Automatic object segmentation from calibrated images

This paper addresses the problem of automatically obtaining the object/background segmentation of a rigid 3D object observed in a set of images that have been calibrated for camera pose and intrinsics. Such segmentations can be used to obtain a shape representation of a potentially texture-less object by computing a visual hull. We propose an automatic approach where the object to be segmented is identified by the pose of the cameras instead of user input such as 2D bounding rectangles or brush-strokes. The key behind our method is a pairwise MRF framework that combines (a) foreground/background appearance models, (b) epipolar constraints and (c) weak stereo correspondence into a single segmentation cost function that can be efficiently solved by Graph-cuts. The segmentation thus obtained is further improved using silhouette coherency and then used to update the foreground/background appearance models which are fed into the next Graph-cut computation. These two steps are iterated until segmentation convergences. Our method can automatically provide a 3D surface representation even in texture-less scenes where MVS methods might fail. Furthermore, it confers improved performance in images where the object is not readily separable from the background in colour space, an area that previous segmentation approaches have found challenging. © 2011 IEEE.

Pothole detection in asphalt pavement images

Pavement condition assessment is essential when developing road network maintenance programs. In practice, the data collection process is to a large extent automated. However, pavement distress detection (cracks, potholes, etc.) is mostly performed manually, which is labor-intensive and time-consuming. Existing methods either rely on complete 3D surface reconstruction, which comes along with high equipment and computation costs, or make use of acceleration data, which can only provide preliminary and rough condition surveys. In this paper we present a method for automated pothole detection in asphalt pavement images. In the proposed method an image is first segmented into defect and non-defect regions using histogram shape-based thresholding. Based on the geometric properties of a defect region the potential pothole shape is approximated utilizing morphological thinning and elliptic regression. Subsequently, the texture inside a potential defect shape is extracted and compared with the texture of the surrounding non-defect pavement in order to determine if the region of interest represents an actual pothole. This methodology has been implemented in a MATLAB prototype, trained and tested on 120 pavement images. The results show that this method can detect potholes in asphalt pavement images with reasonable accuracy.

Parameter optimization for automated concrete detection in image data

Several research studies have been recently initiated to investigate the use of construction site images for automated infrastructure inspection, progress monitoring, etc. In these studies, it is always necessary to extract material regions (concrete or steel) from the images. Existing methods made use of material's special color/texture ranges for material information retrieval, but they do not sufficiently discuss how to find these appropriate color/texture ranges. As a result, users have to define appropriate ones by themselves, which is difficult for those who do not have enough image processing background. This paper presents a novel method of identifying concrete material regions using machine learning techniques. Under the method, each construction site image is first divided into regions through image segmentation. Then, the visual features of each region are calculated and classified with a pre-trained classifier. The output value determines whether the region is composed of concrete or not. The method was implemented using C++ and tested over hundreds of construction site images. The results were compared with the manual classification ones to indicate the method's validity.

Concrete column recognition in images and videos

The automated detection of structural elements (e.g., columns and beams) from visual data can be used to facilitate many construction and maintenance applications. The research in this area is under initial investigation. The existing methods solely rely on color and texture information, which makes them unable to identify each structural element if these elements connect each other and are made of the same material. The paper presents a novel method of automated concrete column detection from visual data. The method overcomes the limitation by combining columns’ boundary information with their color and texture cues. It starts from recognizing long vertical lines in an image/video frame through edge detection and Hough transform. The bounding rectangle for each pair of lines is then constructed. When the rectangle resembles the shape of a column and the color and texture contained in the pair of lines are matched with one of the concrete samples in knowledge base, a concrete column surface is assumed to be located. This way, one concrete column in images/videos is detected. The method was tested using real images/videos. The results are compared with the manual detection ones to indicate the method’s validity.

Improving Pothole Recognition through Vision Tracking for Automated Pavement Assessment

Pavement condition assessment is essential when developing road network maintenance programs. In practice, pavement sensing is to a large extent automated when regarding highway networks. Municipal roads, however, are predominantly surveyed manually due to the limited amount of expensive inspection vehicles. As part of a research project that proposes an omnipresent passenger vehicle network for comprehensive and cheap condition surveying of municipal road networks this paper deals with pothole recognition. Existing methods either rely on expensive and high-maintenance range sensors, or make use of acceleration data, which can only provide preliminary and rough condition surveys. In our previous work we created a pothole detection method for pavement images. In this paper we present an improved recognition method for pavement videos that incrementally updates the texture signature for intact pavement regions and uses vision tracking to track detected potholes. The method is tested and results demonstrate its reasonable efficiency.

Alternative methods for surface texturing

Surface texturing has a great potential to improve tribological performance. First, possible texturing methods were identified and classified according to their physical principles. In sequence, some alternative texturing methods are presented. Some of them are already currently used either in industry or in laboratory, and innovations or simplifications are described for them. Others are innovative techniques. Some were explored only tentatively, where basic ideas and simple experimental investigations were developed to check their validity. Others were explored in more detail, so that their practical applicability could be identified. The first texturing method was photochemical texturing using a simple and cheap apparatus. Masking with inkjet printing before chemical etching was also successful to texture metallic samples. A new method involving electrochemical texturing, without the need to previously mask the samples to be textured have been studied in terms of voltage, current, mechanical configuration of the apparatus and electrolyte flushing. Another method aims to generate randomly distributed circular pockets on steel surfaces and involves dispersion of small acid droplets in oil. The final method involves the selective formation of hard areas on a steel surface by locallised diffusion, which should then develop into a texture during wear.

The parameter space of graphene chemical vapor deposition on polycrystalline Cu

A systematic study of the parameter space of graphene chemical vapor deposition (CVD) on polycrystalline Cu foils is presented, aiming at a more fundamental process rationale in particular regarding the choice of carbon precursor and mitigation of Cu sublimation. CH 4 as precursor requires H 2 dilution and temperatures ≥1000 °C to keep the Cu surface reduced and yield a high-quality, complete monolayer graphene coverage. The H 2 atmosphere etches as-grown graphene; hence, maintaining a balanced CH 4/H 2 ratio is critical. Such balance is more easily achieved at low-pressure conditions, at which however Cu sublimation reaches deleterious levels. In contrast, C 6H 6 as precursor requires no reactive diluent and consistently gives similar graphene quality at 100-150 °C lower temperatures. The lower process temperature and more robust processing conditions allow the problem of Cu sublimation to be effectively addressed. Graphene formation is not inherently self-limited to a monolayer for any of the precursors. Rather, the higher the supplied carbon chemical potential, the higher the likelihood of film inhomogeneity and primary and secondary multilayer graphene nucleation. For the latter, domain boundaries of the inherently polycrystalline CVD graphene offer pathways for a continued carbon supply to the catalyst. Graphene formation is significantly affected by the Cu crystallography; i.e., the evolution of microstructure and texture of the catalyst template form an integral part of the CVD process. © 2012 American Chemical Society.

Learning based automatic face annotation for arbitrary poses and expressions from frontal images only

Statistical approaches for building non-rigid deformable models, such as the Active Appearance Model (AAM), have enjoyed great popularity in recent years, but typically require tedious manual annotation of training images. In this paper, a learning based approach for the automatic annotation of visually deformable objects from a single annotated frontal image is presented and demonstrated on the example of automatically annotating face images that can be used for building AAMs for fitting and tracking. This approach employs the idea of initially learning the correspondences between landmarks in a frontal image and a set of training images with a face in arbitrary poses. Using this learner, virtual images of unseen faces at any arbitrary pose for which the learner was trained can be reconstructed by predicting the new landmark locations and warping the texture from the frontal image. View-based AAMs are then built from the virtual images and used for automatically annotating unseen images, including images of different facial expressions, at any random pose within the maximum range spanned by the virtually reconstructed images. The approach is experimentally validated by automatically annotating face images from three different databases. © 2009 IEEE.

A performance evaluation of volumetric 3D interest point detectors

This paper presents the first performance evaluation of interest points on scalar volumetric data. Such data encodes 3D shape, a fundamental property of objects. The use of another such property, texture (i.e. 2D surface colouration), or appearance, for object detection, recognition and registration has been well studied; 3D shape less so. However, the increasing prevalence of 3D shape acquisition techniques and the diminishing returns to be had from appearance alone have seen a surge in 3D shape-based methods. In this work, we investigate the performance of several state of the art interest points detectors in volumetric data, in terms of repeatability, number and nature of interest points. Such methods form the first step in many shape-based applications. Our detailed comparison, with both quantitative and qualitative measures on synthetic and real 3D data, both point-based and volumetric, aids readers in selecting a method suitable for their application. © 2012 Springer Science+Business Media, LLC.

Fabrication, densification, and replica molding of 3D carbon nanotube microstructures

The introduction of new materials and processes to microfabrication has, in large part, enabled many important advances in microsystems, labon- a-chip devices, and their applications. In particular, capabilities for cost-effective fabrication of polymer microstructures were transformed by the advent of soft lithography and other micromolding techniques 1,2, and this led a revolution in applications of microfabrication to biomedical engineering and biology. Nevertheless, it remains challenging to fabricate microstructures with well-defined nanoscale surface textures, and to fabricate arbitrary 3D shapes at the micro-scale. Robustness of master molds and maintenance of shape integrity is especially important to achieve high fidelity replication of complex structures and preserving their nanoscale surface texture. The combination of hierarchical textures, and heterogeneous shapes, is a profound challenge to existing microfabrication methods that largely rely upon top-down etching using fixed mask templates. On the other hand, the bottom-up synthesis of nanostructures such as nanotubes and nanowires can offer new capabilities to microfabrication, in particular by taking advantage of the collective self-organization of nanostructures, and local control of their growth behavior with respect to microfabricated patterns. Our goal is to introduce vertically aligned carbon nanotubes (CNTs), which we refer to as CNT "forests", as a new microfabrication material. We present details of a suite of related methods recently developed by our group: fabrication of CNT forest microstructures by thermal CVD from lithographically patterned catalyst thin films; self-directed elastocapillary densification of CNT microstructures; and replica molding of polymer microstructures using CNT composite master molds. In particular, our work shows that self-directed capillary densification ("capillary forming"), which is performed by condensation of a solvent onto the substrate with CNT microstructures, significantly increases the packing density of CNTs. This process enables directed transformation of vertical CNT microstructures into straight, inclined, and twisted shapes, which have robust mechanical properties exceeding those of typical microfabrication polymers. This in turn enables formation of nanocomposite CNT master molds by capillary-driven infiltration of polymers. The replica structures exhibit the anisotropic nanoscale texture of the aligned CNTs, and can have walls with sub-micron thickness and aspect ratios exceeding 50:1. Integration of CNT microstructures in fabrication offers further opportunity to exploit the electrical and thermal properties of CNTs, and diverse capabilities for chemical and biochemical functionalization 3. © 2012 Journal of Visualized Experiments.

Characterization of the surface acoustic wave devices based on ZnO/nanocrystalline diamond structures

Nanocrystalline ZnO films with strong (0002) texture and fine grains were deposited onto ultra-nanocrystalline diamond (UNCD) layers on silicon using high target utilization sputtering technology. The unique characteristic of this sputtering technique allows room temperature growth of smooth ZnO films with a low roughness and low stress at high growth rates. Surface acoustic wave (SAW) devices were fabricated on ZnO/UNCD structure and exhibited good transmission signals with a low insertion loss and a strong side-lobe suppression for the Rayleigh mode SAW. Based on the optimization of the layered structure of the SAW device, a good performance with a coupling coefficient of 5.2% has been realized, promising for improving the microfluidic efficiency in droplet transportation comparing with that of the ZnO/Si SAW device. An optimized temperature coefficient of frequency of -23.4 ppm°C-1 was obtained for the SAW devices with the 2.72 μm-thick ZnO and 1.1 μm-thick UNCD film. Significant thermal effect due to the acoustic heating has been redcued which is related to the temperature stability of the ZnO/UNCD SAW device. © 2013 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.

Modeling the carbon nanofiber addressed liquid crystal microlens array from experimentally observed optical phenomena

This paper presents a novel method of using experimentally observed optical phenomena to reverse-engineer a model of the carbon nanofiber-addressed liquid crystal microlens array (C-MLA) using Zemax. It presents the first images of the optical profile for the C-MLA along the optic axis. The first working optical models of the C-MLA have been developed by matching the simulation results to the experimental results. This approach bypasses the need to know the exact carbon nanofiber-liquid crystal interaction and can be easily adapted to other systems where the nature of an optical device is unknown. Results show that the C-MLA behaves like a simple lensing system at 0.060-0.276 V/μm. In this lensing mode the C-MLA is successfully modeled as a reflective convex lens array intersecting with a flat reflective plane. The C-MLA at these field strengths exhibits characteristics of mostly spherical or low order aspheric arrays, with some aspects of high power aspherics. It also exhibits properties associated with varying lens apertures and strengths, which concur with previously theorized models based on E-field patterns. This work uniquely provides evidence demonstrating an apparent "rippling" of the liquid crystal texture at low field strengths, which were successfully reproduced using rippled Gaussian-like lens profiles. © 2014 Published by Elsevier B.V.

Surface texturing through cylinder buckling

We consider the axial buckling of a thin-walled cylinder fitted onto a mandrel core with a prescribed annular gap. The buckling pattern develops fully and uniformly to yield a surface texture of regular diamond-shaped buckles, which we propose for novel morphing structures. We describe experiments that operate well into the postbuckling regime, where a classical analysis does not apply; we show that the size of buckles depends on the cylinder radius and the gap width, but not on its thickness, and we formulate simple relationships from kinematics alone for estimating the buckle proportions during loading. © 2014 by ASME.

Dense active appearance models using a bounded diameter minimum spanning tree

We present a method for producing dense Active Appearance Models (AAMs), suitable for video-realistic synthesis. To this end we estimate a joint alignment of all training images using a set of pairwise registrations and ensure that these pairwise registrations are only calculated between similar images. This is achieved by defining a graph on the image set whose edge weights correspond to registration errors and computing a bounded diameter minimum spanning tree (BDMST). Dense optical flow is used to compute pairwise registration and we introduce a flow refinement method to align small scale texture. Once registration between training images has been established we propose a method to add vertices to the AAM in a way that minimises error between the observed flow fields and a flow field interpolated between the AAM mesh points. We demonstrate a significant improvement in model compactness using the proposed method and show it dealing with cases that are problematic for current state-of-the-art approaches.