Image Processing & Applications

Two different closed-cell aluminium foams, manufactured by compact powder technology starting from 7075 and 6061 alloys, were studied with the aim of investigating the effect of heat treatments on their axial crushing behaviour. By means of scanning electron microscopy/energy-dispersive spectroscopy analyses, the crushing response was investigated in an attempt to understand how heat treatments may improve the absorbing efficiency of crashboxes. © 2009 Acta Materialia Inc.

Keywords: Aluminium alloy , Aluminium foam , Crashworthy design , Heat treatment , Microstructure

Abstract: This paper presents some experiments on different image processing techniques combined and adapted to define a procedure aimed at improving the readability of ancient degraded documents. The focus is on documents affected by the bleed-through effect, which significantly reduces readability. The availability of the recto and verso scans of any single page is assumed. The proposed procedure is composed of three steps: recto-verso registration, image enhancement through statistical decorrelation and, finally, image segmentation. As the registration step can strongly influence the results obtained during the subsequent phases, we carried out a comparison of different registration algorithms, in order to evaluate their overall effectiveness on recto-verso processing. © 2009 IEEE.

Keywords: Ancient documents. , Binarization , Image enhancement , Registration

Chronic wounds represent one of the most frequent pathologies affecting elderly people. In the last few years, several methods have been proposed with the aim at developing systems able to produce automatic and objective assessment of skin ulcers. Despite that, none of the existing technologies seems to be robust enough to work for all the ulcer typologies. In the majority of the clinical contexts, the evaluation of chronic wounds still depends on the experience of physicians rather than on the use of numerically objective attributes. This paper presents a non-invasive methodology to allow automatic measurements of chronic wounds. The proposed system involves the development of a 3D optical scanner, based on a structured light approach, combined with an infrared detector. This arrangement permits the acquisition of geometrical data including both chromatic texture and temperature information. A tool for the automatic detection of damaged skin areas has been developed by segmenting both color and thermal images, making possible to reliably quantify parameters characterizing wounds, such as area, depth and volume, even in absence of a perceptible color characterization. The described technology has been finally experienced in medical environment, proving its effectiveness for an objective assessment of wound healing.

Keywords: 3D segmentation , 3D thermography , Imaging , Wound modelling

Three different aluminium foams, manufactured by compact powder technology starting from 7075, 6061 and AlSi7 alloys were studied by performing microstructural and morphometric analyses, with the aim of explaining their different behaviour during axial crushing. Void distribution coupled with material microstructure justifies the behaviour of load-displacement curves obtained during axial crushing of the foams. The results show that 7075 alloy seems to be the material having the best behaviour during crushing, at least when the foam is removed of the external walls. Despite that outer skin presence coupled with the intrinsic brittle behaviour of this alloy may cause instability, if it is used to fill hollow components like crashboxes. During deformation process 6061 and AlSi7 alloys that are more ductile, give in and maintain contact adapting to the encasement deformation. © 2007 Elsevier B.V. All rights reserved.

Keywords: Aluminium alloy , Aluminium foam , Crashworthy design , Heat treatment , Microstructure

Recycling of clothes is a straightforward approach for the supply of a coloured raw material which does not involve the cost of the colouring process. A real time and completely automated colour classification tool for woollen clothes to be recycled is proposed. The tool uses the combination of a statistical method, called matrix approach, of a self-organizing feature map (SOFM) and a feed-forward backpropagation artificial neural network (FFBP ANN)-based approach, to correctly classify the clothes by respecting the selection criteria provided by human know-how. The developed tool, which uses an appositely developed workbench with a spectrophotometer, is aware of the way the different coloured clothes to be recycled combine each other to create a new one. The tool has been validated using a set of 5,000 differently coloured clothes to be recycled and the classification error in classifying the clothes is within 5%, i.e., lower than the one resulting from the use of an expert human operator. © 2007 Springer-Verlag London Limited.

Keywords: Colour , Neural networks , Picking , Recycling , Spectrophotometry

This work describes an automated artificial vision inspection (AVI) system for real-time detection and classification of defects on textile raw fabrics. The tool (software + hardware) is directly attached to an appositely developed appraisal equipment machine (weave room monitoring system) and the inspection is performed online. The developed tool performs (1) the image acquisition of the raw fabric, (2) the extraction of some critical parameters from the acquired images, (3) an artificial neural network (ANN)-based approach able to detect and classify the most frequently occurring types of defects occurring on the raw fabric and (4) a standard image processing algorithm that allows the measurement of the geometric properties of the detected defects. The reliability of the tool is about 90% (defect detected vs. effectively existing defects), that is, similar to the performance obtained by human experts. Once detected the defects are correctly classified in 88% of cases and their geometrical properties are measured with a sub-pixel precision.

Keywords: Artificial neural network , Image processing , Raw fabrics , Real time

This work provides a software based on the combination of a artificial neural network (ANN) approach and a numerical method (called "refining algorithm") for the estimation of the acidity level and of peroxides number of olive oil extracted by a continuous extraction process. The estimation is achieved through the measurement of some agronomical and technological parameters commonly measured by the technicians working at the oil mills. The ANN based approach is able to perform a rough prediction of the two parameters; this prediction is then refined by means of the numerical approach. The devised software, developed using a database of experimental data collected in the years 1997-2004 in the Tuscany Region (Italy), is able to estimate the two parameters during the continuous oil extraction, thus allowing a real-time control of the oil quality during the extraction process. The experimental data are related to different kinds of extraction systems: the data collected before 2003 are mainly referred to three phases decanting while the one collected in years 2003-2004, are referred to a two phases decanter. The goal of this approach is to build up a fast and reliable quality control without performing the standard laboratory analysis. The results of the estimation achieved by the ANN based system are compared with the results of the chemical analyses carried out by Florence Commerce Chamber "Laboratorio Chimico Merceologico-Azienda Speciale CCIAA di Firenze" according in force to European Union Rules standards. The mean error in estimation of the parameters with the combination of the ANN approach and the refining algorithm, in comparison with the standard chemical analysis, is in the range 6-8% for an oil extraction with a three phases decanting system and about 6-7% for an oil extraction with a two phases decanting system. The maximum error affecting the method (ANN + Refining) described in this work is about 15% for the first case and about 11% for the second one. The system developed in the present work is actually running on the oil mill "TEM Toscana Enologica Mori" of Florence, Italy. The present work has been financed by the Tuscany Regional Agricultural Development and Innovation Office (ARSIA: Azienda Regionale per lo Sviluppo e l'Innovazione dell'Agricoltura) and is a part of a 3-year project whose objective is to create an entirely software + hardware controlled oil mill. © 2007 Elsevier B.V. All rights reserved.

Keywords: Artificial neural network , Olive oil , Olive oil acidity level , Olive oil centrifugal process , Peroxides number

The sequential approach to colour texture classification relies on colour histogram clustering before extracting texture features from indexed images. The basic idea of such methods is to replace the colour triplet (RGB, HSV, Lab, etc.) associated to a pixel, by a scalar value, which represents an index of a colour palette. In this paper we studied different implementations of such approach. An experimental campaign was carried out over a database of 100 textures. The results show that the choice of a particular colour representation can improve classification performance with respect to grayscale conversion. We also found strong interaction effects between colour representation and feature space. In order to improve accuracy and robustness of classification, we have tested three well known expert fusion schemes: weighted vote, and a posteriori probability fusion (sum and product rules). The results demonstrate that combining different sequential approaches through classifier fusion is an effective strategy for colour texture classification. © Springer-Verlag Berlin Heidelberg 2007.

Keywords: Classifier fusion , Colour texture classification

Gabor filtering is a widely adopted technique for texture analysis. The design of a Gabor filter bank is a complex task. In texture classification, in particular, Gabor filters show a strong dependence on a certain number of parameters, the values of which may significantly affect the outcome of the classification procedures. Many different approaches to Gabor filter design, based on mathematical and physiological consideration, are documented in literature. However, the effect of each parameter, as well as the effects of their interaction, remain unclear. The overall aim of this work is to investigate the effects of Gabor filter parameters on texture classification. An extensive experimental campaign has been conducted. The outcomes of the experimental activity show a significant dependence of the percentage of correct classification on the smoothing parameter of the Gabor filters. On the contrary, the correlation between the number of frequencies and orientations used to define a filter bank and the percentage of correct classification appeared to be poor. © 2007 Pattern Recognition Society.

Keywords: Design of experiments , Gabor filters , Texture classification

A reliable prediction of ductile failure in metals is still a wide-open matter of research. Several models are available in the literature, ranging from empirical criteria, porosity-based models and continuum damage mechanics (CDM). One major issue is the accurate identification of parameters which describe material behavior. For some damage models, parameter identification is more or less straightforward, being possible to perform experiments for their evaluation. For the others, direct calibration from laboratory tests is not possible, so that the approach of inverse methods is required for a proper identification. In material model calibration, the inverse approach consists in a non-linear iterative fitting of a parameter-dependent load-displacement curve (coming from a FEM simulation) on the experimental specimen response. The test is usually a tensile test on a round-notched cylindrical bar. The present paper shows a novel inverse procedure aimed to estimate the material parameters of the Gurson-Tvergaard-Needleman (GTN) porosity-based plastic damage model by means of experimental data collected using image analysis. The use of digital image processing allows to substitute the load-displacement curve with other global quantities resulting from the measuring of specimen profile during loading. The advantage of this analysis is that more data are available for calibration thus allowing a greater level of confidence and accuracy in model parameter evaluation. © Springer Science+Business Media B.V. 2007.

Keywords: Damage mechanics , Image analysis , Inverse methods , Machine design , Parameter identification

A real-time and robust algorithm for 3D camera registration in a Mixed Reality (MR) environment is described in this paper. The most used technique for camera pose (position and orientation with respect to a fixed or moving object) is based on fiducial marker tracking. This method guarantees good results in real-time with a single camera, but needs several high contrast printed markers on external world in order to make possible the calculation of camera parameters and positioning. Thus real 3D geometric data are grabbed only through already known markers. The aim of this research is a real-time monocular camera tracking and registration through automatic image features extraction from video streaming. The first implementation of the method, several examples and confrontation with non interactive algorithm for SFM (Structure From Motion) have demonstrated that this meets the real-time response and sufficient precision needed by a Mixed Reality environment. © The Eurographics Association 2007.

This paper deals with the calorimetric analysis of fatigue mechanisms of an aluminium alloy. Local 1D and 2D expressions of the heat diffusion equation were used to separately estimate the coupling and dissipative sources accompanying the fatigue test. The image processing was based on local approximations of surface temperature fields provided by an IR camera. The set of approximation functions accounted for the spectral properties of the sought sources. The analysis of thermoelastic source distribution allowed us to point out the possible heterogeneous character of fatigue loading from the beginning of the test. We observed that the zones where thermoelastic source amplitudes were the highest were potential loci of fatigue crack onset. On the other hand, such fine results on the dissipative source distribution were hard to obtain because of low signal-to-noise ratios. Nevertheless, we observed that the global dissipated energy rates increased linearly with the loading frequency. © 2006 Elsevier Ltd. All rights reserved.

Keywords: Aluminium alloy , Calorimetric analysis , Fatigue dissipation , Thermal images , Thermoelastic sources

The present work presents the devising of a new highly automated artificial vision inspection (AVI) tool for real-time defect detection and classification on circular knitting machines. The tool is based on the combination of statistical analysis, Image Processing and an artificial neural network (ANN) approach. The tool (software + hardware) is directly attached to a circular knitting machine and the inspection is performed on-line during other common operations like laser cutting and ironing. The automatic inspection allows the real-time detection and classification of the most frequently occurring types of defects on knitted fabrics, which are significant for purposes of quality control and fabric grading. The reliability of the detection tool, i.e. the ratio between defect detected and effective defects is about 93%. The AVI system has been developed by the Department of Mechanical Engineering, University of Florence (Italy), and the textile research centre Tecnotessile s.r.l. of Prato (Italy).

Keywords: Image processing , Knitting machines , Neural network , Radon transform , Real time , Skewness

This work provides an automatic and non-intrusive tool to objectively monitoring the raising process by measuring the height and the density of the fibres emerging from a raised cloth (pile). These parameters are assessed by a numerical procedure, which elaborates the images provided by an appositely developed machine-vision system. The proposed approach allows the investigation and the control of the raising process and has been validated by experimental measurements performed on a set of specimens (cloths) with several raising degrees. The comparison between the results obtained by the proposed procedure and the ones coming from a widely accepted textile-measuring device (fabric assurance by simple testing, FAST) is also provided. © 2005 Elsevier B.V. All rights reserved.

Keywords: Cloth raising , Image processing , Machine-vision design , Pile density , Pile height

In this paper the results of an experimental investigation on the effect of subcritical damage on the residual strength properties of notched composite laminates are presented. A procedure based on the digital image correlation method was applied to laminates subjected to static and fatigue tensile loading. The digital image correlation method (DICM) is a whole-field technique that calculates surface displacements and strains from digital images characterized by a random distribution of intensity grey levels. Graphite/PEEK (polyether ether ketone) and graphite/epoxy laminates with different stacking sequences were analysed and the damage progression near the stress riser was evaluated by means of the strain maps obtained by digital image correlation. It was found that damage developing before final fracture may significantly affect the structural performance of composite laminates. The digital image correlation technique allowed clarification of the beneficial or detrimental role played by the different failure mechanisms on the strain redistribution around the hole and, as a consequence, on the residual strength and fatigue life of notched samples. The findings of the investigation suggest that the DICM is an efficient and reliable tool for full-field monitoring and detailed damage characterization of structural composite elements. © IMechE 2005.

Keywords: Composite materials , Damage , Digital image correlation , Fatigue , Notch

Research was carried out by Tecnotessile Societa Nazionale di Ricerca Tecnologica r.l. in collaboration with the Department of Mechanics and Industrial Technology of the University of Florence to develop a methodology and software for measuring the pile depth and pile density of fabrics in real time. The research concentrated on napped wool and wool blend fabrics. The method is based on the use of artificial vision and was compared with results produced by FAST testing. The results of pile depth measurement are reported in the present article. The research comprised the following phases: HR image acquisition, HR image processing, processing of numerical information and definition of required values, low resolution image acquisition and processing, creation of Feed Forward Back Propagation neural network, and development of suitable software.

Experimental, analytical and numerical methods of measuring the surface density of the pile of napped fabrics are presented. Surface density of pile is defined as the ratio between the weight of the pile extending above the fabric and the area of the sample to which it belongs. Samples were weighed, shaved and reweighed and the area of the sample was calculated from the number of pixels on a high-resolution image. The density of the fibre removed was estimated as the difference in density of the sample before and after shaving. A non-intrusive analytical method based on the height, diameter and number of pile fibres and a numerical method are described. Calculations based on pure wool and wool blend are presented and compared with the results obtained using the experimental method.

This paper presents a 3D optical digitiser based on an active stereo vision aproach. The system is composed of a digital camera, a multimedia projector and software to control the hardware and process the images. The proposed solution integrates an active coded light method with a photogrammetric procedure in order to allow the acquisition of complex surfaces. In the work, experimental tests have been conducted with a nominal model and styling components of two-wheeler vehicles. The measurement process and the experimental results have been analysed to verify usability and accuracy of the methodology.

Keywords: Coded light , Photogrammetry , Reverse Engineering , Stereovision

The planning of the hip prosthesis surgical operation is usually performed manually by the surgeon, who 'draws' on a patient's X-ray image the outline of the prosthetic stem in order to choose the one most suitable for the case at hand. In an attempt to give some repeatability and objectivity to the planning phase, a procedure has been devised for hip prosthesis' stem selection based on the extraction of the femoral relevant outlines. This work presents a computer aided method aimed at automatically extracting the medulla duct outlines from a human femur radiographic image. The outlines are retrieved by referring to a suitable geometric model of the generic femoral cross-section; the projection function obtained by simulating the radiographic acquisition of such a model is fitted on the grey-level functions corresponding to the rows of the actual digitised radiographic image by means of a least squares algorithm. The resulting outlines are used in a software tool performing the hip prosthesis pre-operational planning. © 2003 Published by Elsevier B.V.

Keywords: Medical images processing , Model-based image processing , Surgery aiding tools