MCBoost: Multiple classifier boosting for perceptual co-clustering of images and visual features

We present a new co-clustering problem of images and visual features. The problem involves a set of non-object images in addition to a set of object images and features to be co-clustered. Co-clustering is performed in a way that maximises discrimination of object images from non-object images, thus emphasizing discriminative features. This provides a way of obtaining perceptual joint-clusters of object images and features. We tackle the problem by simultaneously boosting multiple strong classifiers which compete for images by their expertise. Each boosting classifier is an aggregation of weak-learners, i.e. simple visual features. The obtained classifiers are useful for object detection tasks which exhibit multimodalities, e.g. multi-category and multi-view object detection tasks. Experiments on a set of pedestrian images and a face data set demonstrate that the method yields intuitive image clusters with associated features and is much superior to conventional boosting classifiers in object detection tasks.

Initializing Vision Based Trackers Using Semantic Texton Forests

Vision based tracking can provide the spatial location of project related entities such as equipment, workers, and materials in a large-scale congested construction site. It tracks entities in a video stream by inferring their motion. To initiate the process, it is required to determine the pixel areas of the entities to be tracked in the following consecutive video frames. For the purpose of fully automating the process, this paper presents an automated way of initializing trackers using Semantic Texton Forests (STFs) method. STFs method performs simultaneously the segmentation of the image and the classification of the segments based on the low-level semantic information and the context information. In this paper, STFs method is tested in the case of wheel loaders recognition. In the experiments, wheel loaders are further divided into several parts such as wheels and body parts to help learn the context information. The results show 79% accuracy of recognizing the pixel areas of the wheel loader. These results signify that STFs method has the potential to automate the initialization process of vision based tracking.

ID ENGINE: Image Classifier

C++ Prototype implementation of multi-modal image classification and retrieval method for construction site images

A coordination-based approach to elasticity of floppy and stiff random networks

We study the role of connectivity on the linear and nonlinear elastic behavior of amorphous systems using a two-dimensional random network of harmonic springs as a model system. A natural characterization of these systems arises in terms of the network coordination relative to that of an isostatic network $\delta z$; a floppy network has $\delta z<0$, while a stiff network has $\delta z>0$. Under the influence of an externally applied load we observe that the response of both floppy and rigid network are controlled by the same critical point, corresponding to the onset of rigidity. We use numerical simulations to compute the exponents which characterize the shear modulus, the amplitude of non-affine displacements, and the network stiffening as a function of $\delta z$, derive these theoretically and make predictions for the mechanical response of glasses and fibrous networks.

Making a shallow network deep: Conversion of a boosting classifier into a decision tree by boolean optimisation

This paper presents a novel way to speed up the evaluation time of a boosting classifier. We make a shallow (flat) network deep (hierarchical) by growing a tree from decision regions of a given boosting classifier. The tree provides many short paths for speeding up while preserving the reasonably smooth decision regions of the boosting classifier for good generalisation. For converting a boosting classifier into a decision tree, we formulate a Boolean optimization problem, which has been previously studied for circuit design but limited to a small number of binary variables. In this work, a novel optimisation method is proposed for, firstly, several tens of variables i.e. weak-learners of a boosting classifier, and then any larger number of weak-learners by using a two-stage cascade. Experiments on the synthetic and face image data sets show that the obtained tree achieves a significant speed up both over a standard boosting classifier and the Fast-exit-a previously described method for speeding-up boosting classification, at the same accuracy. The proposed method as a general meta-algorithm is also useful for a boosting cascade, where it speeds up individual stage classifiers by different gains. The proposed method is further demonstrated for fast-moving object tracking and segmentation problems. © 2011 Springer Science+Business Media, LLC.

Adhesive properties of gecko-inspired mimetic via micropatterned carbon nanotube forests

The adhesive properties of the gecko foot have inspired designs of advanced micropatterned surfaces with increased contact areas. We have fabricated micropatterned pillars of vertically aligned carbon nanotube forests with a range of pillar diameters, heights, and spacings (or pitch). We used nanoindentation to measure their elastic and orthogonal adhesion properties and derive their scaling behavior. The patterning of nanotube forests into pillar arrays allows a reduction of the effective modulus from 10 to 15 MPa to 0.1-1 MPa which is useful for developing maximum conformal adhesion. © 2012 American Chemical Society.

Multiple classifier boosting and tree-structured classifiers

Visual recognition problems often involve classification of myriads of pixels, across scales, to locate objects of interest in an image or to segment images according to object classes. The requirement for high speed and accuracy makes the problems very challenging and has motivated studies on efficient classification algorithms. A novel multi-classifier boosting algorithm is proposed to tackle the multimodal problems by simultaneously clustering samples and boosting classifiers in Section 2. The method is extended into an online version for object tracking in Section 3. Section 4 presents a tree-structured classifier, called Super tree, to further speed up the classification time of a standard boosting classifier. The proposed methods are demonstrated for object detection, tracking and segmentation tasks. © 2013 Springer-Verlag Berlin Heidelberg.

Boundary effects on the vibration statistics of a random plate

This paper considers the estimation of statistics of displacement of a vibrating rectangular plate with random wave scatterers. The influence of uncertainty is investigated using point impedance theory. Coherent boundary effects are seen, which decrease when the number of scatterers increases. The boundary effect is investigated using images and the first side and corner reflections are found to be a minimum requirement to estimate the spatial correlation. Statistics for point driven response are investigated under the assumption that the statistics of the natural frequencies follow those of the Gaussian Orthogonal Ensemble (GOE). The estimates are compared with Monte Carlo simulation results, and they show good agreement. © 2012 Elsevier Ltd. All rights reserved.

Cylindrical Fresnel lenses based on carbon nanotube forests

The forests of carbon nanotubes have been termed as the darkest man-made materials. Such materials exhibit near-perfect optical absorption (reflectance∼0.045%) due to low reflectance and nanoscale surface roughness. We have demonstrated the utilization of these perfectly absorbing forests to produce binary amplitude cylindrical Fresnel lenses. The opaque Fresnel zones are defined by the dark nanotube forests and these lenses display efficient focusing performance at optical wavelengths. Lensing performance was analyzed both computationally and experimentally with good agreement. Such nanostructure based lenses have many potential applications in devices like photovoltaic solar cells. © 2012 American Institute of Physics.

Optimized vertical carbon nanotube forests for multiplex surface-enhanced raman scattering detection

The highly sensitive and molecule-specific technique of surface-enhanced Raman spectroscopy (SERS) generates high signal enhancements via localized optical fields on nanoscale metallic materials, which can be tuned by manipulation of the surface roughness and architecture on the submicrometer level. We investigate gold-functionalized vertically aligned carbon nanotube forests (VACNTs) as low-cost straightforward SERS nanoplatforms. We find that their SERS enhancements depend on their diameter and density, which are systematically optimized for their performance. Modeling of the VACNT-based SERS substrates confirms consistent dependence on structural parameters as observed experimentally. The created nanostructures span over large substrate areas, are readily configurable, and yield uniform and reproducible SERS enhancement factors. Further fabricated micropatterned VACNTs platforms are shown to deliver multiplexed SERS detection. The unique properties of CNTs, which can be synergistically utilized in VACNT-based substrates and patterned arrays, can thus provide new generation platforms for SERS detection. © 2012 American Chemical Society.

Random function priors for exchangeable arrays with applications to graphs and relational data

A fundamental problem in the analysis of structured relational data like graphs, networks, databases, and matrices is to extract a summary of the common structure underlying relations between individual entities. Relational data are typically encoded in the form of arrays; invariance to the ordering of rows and columns corresponds to exchangeable arrays. Results in probability theory due to Aldous, Hoover and Kallenberg show that exchangeable arrays can be represented in terms of a random measurable function which constitutes the natural model parameter in a Bayesian model. We obtain a flexible yet simple Bayesian nonparametric model by placing a Gaussian process prior on the parameter function. Efficient inference utilises elliptical slice sampling combined with a random sparse approximation to the Gaussian process. We demonstrate applications of the model to network data and clarify its relation to models in the literature, several of which emerge as special cases.

Band alignment between Ta2O5 and metals for resistive random access memory electrodes engineering

Band alignment of resistive random access memory (RRAM) switching material Ta2O5 and different metal electrode materials was examined using high-resolution X-ray photoelectron spectroscopy. Schottky and hole barrier heights at the interface between electrode and Ta2O 5 were obtained, where the electrodes consist of materials with low to high work function (Φ m, v a c from 4.06 to 5.93 eV). Effective metal work functions were extracted to study the Fermi level pinning effect and to discuss the dominant conduction mechanism. An accurate band alignment between electrodes and Ta2O5 is obtained and can be used for RRAM electrode engineering and conduction mechanism study. © 2013 American Institute of Physics.