Image Processing & Applications

We present a novel framework for computer aided measurements which supports physicians during 3D virtual colonoscopy. We propose a four steps methodology: (i) segment colon regions, (ii) search for colorectal polyps by means of a supervised neural network architecture, (iii) 3D reconstruction of polyp geometry using a synthetic model and (iv) qualitative and quantitative geometrical analysis. Doctor can graphically visualize and inspect the surface and search for irregularities. This qualitative visual support is extended with mathematical tools which computes diagnostic indicators. The proposed tool was tested with two different data sets of polyps. The proposed approach can be an effective diagnostic tool to evaluate the degree of malignancy of the cancer. Due to its generality, the presented framework can be easily extended to search for other types of cancers. © 2011 IEEE.

Keywords: 2D segmentation , 3D rendering and 2D and 3D measurements , neural networks , polyps , Virtual Conolonoscopy

Nowadays three-dimensional Computer Aided modeling is of out-standing importance in the mechanical design process since it impacts on several issues like visualization, simulation, machining, etc. Anyway, multi orthographic view engineering drawings have been widely used up to latest decade and still are, so they play an essential role in traditional engineering. The conversion from 2D drawings to 3D CAD models is still a key task in a wide range of applications. In order to cope with this issue a number of works have been proposed in the last decades, providing a series of methodologies for solving the reconstruction prob-lem. On the basis of such methodologies the main aim of the present paper is to suggest a comprehensive, orderly, unambiguous and automatic procedure meant to help researchers and practitioners who want to deal with the reconstruction problem. The procedure, by using an appropriate formal mathematic language, systematize and integrates some of the methods proposed so far. © 2011 Heldermann Verlag.

Keywords: 3D reconstruction , Computational geometry , Computer aided design , Engineering drawings , Or-thographic projections , Pseudo-wireframe

Marbling effect on fabrics is a relevant aesthetic feature, increasing its diffusion specially in the field of textiles for technical applications. The fabric aesthetic anisotropy, characterizing the marbling effect, has a strong impact on the perceived quality: a high-quality marbled fabric to be used in automotive textiles, for instance, is characterized by a tiny quantity of veins and spotted areas. A large amount of "veins" and/or discolored areas may induce a customer to consider the fabric as "defected". In common practice, the identification of whether the fabric is defective or not is performed by human experts by means of visual inspection. As a consequence, fabric inspection is performed in a qualitative and unreliable way; thereby the definition of a method for the automatic and objective inspection is advisable. On the basis of the state of the art, the present work aims to describe a computer-based approach for the automated inspection of marbling effect on fabrics, resulting in the classification of fabrics into three quality classes. The devised apparatus is composed by a machine vision system provided with an image processing-based software. The processing software is able to determine the anisotropy of a fabric using edge segmentation and image entropy and defining a "fabric entropy curve" The proposed method proves to be able to classify the fabrics into the correct quality class in 90% of the cases, with respect to the selection criteria provided by human operators.

Keywords: Artificial vision , Edge detection , Fabric inspection , Image entropy , Marbling effect

Digital applications such as CG, CAD and GIS are based on vectorial data since all the information about shape, size, topology etc. are provided in such kind of data representation rather than raster one. Turning raster images into vector ones is a key issue which has been addressed by a number of authors but still far to be exhaustively worked out. Especially in the case of 2D images representing technical drawings, fitting analytical curves to point clouds (pixel sets) is a critical matter. The present paper provides a novel approach to fit unordered point cloud data. Such an approach integrates a PCA-based method, for detecting the main local directions of the point cloud and to order the points, with and a weighted approximation of a B-spline curve to the original data, based on pixel gray levels. The methodology, tested against alternative techniques based on Least Square (LS) B-spline approximation and on image thinning, proved to be effective in preserving the original shape according to human perception.

Keywords: Curve reconstruction , Image processing , PCA , Unorganized points , Weighted least-squares

One of the most critical phases of the production of carded fiber is the comparison between the colour of the finished product and the colour desired by a customer or provided in a catalog. In most cases, the required proportions of coloured fibers for obtaining a desired final colour of a fiber are evaluated by trial and error techniques. This approach, which often lacks a computer-aided tool to assist colourists, allows the assessment of a recipe to reproduce a target colour. The companies producing carded fibers often perform a colorimetric control of their product. This is obtained by evaluating the colorimetric distance between the final product, obtained by mixing some coloured raw fibers in order to obtain a homogeneous colour, and the desired one. When this colorimetric distance exceeds a value stated by the customer, the companies have to modify the recipe in order to reduce the gap between the colour of the final product and the desired one. The present work describes a novel approach for a reliable prediction of the spectrophotometric response of a carded fiber based on the spectrophotometric response of the raw materials composing the fiber itself. This is obtained for any variation of an original recipe. The developed system is, first, based on a theoretical approach used for the prediction of reflectance factors obtained by mixing the same kind of differently coloured raw material. A more general approach, based on Artificial Neural Networks (ANNs), is then described. This second approach performs a reliable prediction of the reflectance factors of a carded fiber composed by several kind of differently coloured raw materials. The two approaches were used to predict the reflectance factors of 500 different blends. Colour differences between the predicted reflectance factors and the real spectrophotometric response of the carded fibers are measured and expressed in CIELAB and CMC(2:1) distances under several standard illuminants. Referring to the test performed in the present work, the mean predicted colour difference in terms of CIELAB is less than 0.24. The maximum distance in terms of CMC(2:1) is less than 0.25. Experimental validation demonstrates that the proposed approach is suitable for a reliable, fast and practical prediction of the recipes for carded fibers. For these reasons, the developed system is actually running into the Laboratory of an important textile company located in Prato (Italy): the New Mill S.p.A. © 2010 Wiley Periodicals, Inc.

Keywords: artificial neural network , carded fiber , spectrophotometry

An automatic 3D model retrieval from freehand conceptual sketches is a key target for both commercial software houses and academic research. Unfortunately, most of the approaches are not suitable for properly translating stylistic sketches into 3D models. In order to carry out this 3D model conversion, the first task to be dealt with is to turn raster data (3D or 2D free-form curves) into vectorial ones. Such a task represents a key issue which has been addressed by a number of authors but still far to be exhaustively worked out. To address this challenge, this work presents a new method that allows to fit 2D unordered point cloud data with Multiple Incident Splines (MISs). At the heart of the proposed approach are two main procedures: the first one is based on Euclidean Minimum Spanning Tree (EMST) and Principal Component Analysis (PCA) for detecting the main local directions of the point cloud and to order its points while preserving original topology; the second is meant to fit ordered point clouds with spline curves providing a robust intersection and vertex detection. The proposed methodology, tested on a number of case studies, proves to preserve the original topology more efficiently than alternative techniques supplied by commercial vectorization software packages.

Keywords: Curve reconstruction , Freehand sketches , Image Processing , Styling , Unorganized point cloud

Car seat fabrics are uniquely fashioned textiles. A number of them is branded by a sponged-like appearance, characterized by spots and slightly discoloured areas. Their surface anisotropy is considered to be a relevant aesthetic feature since it has a strong impact on customer perceived quality. A first-rate car seat fabric requires a "small" quantity of spots and discoloured areas while fabrics characterized either by a large number or by a low number of spots, are considered to be of lower quality. Therefore, car seat fabric quality grading is a relevant issue to be dealt with downstream to the production line. Nowadays, sponged-like fabric grading is performed by human experts by means of manual inspection and classification; though this manual classification proves to be effective in fabric grading, the process is subjective and its results may vary depending on the operator skills. Accordingly, the definition of a method for the automatic and objective grading of sponged-like fabrics is necessary. The present work aims to provide a computer-based tool capable of classifying sponged-like fabrics, as closely as possible to classifications performed by skilled operators. Such a tool, composed by an appositely devised machine vision system, is capable of extracting a number of numerical parameters characterizing the fabric veins and discoloured areas. Such parameters are, then, used for training an Artificial Neural Network (ANN) with the aim of classifying the fabrics in terms of quality. Finally, a comparison between the ANN-based classification and the one provided by fabric inspectors is performed. The proposed method, tested on a validation set composed by 65 sponged-like fabrics, proves to be able to classify the fabrics into the correct quality class in 93.8% of the cases, with respect to the selection provided by human operators.

Keywords: Artificial neural networks , Car seat fabrics , Grading , Machine vision

Classification of wasted woollen textiles on the basis of their colour is a basic approach for the supply of a raw material which does not involve the cost of the colouring process. Colour classification is a very difficult task, especially when a fabric is composed by differently coloured fibre (melange fabric). Many systems have been developed in the last years for colour classification of textiles. Unfortunately such colour classification systems are not able to correctly classify melange fabrics. In the present work a method for real-time classification of melange colour woollen fabrics is proposed. The provided approach, that is suitable also for classifying solid colour fabrics, integrates a Machine Vision (MV) system, able to acquire high resolution images, with a clustering algorithm capable of mapping the colour pixel of fabric images into a series of colour classes. The proposed system provides a colour classification with a misclassification less than 10% when compared with the classification resulting from apanel of expert human operators. A comparison between the proposed method and some tools stated in scientific literature is also afforded. © 2011 Asian Network for Scientific Information.

Keywords: Machine vision , Mapping , Melange , Recycling , Textiles

In the field of 3D scanning, there is an increasing need for more accurate technologies to acquire 3D models of close range objects. Underwater exploration, for example, is very hard to perform due to the hostile conditions and the bad visibility of the environment. Some application fields, like underwater archaeology, require to recover tridimensional data of objects that cannot be moved from their site or touched in order to avoid possible damages. Photogrammetry is widely used for underwater 3D acquisition, because it requires just one or two digital still or video cameras to acquire a sequence of images taken from different viewpoints. Stereo systems composed by a pair of cameras are often employed on underwater robots (i.e. ROVs, Remotely Operated Vehicles) and used by scuba divers, in order to survey archaeological sites, reconstruct complex 3D structures in aquatic environment, estimate in situ the length of marine organisms, etc. The stereo 3D reconstruction is based on the triangulation of corresponding points on the two views. This requires to find in both images common points and to match them (correspondence problem), determining a plane that contains the 3D point on the object. Another 3D technique, frequently used in air acquisition, solves this point-matching problem by projecting structured lighting patterns to codify the acquired scene. The corresponding points are identified associating a binary code in both images. In this work we have tested and compared two whole-field 3D imaging techniques (active and passive) based on stereo vision, in underwater environment. A 3D system has been designed, composed by a digital projector and two still cameras mounted in waterproof housing, so that it can perform the various acquisitions without changing the configuration of optical devices. The tests were conducted in a water tank in different turbidity conditions, on objects with different surface properties. In order to simulate a typical seafloor, we used various concentrations of clay. The performances of the two techniques are described and discussed. In particular, the point clouds obtained are compared in terms of number of acquired 3D points and geometrical deviation.

Keywords: Dense reconstruction , Stereo vision , Structured-light , Underwater imaging

This paper presents a new 3D design paradigmfor the development of specific custom-fit products, such as the soft socket of prostheses for lower limb amputees. It is centered on the digitalmodel of the human body and, contrarily to the traditional process almost manually based, it considers the integration of methods and tools coming from different research and application fields: Reverse Engineering, Medical Imaging, Virtual Prototyping, Physics-based Simulation, and Rapid Prototyping. The paper describes the techniques adopted to acquire and create the digital model of the residual limb, the procedure to generate the socket model, the strategy developed for the functional simulation of the socket-stump interaction and, finally, the realization of the physical prototype. Each design step is described with the related problems and the obtained results. Both trans-tibial and trans-femoral amputees have been considered; however, for now the complete process has been validated for trans-tibial prostheses. Major outcomes of the proposed approach stand in a better quality of the final product, in a shorter involvement of the amputee implying a lower psychological impact, in a limited use of physical prototypes, and in a shorter development time. Moreover, the resulting paradigm answers to the Collaborative Engineering guidelines by optimizing the interaction between different domains and enhancing their contributions in a homogeneous development framework. © 2010 Elsevier B.V. All rights reserved.

Keywords: Custom-fit products , Human-centric soft product design , Physics-based modeling for soft products , Prosthesis development process

Color match prediction is one of the most important aspects to be considered by industries dealing with colorants. Several generally applicable theoretical models have been proposed so far for helping the colorists in achieving an exact color match. Such approaches, often based on extensive experimental tests and provided of exhaustive results, are differentiated by a specific range of application (textiles, study, paintings, etc.). Therefore, the results are subjected to restrictions or constraints (number of colorants, reliability of the prediction, etc.). The present paper describes, into a mathematical form, three widely known techniques adopted in the scientific literature for evaluating the spectrophotometric color match prediction of a target shade: Kubelka-Munch, Steams-Noechel and Artificial Neural Networks. The proposed method starts from such wide known methodologies and by means of mathematical assessment provides some useful equations to be straightforwardly used for color matching. Moreover an Artificial Neural Network based formulation is provided. The results of the work shows that the expected color distance between the predicted the real color of a shade is less than 0.8, in terms of CTEL*a*b* distance. © 2010 Asian Network for Scientific Information.

Keywords: Absorptivity and the scattering , Color distance , Color formulation , Color matching transfer function , Reflectance

The color and the color stability of a fabric dyed with a jigger machine are highly dependent on the color of the dyes and on technological parameters, such as the process temperature, pH of the dyebath, and time of dyeing. On the basis of their skill, colorists are capable of choosing the technological parameters and the recipes for dyes in order to obtain the desired dyed fabric. However, they need to perform many tests before reaching the preferred product. As a matter of fact, a straightforward relationship between the color of dyes, process parameters, and final color and color solidity is not found in literature. The present work aims to help the colorist in predicting the color and the color stability of the fabric, without the needing to physically dye it. In detail the paper describes a tool, based on the Cascade Neural Network (CNN), that is able to receive the value of some colorimetric and technological parameters as input and to predict, as output, the color appearance and the color solidity of a fabric dyed with a jigger machine. The CNN is composed of a “color prediction module” (ANN I) and a “color solidity prediction module” (ANN II) that work in cascade. The CNN has been validated by means of a large set of experiments. The mean error between the color prediction and the real values of the dyed fabric, in terms of CIE 1976 (L*, a*, b*) color distance, is equal to 0.47 with a variance of 0.031. The maximum color solidity prediction error is 0.5. © 2010, SAGE Publications. All rights reserved.

In this paper we study the feasibility of developing a search engine capable of retrieving images from a granite image database based on a query image that is similar to the intended targets. The main focus was on the determination of the set of colour and/or texture features which yields highest retrieval accuracy. To assess the performance of the considered image descriptors we created a granite image database, formed by images recorded at our laboratory as well as taken from the Internet. Experimental results show that colour and texture features can be successfully employed to retrieve granite images from a database. We also found that improved accuracy is achieved by combining different colour and texture feature sets through classifier fusion schemes.

Keywords: CBIR , Colour , Granite , Image retrieval systems , Texture , Visual appearance

Style sketches are an essential tool for the expression and the definition of the shape of industrial products. We present a novel wavelet approach for sketch segmentation and editing expressly addressed to the designer needs. Starting from a 2D curve drawn by a user with a graphics tablet, our algorithm converts it in a B-spline representation, detects the style features, allows interactive smoothing and continuous local and global editing. The novelty compared to the existing literature is the combination of: (i) unified wavelet approach to segmentation, smoothing and editing, (ii) support for endpoint C0 and C1 continuity constraints and (iii) implementation with an interactive pen-tablet interface. Our approach requires low memory footprints with time complexity linear in the number of control points. We validated the algorithm with synthetic and practical tests. © 2010 World Scientific Publishing Company.

Keywords: 2D interactive sketch , multi-level segmentation , multiresolution editing

The influence of ripening degree of drupes during the harvesting period is well established in olive oil sector. A range of methods for expressing the stage of maturity of olives have been proposed in scientific literature. One of the most commonly adopted methods provides the evaluation of a Ripening Index (RI) on the basis of olive skin and pulp colour. Unfortunately, the RI evaluation technique is time-consuming, subjective (depending on expert skill) and depends on environmental conditions that may affect colour appearance of olives. This work describes a novel method for rapid, automatic and objective prediction of the Ripening Index of an olive lot. The method integrates a Machine Vision system, capable of performing a colour-based raw prediction of RI, with an Artificial Neural Network (ANN) based algorithm to refine it. Such a refinement is based on a set of chemical parameters (oil content, sugar content and phenol content) which are provided as input to the ANN and which can be obtained by historical curves for the region where the RI needs to be predicted. Experimental results demonstrate the effectiveness of the proposed approach. © 2010 Elsevier Ltd. All rights reserved.

Keywords: Artificial neural networks , Machine vision system , Ripening index

Historical documents often undergo various changes over time, that alter their original state and reduce their legibility. Digital techniques are widely applied for document preservation, archiving, analysis and dissemination over the Internet. The goal of restoration is to improve the document legibility for both human and automatic operators, to extract patterns and enhance colour reproduction. Our paper presents a framework for virtual restoration of ancient documents based on a combination of multispectral acquisition, 3D imaging and digital image analysis. The proposed framework consists of several steps. First, digital representations of the documents are acquired as multispectral images and 3D surface maps, the latter reconstructed by a structured light technique. A mul-tispectral camera and a digital projector are used in triangular configuration for 2D and 3D data acquisition. Then the multispectral images are registered against possible misalignments, and the 3D surface representation is used to correct geometrical distortions. Document flattening is then performed by 3D surface parameterization and texture mapping. Statistical techniques of decorrelation are applied to extract individual context parts of the document patterns (stamp, text, etc.) and to attenuate interferences. The processed data are then binarized by the proper segmentation technique. The entire digital object history - all the acquisition and processing steps, with the corresponding parameters - is stored in metadata files. These data can be exploited in a future evaluation of the restoration process and can be used for either the creation of an expert knowledge database or the extraction of cross-document observations and conclusions. © The Eurographics Association 2010.

In this paper, the displacement field induced by the split-sleeve cold expansion of holes was measured using both digital image correlation (DIC) and digital speckle pattern interferometry (DSPI) techniques. Thus, the experimental results, which were evaluated on the inlet surface of a 6082-T6 aluminium plate, were compared with those from theoretical prediction. DIC provided accurate measurements up to the elastic-plastic boundary, whereas the DSPI technique highlighted the changes of displacement in the elastic domain. Prediction of the displacement based on the existing analytical model agreed with the experimental results achieved with both techniques. Possible explanations for the differences are discussed.

Keywords: cold expansion of holes , digital image correlation , digital speckle pattern interferometry , displacement measurement

We present a set of multiscale, multidirectional, rotation-invariant features for color texture characterization. The proposed model is based on the ranklet transform, a technique relying on the calculation of the relative rank of the intensity level of neighboring pixels. Color and texture are merged into a compact descriptor by computing the ranklet transform of each color channel separately and of couples of color channels jointly. Robustness against rotation is based on the use of circularly symmetric neighborhoods together with the discrete Fourier transform. Experimental results demonstrate that the approach shows good robustness and accuracy. © 2009 SPIE and IS&T.

The Coordinated Clusters Representation (CCR) is a texture descriptor based on the probability of occurrence of elementary binary patterns (texels) defined over a square window. The CCR was originally proposed for binary textures, and it was later extended to grayscale texture images through global image thresholding. The required global binarization is a critical point of the method, since this preprocessing stage can wipe out textural information. Another important drawback of the original CCR model is its sensitivity against rotation. In this paper we present a rotation-invariant CCR-based model for colour textures which yields a twofold improvement over the grayscale CCR: first, the use of rotation-invariant texels makes the model insensitive against rotation; secondly, the new texture model benefits from colour information and does not need global thresholding. The basic idea of the method is to describe the textural and colour content of an image by splitting the original colour image into a stack of binary images, each one representing a colour of a predefined palette. The binary layers are characterized by the probability of occurrence of rotation-invariant texels, and the overall feature vector is obtained by concatenating the histograms computed for each layer. In order to quantitatively assess our approach, we performed experiments over two datasets of colour texture images using five different colour spaces. The obtained results show robust invariance against rotation and a marked increase in classification accuracy with respect to grayscale versions of CCR and LBP. © 2009 Elsevier B.V. All rights reserved.

Keywords: CCR , Colour texture classification , Rotation invariance

The ability to utilize a conventional far field microscope to perform highly accurate measurements at the nano scale is presented. The free electrons on a metallic surface can be excited by evanescent waves producing surface plasmons. The surface plasmons are electromagnetic waves that by decaying can generate photons. These photons produce propagating waves that feed energy to an oscillating cavity. From this effect a Fabry-Perot like interferometer is generated. Essentially by using the transformation of evanescent waves into propagating waves via the plasmon generation it is possible to perform high accuracy contouring of any metal surface as well as the determination of its contact stresses. A laser beam illuminates a glass-air-metal interface and through proper setup it can produce double evanescent illumination. This illumination produces interference fringes that contain surface depth information as well as in-plane strain information. Some applications are discussed briefly to demonstrate the technique's unique capabilities. ©2009 Society for Experimental Mechanics Inc.