Nombre total de résultats : 503
Pertinence Titre A-Z Plus récents Plus anciens
10 25 50
Année de publication

Unsupervised weld defect classification in radiographic images usingmultivariate generalized Gaussian mixture model with exactcomputation of mean and shape parameters

Nafaa Nacereddine, Aicha Baya Goumeidane, Djemel Ziou  (2019)

In industry, the welding inspection is considered as a mandatory stage in the process of quality assurance/quality control. This inspection should satisfy the requirements of the standards and codes governing themanufacturing process in order to prevent unfair harm to the industrial plant in construction. For thispurpose, in this paper, a software specially conceived for computer-aided diagnosis in weld radiographictesting is presented, where a succession of operations of preprocessing, image segmentation, featureextraction andfinally defects classification is carried out on radiographic images. The last operationwhich is the main contribution in this paper consists in an unsupervised classifier based on afinitemixture model using the multivariate generalized Gaussian distribution (MGGD). This classifier is newlyapplied on a dataset of weld defect radiographic images. The parameters of the nonzero-mean MGGDbasedmixture model are estimated using the Expectation-Maximization algorithm where, exactcomputations of mean and shape parameters are originally provided. The weld defect database representfour weld defect types (crack, lack of penetration, porosity and solid inclusion) which are indexed by ashape geometric descriptor composed of geometric measures. An outstanding performance of theproposed mixture model, compared to the one using the multivariate Gaussian distribution, is shown,where the classification rate is improved by 3.2% for the whole database, to reach more than 96%. Theefficiency of the proposed classifier is mainly due to theflexiblefitting of the input data, thanks to theMGGD shape parameter. Voir les détails

Mots clés : Mixture model, Multivariate GGD, radiography, weld defect, classification

Video Processing and Analysisfor Endoscopy-Based InternalPipeline Inspection

Nafaa Nacereddine, Aissa Boulmerka, Nadia MHAMDA  (2019)

Because of the increasing requirements in regards to the pipeline transport regulations, the operators take care to the rigorous application of checking routines that ensure nonoccurrence of leaks and failures. In situ pipe inspection systems such as endoscopy, remains a reliable mean to diagnose possible abnormalities in the interior of a pipe such as corrosion. Through digital video processing, the acquired videos and images are analyzed and interpreted to detect the damaged and the risky pipeline areas. Thus, the objective of this work is to bring a powerful analysis tool for a rigorous pipeline inspection through the implementation of specific algorithms dedicated to this application for a precise delimitation of the defective zones and a reliable interpretation of the defect implicated, in spite of the drastic conditions inherent to the evolution of the endoscope inside the pipeline and the quality of the acquired images and videos. Voir les détails

Mots clés : video processing, endoscopy, Pipeline inspection

Asymmetric Generalized Gaussian DistributionParameters Estimation based on MaximumLikelihood, Moments and Entropy

Nafaa Nacereddine, Aicha Baya Goumeidane  (2019)
Article de conférence

In this paper, we address the problem of estimatingthe parameters of Asymmetric Generalized Gaussian Distribution(AGGD) using three estimation mehods, namely, Maximum LikelihoodEstimation (MLE), Moment Matching Estimation (MME)and Entropy Matching Estimaion (EME). For this purpose, thesemethods are applied on an unimodal histogram fitting of animage corrupted with AGGD noise. Experiments show that theeffectiveness of each method comparatively to the other onedepends on the variation range of the shape factor. Voir les détails

Mots clés : Asymmetric generalized Gaussian distribution, Parameter estimation, maximum likelihood, Moments, Entropy.

A novel correlation filter based on variational calculus

Djemel Ziou, Dayron Rizo Rodriguez, Nafaa Nacereddine, Salvatore Tabbone  (2019)

Correlation filters have been a popular technique for tackling image classification problems. The traditionalcriteria used to design correlation filters overlook some properties that can improve their discriminative power.Therefore, new criteria are proposed to design a novel correlation filter. Such criteria take advantage ofnegative samples, spatial information and the smoothness of the correlation output space. A closed formis derived from the criteria proposed using variational calculus. Moreover, it is shown that the resultingcorrelation filter is a bandpass filter. Experiments are conducted for face identification under illuminationvariation for a single training image per subject and head pose classification. The correlation filter proposeddelivers favorable scores when compared to other correlation filters and state-of-the-art approaches Voir les détails

Mots clés : Correlation filter, Variational calculus, Face identification, Illumination variation, Single training image, Pose classification

Scale space Radon transform

Djemel Ziou, Nafaa Nacereddine, Aicha Baya Goumeidane  (2021)

An extension of Radon transform by using a measure function capturing the user need isproposed. The new transform, called scale space Radon transform, is devoted to the casewhere the embedded shape in the image is not ?liform. A case study is brought on a straightline and an ellipse where the SSRT behaviour in the scale space and in the presence of noiseis deeply analyzed. In order to show the effectiveness of the proposed transform, the exper-iments have been carried out, ?rst, on linear and elliptical structures generated syntheticallysubjected to strong altering conditions such blur and noise and then on structures imagesissued from real-world applications such as road traf?c, satellite imagery and weld X-rayimaging. Comparisons in terms of detection accuracy and computational time with well-known transforms and recent work dedicated to this purpose are conducted, where theproposed transform shows an outstanding performance in detecting the above-mentionedstructures and targeting accurately their spatial locations even in low-quality images. Voir les détails

Mots clés : radon transform, line, ellipse, scale space, noise

Tool combination for the description of steel surface image and defect classification

Zoheir MENTOURI, Hakim DOGHMANE, Kaddour Gherfi, Rachid Zaghdoudi, Hocine Bourouba  (2021)
Article de conférence

In industry, the automatic recognition of surface defects of flat steel products still represents a real challenge. Indeed, in addition to constraints such as the image noise or blur, there is neither an agreed standard of these defects nor a standard method that can ensure the defect identification, whatever are their size, shape, orientation and location. Thus, the complexity of the algorithm that deals with this matter always depends on specific needs of the application. In this paper, we give details on an approach that combines Gabor wavelets (GW) and the local phase quantization technique (LPQ), to describe the steel surface images, and uses the histogram to extract their characteristics. The defect classification is carried out by means of two classifiers, namely the nearest neighbors and the support vector machine. The method assessment is based on testing different parameter values of the used tools. The approach shows a good performance in terms of recognition rates and feature vector length, which impacts the computing time. Also, the study reveals its suitability for an online steel surface defect recognition application. Voir les détails

Mots clés : Quality control, Computer vision, metal surface imaging, Filter bank application, pattern analysis and recognition

Surface Flaw Classification Based on Dual Cross Pattern

Zoheir MENTOURI, Hakim DOGHMANE, Abdelkrim Moussaoui, Djalil BOUDJEHEM  (2020)
Article de conférence

The evaluation of flat steel surface quality is mainly concerned with detecting and identifying product surface defects. Although the variety of the implemented techniques, this type of control still presents a challenge. In this paper, we assess the Dual Cross Pattern technique, as a feature descriptor, that should be quite discriminative, to ease the steel surface defect classification. The histograms extracted from the captured DCP features are concatenated to represent the global image feature vector. The procedure parameters, as the DCP circle radius, the number of the training images and their choice, are considered to show their impact on the results. The experiment conducted on the NEU published defect database shows that, compared to the other used techniques, the proposed approach reveals not only interesting recognition rates but presents advantages in time coast too. Voir les détails

Mots clés : Image description, pattern recognition, Product quality, steel surface defects, hot rolling process

Steel Strip Surface Defect Identi?cation using MultiresolutionBinarized Image Features

Zoheir MENTOURI, Abdelkrim Moussaou, Djalil BOUDJEHEM, Hakim DOGHMANE  (2020)

The shaped steel strip, in the hot rolling process,may exhibit some surface ?aws. Their origin could bethe internal discontinuities in the input product or thethermomechanical transformation of the material, duringthe shaping process. Such defects are of a random occurrenceand may lead to costly rework operations or to adowngrading of the ?nal product. So, they should bedetected and identi?ed as soon as possible, to allow atimely decision-making. For such a quality monitoring, theused vision systems are mainly based on an imagedescription and a reliable classi?cation. In this paper, weexplore pre-de?ned image ?lters and work on a procedureto extract a discriminant image feature, while realizing thebest trade-off between the improved recognition rate of thesurface defects and the computing time. The proposedmethod is a multiresolution approach, based on theBinarized Statistical Image Features method, employed todate in biometrics. The ?lters, pre-learnt from naturalimages, are applied to steel defect images as a new surfacestructure indicator. They provide a quite discriminating image description. A relevant data reduction is used togetherwith a classi?er to allow an ef?cient recognition rate ofthe defective hot rolled products. Voir les détails

Mots clés : Computer vision, statistical features, Classi?cation, strip surface defects, hot rolling process

Improved cross pattern approach for steel surface defect recognition

Zoheir MENTOURI, Hakim DOGHMANE, Abdelkrim Moussaoui, Hocine Bourouba  (2020)

In steel-making processes, different methods are used for online surface product monitoring. Such a control has become anecessity to avoid additional costs resulting from the poor quality of the final product. With the reported performance that variesfrom one application to another, all the applied methods have to meet a minimum of criteria as accuracy and speed. Thiseffectiveness is assured thanks to a relevant image description and efficient defect classification algorithms. The Dual CrossPattern technique, successfully applied in face recognition, is a concept that relies on coding pixels to provide such a discriminatingdescription of the image. Its principle can perfectly be used in industrial vision applications for surface defect recognition.In this study, the relevance of this method of describing defect images is evaluated, and improvements are proposed to increase itsefficiency. The experimental study shows that the pixel coding that considers the variations of the intensity in several directionsand captures the information from more than one pixel-neighborhood level makes it possible to better detect the variability in thedefect image and helps to increase the defect recognition rate. The experiments are carried out with the use of the publishedNortheastern University (NEU) database for the comparison and with a new constructed database to better show the improvementsbrought by the proposed approach. Voir les détails

Mots clés : Computer vision, Image description, Surface defect classification, Steel process

The importance of using dual-channel heterostructure in strained P-MOSFETs

Amine Mohammed TABERKIT, Ahlam GUEN-BOUAZZA, Mohamed HORCH  (2018)
Article de conférence

We present in this work a dual-channel heterostructure strained structure, introduce the high carrier mobility Awaited in heterostructure devices while using several models which are: CVT, SHIRAHATA, and WATT, we present a two-dimensional simulation of dual strained channel heterostructure P-MOSFETs. This study is accomplished usingSILVACO-TCAD simulation software, the comparison of the effect of using strain technique on P-MOSFET transistors will demonstrate the importance of using strain technique especially in dual channel heterostructure MOSFET. The simulation of fabrication steps and the extraction of the electronic proprieties in terms of transfer and output characteristics, transconductance, and the quasi-static capacitance allow understanding and interpreting these enhancements Voir les détails

Mots clés : Strained Silicon, SiGe layer, MOSFET; Heterostructure, simulation, Silvaco