An experimental evaluation of visual similarity for HDR images

In this paper, we investigate visual similarity for high dynamic range (HDR) images. We collect crowdsourcing data through a web-based experimental interface, in which the participants are asked to choose one of the two candidate images as being more similar to the query image. Triplets forming the query-and-candidates sets are obtained by random sampling from existing HDR data sets. Experimental control factors include choice of tone mapping operator (TMO), choice of distance metric, and choice of image feature. The image features that we experiment with are chosen from the features that are commonly used in the usual low dynamic range setting including features learned via Convolutional Neural Networks. The set of image features also includes combined features where the combination coefficients are estimated using logistic regression. We compute correlations between human judgments and quantitative features to understand how much each feature contributes to visual similarity. Combined features yield nearly 84% agreement with human judgments when applied on tone mapped images. Though we observed that using common features directly on raw or linearly scaled HDR images yield subpar correlation estimates compared to using them on tone mapped HDR images, we did not observe significant effect due to the choice of TMO on the estimates. As an application, we propose an improvement to style-based tone mapping for more correctly imparting desired styles to HDR images with different characteristics.

     

Toward an efficient and scalable feature selection approach for internet traffic classification

There is significant interest in the network management and industrial security community about the need to identify the “best” and most relevant features for network traffic in order to properly characterize user behaviour and predict future traffic. The ability to eliminate redundant features is an important Machine Learning (ML) task because it helps to identify the best features in order to improve the classification accuracy as well as to reduce the computational complexity related to the construction of the classifier. In practice, feature selection (FS) techniques can be used as a preprocessing step to eliminate irrelevant features and as a knowledge discovery tool to reveal the “best” features in many soft computing applications. In this paper, we investigate the advantages and disadvantages of such FS techniques with new proposed metrics (namely goodness, stability and similarity). We continue our efforts toward developing an integrated FS technique that is built on the key strengths of existing FS techniques. A novel way is proposed to identify efficiently and accurately the “best” features by first combining the results of some well-known FS techniques to find consistent features, and then use the proposed concept of support to select a smallest set of features and cover data optimality. The empirical study over ten high-dimensional network traffic data sets demonstrates significant gain in accuracy and improved run-time performance of a classifier compared to individual results produced by some well-known FS techniques.     

Parabolic curves of evolving surfaces

In this article we show how certain geometric structures which are also associated with a smooth surface evolve as the shape of the surface changes in a 1-parameter family. We concentrate on the parabolic set and its image under the Gauss map, but the same techniques also classify the changes in the dual of the surface. All these have significance for computer vision, for example through their connection with specularities and apparent contours. With the aid of our complete classification, which includes all the phenomena associated with multi-contact tangent planes as well as those associated with parabolic sets, we re-examine examples given by J. Koenderink in his book (1990) under the title of Morphological Scripts.We also explain some of the connections between parabolic sets and ridges of a surface, where principal curvatures achieve turning values along lines of curvature.The point of view taken is the analysis of the contact between surfaces and their tangent planes. A systematic investigation of this yields the results using singularity theory. The mathematical details are suppressed here and appear in Bruce et al. (1993).The third author was supported by the Esprit grant VIVA while this paper was in preparation.     

Dissimilarity between two skeletal trees in a context

Skeletal trees are commonly used in order to express geometric properties of the shape. Accordingly, tree-edit distance is used to compute a dissimilarity between two given shapes. We present a new tree-edit based shape matching method which uses a recent coarse skeleton representation. The coarse skeleton representation allows us to represent both shapes and shape categories in the form of depth-1 trees. Consequently, we can easily integrate the influence of the categories into shape dissimilarity measurements. The new dissimilarity measure gives a better within group versus between group separation, and it mimics the asymmetric nature of human similarity judgements.     

Mumford-Shah Regularizer with Contextual Feedback

We present a simple and robust feature preserving image regularization by letting local region measures modulate the diffusivity. The purpose of this modulation is to disambiguate low level cues in early vision. We interpret the Ambrosio-Tortorelli approximation of the Mumford-Shah model as a system with modulatory feedback and utilize this interpretation to integrate high level information into the regularization process. The method does not require any prior model or learning; the high level information is extracted from local regions and fed back to the regularization step. An important characteristic of the method is that both negative and positive feedback can be simultaneously used without creating oscillations. Experiments performed with both gray and color natural images demonstrate the potential of the method under difficult noise types, non-uniform contrast, existence of multi-scale patterns and textures.

An approach to the measurement of shunt resistance of individual subcells in thin‐film tandem devices

In this paper, the selective illumination approach is adopted to separately extract the shunt resistance of the individual subcells belonging to a tandem cell. The method relies on simple theoretical considerations and is based on the measurement of the current–voltage characteristic of the tandem cell by alternately keeping one of the subcells under dark conditions. Numerical simulations are employed to support the reliability of the technique, which is experimentally tested on micromorph devices deposited onto glass covered by a V‐shaped transparent conducting oxide and subject to different thermal treatments. Copyright © 2013 John Wiley & Sons, Ltd.     

From a Non-Local Ambrosio-Tortorelli Phase Field to a Randomized Part Hierarchy Tree

In its most widespread imaging and vision applications, Ambrosio and Tortorelli (AT) phase field is a technical device for applying gradient descent to Mumford and Shah simultaneous segmentation and restoration functional or its extensions. As such, it forms a diffuse alternative to sharp interfaces or level sets and parametric techniques. The functionality of the AT field, however, is not limited to segmentation and restoration applications. We demonstrate the possibility of coding parts—features that are higher level than edges and boundaries—after incorporating higher level influences via distances and averages. The iteratively extracted parts using the level curves with double point singularities are organized as a proper binary tree. Inconsistencies due to non-generic configurations for level curves as well as due to visual changes such as occlusion are successfully handled once the tree is endowed with a probabilistic structure. As a proof of concept, we present (1) the most probable configurations from our randomized trees; and (2) correspondence matching results between illustrative shape pairs. The work is a significant step towards establishing exponentially decaying diffuse distance fields as bridges between low level visual processing and shape computations.     

Experimental investigation of tornado-like vortex dynamics with swirl ratio: The mean and turbulent flow fields

An experimental simulation of tornado-like vortices is conducted in a small tornado vortex simulator in order to study the effect of swirl ratio on flow characteristics. Particle Image Velocimetry (PIV) method is employed to quantitatively determine the tornado-vortex velocity field for swirl ratios ranging from 0.08 to 1. The radial and tangential components of velocity as well as the core radius of the tornado increase with increase in swirl ratio. The location of the maximum radial and tangential velocities is adjacent to the ground where the tornado vortex interacts with the surface. The values of normal and shear turbulent stresses indicate the existence of a laminar core for small swirl. As expected the shear stresses increase with swirl ratio as the vortex becomes turbulent. The highest turbulent production corresponds to the critical case of vortex touchdown.     

Austenite Reconstruction Elucidates Prior Grain Size Dependence of Toughness in a Low Alloy Steel

Texture and grain size distribution of the austenite phase at high temperature have distinct effects on the subsequent transformed microstructure and mechanical properties of steel alloys at room temperature. Measurement of austenite at high temperature is not, however, a trivial task especially if orientation maps are desired. Therefore, a technique for determining austenite texture and grain size that was present at high temperature is highly desirable. In this work, we review several pipe samples intended for use as well casing with large variations in toughness and martensitic microstructures at room temperature. Microstructural analysis of the martensite structures could not easily explain these differences due to how martensite forms from austenite. An algorithm was developed to reconstruct the parent austenite at high temperature from martensite microstructure at room temperature. This technique successfully reconstructed orientation maps for the prior austenite in these samples, demonstrating differences in texture, grain orientation spread, and austenite grain size that could account for the differences in mechanical properties.