Image Processing & Applications

The diffusion of depth sensors to sense people and objects constitutes an outstanding opportunity in those fields in which the benefits of optical marker-less solutions for scanning or tracking are requested. This paper shows how two different applications based on MS Kinect device can be accomplished in the domain of lower limb prosthesis design and test. The first one refers to the use of a depth camera as a three-dimensional scanner to acquire the geometry of residual limbs or of custom-fit components. The second application is related to the motion capture of patients' gait with the prosthesis. In both cases, the technology resulted to be better than many traditional ones mainly for its limited invasivity, interesting performance, portability and low cost.

Keywords: 3D scanner , Digital human modelling , Lower limb prosthesis , Motion capture , RGB-D cameras

The diffusion of depth sensors to sense people and objects constitutes an outstanding opportunity in those fields in which the benefits of optical marker-less solutions for scanning or tracking are requested. This paper shows how two different applications based on MS Kinect device can be accomplished in the domain of lower limb prosthesis design and test. The first one refers to the use of a depth camera as a three-dimensional scanner to acquire the geometry of residual limbs or of custom-fit components. The second application is related to the motion capture of patients' gait with the prosthesis. In both cases, the technology resulted to be better than many traditional ones mainly for its limited invasivity, interesting performance, portability and low cost.

Keywords: 3D scanner , Digital human modelling , Lower limb prosthesis , Motion capture , RGB-D cameras

The reconstruction of tooth anatomies is of utmost importance when dental implant surgeries and/or orthodontic corrections must be planned. In the last few years, cone beam CT (CBCT) has gained popularity in dentistry for 3D imaging of jawbones and teeth. However, within CBCT data sets, each tooth is defined by a region, which cannot be easily separated from surrounding tissues (i.e., bone tissue) by only considering pixel's grey-intensity values. For this reason, some enhancement is usually necessary in order to properly segment tooth geometries. In this paper, a semi-automatic approach to reconstruct individual 3D tooth anatomies by processing CBCT-scan data is presented. The methodology is based on the creation of a minimal number of 2D "local ray-sum" images by adding the absorption values of adjacent voxels along the most significant views for each tooth. The knowledge of the specific anatomical patient morphology drives the selection of these significant projection directions. The reconstructed "ray-sum" images greatly enhance the clearness of the root contours, which can then be interactively traced by dentists. A set of meaningful 2D tooth contours is consequently obtained and used to automatically extract a cubic spline curve for each transverse slice, thus approximating the overall 3D tooth profile. The effectiveness of the methodology has been evaluated by comparing the results obtained for the reconstruction of anterior teeth with those obtained by using classical segmentation tools provided within commercial software.

This paper presents a new methodology for computing grey-scale granulometries and estimating the mean size of fine and coarse aggregates. The proposed approach employs area morphology and combines the information derived from both openings and closings to determine the size distribution. The method, which we refer to as bipolar area morphology (BAM), is general and can operate on particles of different size and shape. The effectiveness of the procedure was validated on a set of 13 classes of aggregates of size ranging from 0.125 to 16 mm and made a comparison with standard, fixed-shape granulometry. In the experiments our model consistently outperformed the standard approach and predicted the correct size class with overall accuracy over 92 %. Tests on three classes from real samples also confirmed the potential of the method for application in real scenarios.

Keywords: Aggregates , Area morphology , Granulometry , Image analysis

This paper investigates the problem of learning sets of discriminative patterns from local binary patterns (LBP). Such patterns are usually referred to as 'dominant local binary patterns' (DLBP). The strategies to obtain the dominant patterns may either keep knowledge of the patterns labels or discard it. It is the aim of this work to determine which is the best option. To this end the paper studies the effectiveness of different strategies in terms of accuracy, data compression ratio and time complexity. The results show that DLBP provides a significant compression rate with only a slight accuracy decrease with respect to LBP, and that retaining information about the patterns' labels improves the discrimination capability of DLBP. Theoretical analysis of time complexity revealed that the gain/loss provided by DLBP vs. LBP depends on the classification strategy: we show that, asymptotically, there is in principle no advantage when classification is based on computationally-cheap methods (such as nearest neighbour and nearest mean classifiers), because in this case determining the dominant patterns is computationally more expensive than classifying using the whole feature vector; by contrast, pattern selection can be beneficial with more complex classifiers such as support vector machines.

Keywords: Dominant local binary patterns , Feature selection , Texture classification

The injection of urea-water sprays within selective catalytic reduction systems is currently the leading technique for reducing the emission of nitrogen oxides from Diesel engines. For the process to work properly, it is crucial to guarantee the adequate size, velocity and distribution of the spray droplets upstream of the catalyst. It is therefore extremely important to understand the process of spray formation and evolution as well as possible. In this paper we describe a new methodology for inspecting the behaviour of urea-water sprays in realistic conditions. Our approach is based on a hot-air flow tunnel enabling optical inspection of the spray through phase-Doppler anemometry and back-light imaging. The procedure was employed to investigate the global and local characteristics of urea-water sprays under different flow conditions. The results proved the significant influence exerted by the flow conditions on the spray behaviour, and confirmed that the proposed system can provide considerable insight about the evolution of urea-water sprays. Cross-comparison of the droplet size estimated through phase-Doppler anemometry and back-light imaging showed substantial agreement between the two methods. This result suggests that back-light imaging is a viable alternative in those cases where complex exhaust geometry impedes the use of phase-Doppler anemometry.

Keywords: Image processing , Nitrogen oxides , Phase-Doppler anemometry , Selective catalytic reduction , Urea-water sprays

In this work we propose the use of image features based on visual perception for discriminating epithelium and stroma in histological images. In particular, we assess the capability of the following five visual features to correctly discriminate epithelium from stroma in digitised tissue micro-arrays of colorectal cancer: coarseness, contrast, directionality, line-likeliness and roughness. The use of features directly related to human perception makes it possible to evaluate the tissue's appearance on the basis of a set of meaningful parameters; moreover, the number of features used to discriminate epithelium from stroma is very small. In the experiments we used histologically-verified, well-defined images of epithelium and stroma to train three classifiers based on Support Vector Machines (SVM), Nearest Neighbour rule (1-NN) and Naïve Bayes rule (NB). We optimised SVM's parameters on a validation set, and estimated the accuracy of the three classifiers on a independent test set. The experiments demonstrate that the proposed features can correctly discriminate epithelium from stroma with state-of-the-art accuracy.

Keywords: Colorectal cancer , Epithelium , Image analysis , Perceptual features , Stroma

This paper investigates the use of digital polygons as a replacement for circular interpolated neighbourhoods for extracting texture features through Local Binary Patterns. The use of digital polygons has two main advantages: reduces the computational cost, and avoids the high-frequency loss resulting from pixel interpolation. The solution proposed in this work employs a sub-sampling scheme over Andres' digital circles. The effectiveness of the method was evaluated in a supervised texture classification experiment over eight different datasets. The results showed that digital polygons outperformed interpolated circular neighbourhoods in most cases.

Keywords: Digital circles , Digital polygons , Local Binary Patterns , Rotation invariance , Texture classification

In this paper we investigate the problem of choosing the adequate colour representation for automated surface grading. Specifically, we discuss the pros and cons of different colour spaces, point out some common misconceptions about their use, and propose a number of ‘best practices' for colour conversion. To put the discussion into practice we generated a new dataset of 25 classes of natural stone products which we used to systematically compare and evaluate the performance of seven device-dependent and three device-independent colour spaces through two classification strategies. With the nearest neighbour classifiers no significant difference emerged among the colour spaces considered, whereas with the linear classifier it was found that device-independent Lab and Luv spaces performed significantly better than the others.

Keywords: Colour spaces , Natural stones , Soft colour descriptors , Visual appearance

The objective of this study was to assess frequency and extension of the defects affecting the dentin-post interface after using different combinations of irrigants and sealers. The experimental work was conducted on single-rooted teeth extracted for orthodontic reasons. The specimens were divided into different groups, according to irrigant and endodontic cement utilized, and endodontically instrumented. After fiberglass posts cementation, cross sections were obtained at apical, middle and coronal level of the root and submitted to quantitative analyses. Different types of defects were found: bubbles, bonding defects, polymerization defect, and cement residues. The percent extension of each defect and its frequency were related to the specific irrigant/sealer combination and to the root level. Detachments of the material from dentin were found only at apical and middle levels. Chlorhexidine digluconate seems to have more beneficial effects if compared to sodium hypochlorite: samples prepared with chlorhexidine digluconate showed a higher performance, with roots including null to few defects. In detail, samples treated with chlorhexidine digluconate and Pulp Canal Sealer showed the lowest frequency and the smallest dimension of defects.

Keywords: Adhesive defects , Endodontic cements , Endoposts , Irrigants , Morphometric analysis

Quantification of 3-D deformations of human organs plays an important role in the understanding phenomena that have an impact in medical diagnosis and treatment of diseases. One important example is the mechanics of heart functions. Comparing normal deformation patterns of the cardiac cycle in healthy and diseased individuals can be a diagnostic tool that provides early and accurate indications of the onset of heart diseases. The tagging technique is an experimental mechanics method that makes it possible to utilize the extensive literature existing on the analysis of deformations utilizing the digital moire´ method for accurate and fast quantification of the heart 3-D kinematics. MRI tagging is an imaging technique used in medicine to visualize the structures of tissues of the human body in detail. MRI uses of the phenomenon of nuclear magnetic resonance to image tissues by exciting the nuclei of atoms in the tissue. Because of the different chemical composition of the tissues it can provide details that cannot be visible with CT Scans. By modulating magnetization it is possible to inscribe lattice-patterns in the tissue volume. These lattices are fixed to the under laying tissues for periods of time long enough to follow a cardiac cycle. The objective of this paper is to outline image processing techniques that can be utilized to decode the displacements and strains taking into consideration that one is dealing with large 3-D deformations that form a time sequence of images. These techniques are based on fundamental principles that have been developed in the field of digital moire´.

Keywords: Deformation of cardiac tissues , Digital moiré , Large deformations , MRI , MRI tagging

Human movements express non-verbal communication: The way humans move, live and act within a space influences and reflects the experience with a product. The study of postures and gestures can bring meaningful information to the design process. This paper explores the possibility to adopt Motion Capture technologies to inform the design process and stimulate concept generation with an Experience Design perspective. Motion data could enable designers to tackle Experience-driven design process and come up with innovative designs. However, due to their computational nature, these data are largely inaccessible for designers. This study presents a method to process the raw data coming from the Motion Capture system, with the final goal of reaching a comprehensible visualization of human movements in a modelling environment. The method was implemented and applied to a case study focused on User Experience within the car space. Furthermore, the paper presents a discussion about the conceptualization of human movement, as a way to inform and facilitate Experience-driven design process, and includes some propositions of applicable design domains.

Keywords: Body tracking , Conceptual design , Data visualization , Motion capture , User experience

Beyond ergonomic measurements, the study of human movements can help designers in exploring the rich, non-verbal communication of users' perception of products. This paper explores the ability of human gestures to express subjective experiences and therefore, to inform the design process at its early stages. We will investigate the traditional techniques used in the Experience Design domain to observe human gestures, and propose a method to couple Experience-driven design approach with Motion Capture technique. This will allow integrating qualitative user observations with quantitative and measurable data. However, the richness of information that Motion Capture can retrieve is usually inaccessible for designers. This paper presents a method to visualize human motion data so that designers can make sense of them, and use them as the starting point for concept generation.

Keywords: Body Tracking , Concept design , Data visualization , Motion Capture , User experience

The maturity reached in the development of Unmanned Air Vehicles (UAVs) systems is making them more and more attractive for a vast number of civil missions. Clearly, the introduction of UAVs in the civil airspace requiring practical and effective regulation is one of the most critical issues being currently discussed. As several civil air authorities report in their regulations "Sense and Avoid" or "Detect and Avoid" capabilities are critical to the successful integration of UAV into the civil airspace. One possible approach to achieve this capability, specifically for operations beyond the Line-of-Sight, would be to equip air vehicles with a vision-based system using cameras to monitor the surrounding air space and to classify other air vehicles flying in close proximity. This paper presents an image-based application for the supervised classification of air vehicles. First, several vehicle images, taken from different points of view, are transformed using a descriptor of salient features as to build the five-class database used to train the classification algorithm. Then, the latter compares the descriptor of a vehicle image taken from a random point of view to records in the database. With a positive match, the vehicle will be assigned to one of the following classes: a) civil transport aircrafts, b) military aircrafts, c) general aviation aircrafts, d) helicopters, and e) airships/hot air balloons. The paper provides a possible layout for the algorithm implementation and presents the outcome of several tests performed to evaluate its efficiency and possible exploitation. Indications useful to further studies are presented to help future researches.

A correct recognition of nonverbal expressions is currently one of the most important challenges of research in the field of human computer interaction. The ability to recognize human actions could change the way to interact with machines in several environments and contexts, or even the way to live. In this paper, we describe the advances of a previous study finalized to design, implement and validate an innovative recognition system already developed by some of the authors. It was aimed at recognizing two opposite emotional conditions (resonance and dissonance) of a candidate to a job position interacting with the recruiter during a job interview. Results in terms of the accuracy, resonance rate, and dissonance rate of the three new optimized neural networkbased (NN) classifiers are discussed. Comparison with previous results of three NN classifiers is also presented based on three single domains: facial, vocal and gestural.

Keywords: Facial/vocal/gestural features , Job interview , Neural networks and support vector machines , Nonverbal emotional recognition

Inexpensive, flicker free, flat, large, extremely bright LCD panels can be effectively used to conceal very large objects both stationary and moving. CCDC (Charge Coupled Device Camera) may capture the surrounding environment and reproduce it on the LCD screens. LCDs may easily be installed on the surface of the objects. It is also possible to use the video cards of personal computers and laptops to digitally process the image captured with the CCDS and obtain patterns to disguise the real appearance of the object or the way it is moving. The visual CCD (Camouflage, Concealment, and Deception) system so obtained is extremely effective. In this paper, it is demonstrated that the digital image processing techniques necessary to obtain a successful concealment are elementary and already embedded in the hardware of video card for real time image processing. It is also possible to superimpose patterns at defined frequencies that will obscure the image to the human observer especially in case of use of magnifying optics. The CCD effect in the visual and ultralight field is extremely effective. An example of active CCD of the Ariete MBT (Main Battle Tank) is shown in this paper.

Keywords: Graphically enhancement , Visual concealment

Pilling is an undesired defect of textile fabrics, consisting of a surface characterized by a number of roughly spherical masses made of entangled fibers. Mainly caused by the abrasion of fabric surface occurring during washing and wearing of fabrics, this defect needs to be accurately controlled and measured by companies working in the textile industry. Pilling measurement is traditionally performed using manual procedures involving visual control of fabric surface by human experts. Since the early nineties, great efforts in developing automatic and non-intrusive methods for pilling measurement have been made all around the world with the final aim of overcoming traditional, visual-based and subjective procedures. Machine Vision proved to be among the best options to perform such defect assessment since it provided increasingly performing measurement equipment and tools, serving the purpose of automatic control. In particular, a relevant number of interesting works have been proposed so far, sharing the idea of helping (or even replacing) traditional measurement methods using image processing-based ones. The present work provides a rational and chronological review of the most relevant methods for pilling measurement proposed so far. This work serves the purposes of 1) understanding whether today's automatic machine vision-based pilling measurement techniques are ready for supplanting traditional pilling measurement and 2) providing textile technology researchers with a bird's eye view of the main methods studied to confront with this problem.

Keywords: Artificial neural networks , Fabrics , Image processing , Machine vision , Pilling assessment , Review

Textile companies usually manufacture fabrics using a mix of pre-colored fibers according to a traditional recipe based on their own experience. Unfortunately, mainly due to the fibers dyeing process, the colorimetric distance between the obtained fabric and the desired one results unsatisfactory with respect to a colorimetric threshold established by the technicians. In such cases, colorists are required to slightly change the original recipe in order to reduce the colorimetric distance. This trial and error process is time-consuming and requires the work of highly skilled operators. Computer-based color recipe assessment methods have been proposed so far in scientific literature to address this issue. Unlikely, many methods are still far to be reliably predictive when the fabric is composed by a high number of components. Accordingly, the present work proposes two alternative methods based on Kubelka-Munk and subtractive mixing able to perform a reliable prediction of the spectrophotometric response of a fabric obtained by means of any variation of a recipe. The assessment performed on a prototypal implementation of the two methods demonstrates that they are suitable for reliable prediction of fabric blends spectral response.

Keywords: Computer-based color assessment , Fabric blend , Kubelka-Munk theory , Spectro-photometer

Remotely Operated underwater Vehicles (ROVs) play an important role in a number of operations conducted in shallow and deep water (e.g.: exploration, survey, intervention, etc.), in several application fields like marine science, offshore construction, and underwater archeology. ROVs are usually equipped with different imaging devices, both optical and acoustic. Optical sensors are able to generate better images in close range and clear water conditions, while acoustic systems are usually employed in long range acquisitions and do not suffer from the presence of turbidity, a well-known cause of coarser resolution and harder data extraction. In this work we describe the preliminary steps in the development of an opto-acoustic camera able to provide an on-line 3D reconstruction of the acquired scene. Taking full advantage of the benefits arising from the opto-acoustic data fusion techniques, the system was conceived as a support tool for ROV operators during the navigation in turbid waters, or in operations conducted by means of mechanical manipulators. The paper presents an overview of the device, an ad-hoc methodology for the extrinsic calibration of the system and a custom software developed to control the opto-acoustic camera and supply the operator with visual information.

Keywords: 3D opto-acoustic camera , Optical and acoustic data fusion , ROV guidance , Stereovision system , System calibration