as a means to demonstrate how our methodology can be applied, not only for analyzing institutions as a whole, but also in different disciplinary contexts.

Representing planar curves by using a scale vector

Abstract This paper introduces a new approach to solving the problem of representing planar curves. We describe the 2-D curve C not at all different scales σ, but each curve part C i of C , isolating a different structure at its single scale σ i . Therefore, we represent the planar curve at a scale vector ( σ 1 , …, σ L supposing that the curve is partitioned in L parts C 1 , …, C L ). We propose an automatic method to divide the contour into the number of nonoverlapping parts C 1 , …, C L , each of them showing a different underlying structure. This process requires neither the number of parts in the curve nor the minimum level of homogeneity for the entities within a particular part. The partition is based on three elements: a vector φ of statistical measures calculated to each class, a distance function d ( φ i , φ j ) between vectors corresponding to two different classes, and a halt criterion based on a measure of the improvement in the disimilarity between the elements of the partition.