Liste des communications
Radiographic Image Segmentation Based on Gaussian Mixture Model
In this work, we propose to use an image segmentation method based on Gaussian mixture model. The observed image is considered as a mixture of multivariate densities and the mixture parameters are estimated by the expectation maximization (EM) algorithm. The segmentation is completed by clustering each pixel into a component according to the maximum likelihood (ML) estimation. This method has been applied to a variety of radiographic images of weld defects and satisfactory segmentation results have been reported. Voir les détails
Mots clés : — expectation maximization algorithm, fuzzy C-means algorithm, Gaussian mixture model, image segmentation, radiographic images, weld defect
Unsupervised Algorithm for Radiographic Image Segmentation Based on the Gaussian Mixture Model
In this paper we study an unsupervised algorithm for radiographic image segmentation, based on the Gaussian mixture models (GMMs). Gaussian mixture models constitute a well-known type of probabilistic neural networks. One of their many successful applications is in image segmentation. Mixture model parameters have been trained using the expectation maximization (EM) algorithm. Numerical experiments using radiographic images illustrate the superior performance of EM method in term of segmentation accuracy compared to fuzzy c-means algorithm. Voir les détails
Mots clés : weld defect, radiographic images, image segmentation, Gaussian mixture model, expectation maximization algorithm, fuzzy C-means algorithm
Etude de la permsélctivité d’une membrane échangeuse de cations au contact des cations Cd+2 et Pb+2
La permsélectivité décrit l’aptitude d’une membrane à favoriser la migration d’un type d’ion (contre-ion) tout en empêchant des ions de signe contraire (co-ion). Elle présente en effet un certain nombre de points communs avec la théorie des jonctions liquides des piles de concentration et elle est directement liée à l’étude dynamique des échanges d’ions. Nous nous intéressons dans cette étude à la caractérisation d’une membrane échangeuse de cations de type CDS. Nous avons effectué des mesures de potentiel d’une pile de concentration séparée par la membrane CDS à courant nul ce qui permet le calcul des nombres de transport des cations (Pb+2 et Cd+2). Le potentiel de membrane mesuré égal au potentiel de membrane théorique moins le potentiel de jonction. La détermination du potentiel théorique nécessite la connaissance des valeurs des coefficients d’activités et des forces ioniques des ions existants dans la solution d’équilibre de la membrane étudiée. Ces deux derniers sont calculés à l’aide de l’expression de Debye-Hückel. Les résultats des essais réalisés au laboratoire sur des solutions composées des ions métalliques (Pb+2 et Cd+2) montrent que la membrane CDS est plus permsélective qu’aux ions Cd+2 que les ions Pb+2. Voir les détails
Mots clés : Membrane échangeuse de cations, Permsélectivité, Nombre de transport
Parametric active contour for boundary estimation of weld defects in radiographic testing
In this paper we present a new approach to deal with the defects contours estimation problem in radiographic images using parametric active contours. In this approach we exploit the performance of the GVF as external force and enhance it by joining to it external adaptive pressure forces which bring speed to the snake evolution and less sensitivity to the snake initialization and provides capability of tracking the concavities. Voir les détails
Mots clés : Active contour, GVF, Pression forces, radiographic images
Parametric active contour for weld defects boundary extraction in radiographic testing
Snakes, or active contours, are used extensively in computer vision and image processing applications, particularly to locate object boundaries. Problems associated with initialization and poor convergences to boundary concavities have aroused, which restricts their utility. This paper presents a new approach to deal with the defects contours estimation problem in radiographic images using parametric active contours. In this approach we exploit the performance of the GVF as external force and enhance it by joining to it an external adaptive pressure forces which speeds up to the snake progression, makes it less sensitive to initialization and provides capability of tracking the concavities. Voir les détails
Mots clés : image segmentation, Active contour, Radiographic inspection
Weld defect detection with parametric deformable model
Snakes, or active contours, are used extensively in computer vision and image processing applications, particularly to locate object boundaries. Problems associated with initialization and poor convergence to boundary concavities have aroused, which restricts their utility. This paper presents a new approach to deal with the defects contour estimation problem in radiographic images using parametric active contours. In this approach we exploit the performance of the Gradient Vector Flow (GVF) as external force and enhance it by adding external adaptive pressure forces (APF) which speeds up the snake progression, makes it less sensitive to initialization and provides capability of tracking the concavities. Voir les détails
Mots clés : : Radiographic testing, Active contour, GVF.
EXPLOITATION DU PHENOMENE D’ATTENUATION, DANS LE CAS D’ONDES ULTRASONORES GUIDEES, EN PRESENCE D’UN LIQUIDE
Our work consists to study the propagation of guided waves in an isotropic solid plate, put in contact with a liquid.. To study the sensitivity of the guide when it is in the presence of a liquid, we plotted the dispersion curves of phase and group velocity, of transverse and longitudinal displacement, and the attenuation according to the product frequency-thickness. The attentive study of these curves let us to locate two particulars points, P1 and P2 on the dispersion curves for the S0 and A0 modes respectively. When the guided wave is excited under the point P1, it undergoes a maximum attenuation, and when the wave is excited under the point P2, it undergoes a less attenuation. A range of experiments was carried out to justify the theoretical forecasts. The results obtained are very satisfactory and show clearly, the sensitivity of the wave for well defined products frequency-thickness. Voir les détails
Mots clés : Lamb waves; generation; attenuation; dispersion curves
A LIQUID LEVEL SENSOR USING THE A0 LAMB WAVE MODE
The object of this work consists to study the feasibility of a liquid level sensor using the A0 Lamb wave mode. This sensor is made of a vertically one meter and thirty plate of stainless steel in which one, an A0 Lamb wave mode is generated. To perfect the sensor, we have to choose the good material, the thickness of the plate and the vibration mode of the wave. To do that, we have plot the phase and group velocity curves, the transversal and longitudinal displacements versus the product frequency x thickness. The curves interpretation enables to visualize the best target point for the excitation energy on a suitable guided wave mode at a suitable frequency; this point is called the operating point. When the chosen mode is generated, the echo from the reflexion at the guide-water interface is easily detected by the transmitter-receiver transducer and can be worked efficiently. Voir les détails
Mots clés : : Lamb waves modes; Guided waves; sensor; level
Analyse par Transformée en Ondelette des signaux ultrasonores pour le contrôle non destructif d’un acier inoxydable austénitique à fort bruit de structure
Dans ce travail nous avons développé des méthodes de traitement du signal, basé sur les Transformées en Ondelette Continues et Croisées dans le but de localiser et de caractériser le bruit de structure d’un acier inoxydable austénitique avec ou sans soudure, contenant ou non des défauts. Le caractère dispersif et atténuant de ces matériaux qui résulte de la diffusion ultrasonore par les grains est à l’origine du bruit de structure qui rend le CND de ces matériaux difficile. Sur des éprouvettes d’acier inoxydable austénitique, nous avons acquis plusieurs signaux, contenant du bruit de structure, des échos de défauts et des échos de la soudure. Nous avons appliqué la transformée en ondelette continue sur ces signaux en choisissant l’ondelette de Morlet d’ordre 6, le résultat obtenu nous permet de localiser les différents échos dans le plan temps - fréquence. Afin de mieux isoler le bruit de structure nous avons appliqué la transformée en ondelette croisée à ces signaux. Ceci nous a permis dans certains cas de distinguer le bruit de structure par rapport aux autres discontinuités (défaut, soudure,..). Les résultats obtenus sont prometteurs dans la mesure ou ils ouvrent des perspectives intéressantes, en vue de caractériser et/ou de supprimer éventuellement le bruit de structure pour une meilleure efficacité du CND. Par ailleurs ce travail pourrait contribuer à développer une nouvelle approche pour la caractérisation des microstructures à partir de l'étude du bruit de structure ultrasonore. Voir les détails
Mots clés : Acier inoxydable austénitique, bruit de structure, contrôle non destructif, transformée en ondelette
Time - frequency and time - scale representations for the ultrasonic Non Destructive Testing and Evaluation of materials with high structural noise.
Time - frequency and time-scale methods are proposed to perform the analysis of Non DestructiveTesting and Evaluation (NDTE) ultrasonic signals received during the inspection of stainless steels and thestainless steel welding. The non-homogeneous nature of such materials induces a very high level of structuralnoise which greatly complicates the interpretation of the NDTE signals. In such conditions, classical temporal orspectral signal processing is not effective. By combining the time domain and frequency analysis, the wavelettransform provides simultaneously spectral representation and temporal order of the signal decompositioncomponents and can give relevant information to separate the structural noise from other echoes present in thesignal in particular those of defects. A time-scale approach providing by the continuous wavelet transform is usedand some representations given by the cross wavelet transform are considered. The time-scale analyses show thatwe could separate in the time-frequency and time-scale representations, the structural noise from other echoespresent in the signal. This work opens new fields in the ultrasonic NDTE of materials with high structural noise asaustenitic steel and austenitic steel welding. Voir les détails
Mots clés : Non Destructive Testing and Evaluation, Stainless Steels Austenitic, Structural noise, Ultrasonic, Wavelet transform, welding