Fractographic classification in metallic materials by using 3D processing and computer vision techniques
Failure analysis aims at collecting information about how and why a failure is produced. The first step in this process is a visual inspection on the flaw surface that will reveal the features, marks, and texture, which characterize each type of fracture. This is generally carried out by personnel w...
| Main Authors: | , , |
|---|---|
| Format: | Online |
| Language: | eng |
| Published: |
Universidad Pedagógica y Tecnológica de Colombia
2016
|
| Subjects: | |
| Online Access: | https://revistas.uptc.edu.co/index.php/ingenieria/article/view/5301 |
| _version_ | 1801706075394670592 |
|---|---|
| author | Bastidas-Rodríguez, Maria Ximena Prieto-Ortíz, Flavio A. Espejo-Mora, Édgar |
| author_facet | Bastidas-Rodríguez, Maria Ximena Prieto-Ortíz, Flavio A. Espejo-Mora, Édgar |
| author_sort | Bastidas-Rodríguez, Maria Ximena |
| collection | OJS |
| description | Failure analysis aims at collecting information about how and why a failure is produced. The first step in this process is a visual inspection on the flaw surface that will reveal the features, marks, and texture, which characterize each type of fracture. This is generally carried out by personnel with no experience that usually lack the knowledge to do it. This paper proposes a classification method for three kinds of fractures in crystalline materials: brittle, fatigue, and ductile. The method uses 3D vision, and it is expected to support failure analysis. The features used in this work were: i) Haralick’s features and ii) the fractal dimension. These features were applied to 3D images obtained from a confocal laser scanning microscopy Zeiss LSM 700. For the classification, we evaluated two classifiers: Artificial Neural Networks and Support Vector Machine. The performance evaluation was made by extracting four marginal relations from the confusion matrix: accuracy, sensitivity, specificity, and precision, plus three evaluation methods: Receiver Operating Characteristic space, the Individual Classification Success Index, and the Jaccard’s coefficient. Despite the classification percentage obtained by an expert is better than the one obtained with the algorithm, the algorithm achieves a classification percentage near or exceeding the 60 % accuracy for the analyzed failure modes. The results presented here provide a good approach to address future research on texture analysis using 3D data. |
| format | Online |
| id | oai:oai.revistas.uptc.edu.co:article-5301 |
| institution | Revista Facultad de Ingeniería |
| language | eng |
| publishDate | 2016 |
| publisher | Universidad Pedagógica y Tecnológica de Colombia |
| record_format | ojs |
| spelling | oai:oai.revistas.uptc.edu.co:article-53012022-06-15T16:21:42Z Fractographic classification in metallic materials by using 3D processing and computer vision techniques Clasificación fractográfica de materiales metálicos usando técnicas 3D de procesamiento y visualización en computador Bastidas-Rodríguez, Maria Ximena Prieto-Ortíz, Flavio A. Espejo-Mora, Édgar Artificial Neural Network brittle fracture ductile fracture fracture due to fatigue Support Vector Machine 3D data datos 3D fractura dúctil fractura frágil fractura por fatiga Máquinas de Vectores de Soporte Red Neuronal Artificial Failure analysis aims at collecting information about how and why a failure is produced. The first step in this process is a visual inspection on the flaw surface that will reveal the features, marks, and texture, which characterize each type of fracture. This is generally carried out by personnel with no experience that usually lack the knowledge to do it. This paper proposes a classification method for three kinds of fractures in crystalline materials: brittle, fatigue, and ductile. The method uses 3D vision, and it is expected to support failure analysis. The features used in this work were: i) Haralick’s features and ii) the fractal dimension. These features were applied to 3D images obtained from a confocal laser scanning microscopy Zeiss LSM 700. For the classification, we evaluated two classifiers: Artificial Neural Networks and Support Vector Machine. The performance evaluation was made by extracting four marginal relations from the confusion matrix: accuracy, sensitivity, specificity, and precision, plus three evaluation methods: Receiver Operating Characteristic space, the Individual Classification Success Index, and the Jaccard’s coefficient. Despite the classification percentage obtained by an expert is better than the one obtained with the algorithm, the algorithm achieves a classification percentage near or exceeding the 60 % accuracy for the analyzed failure modes. The results presented here provide a good approach to address future research on texture analysis using 3D data. El análisis de falla tiene como objetivo recolectar información sobre cómo y porqué una falla es generada. El primer paso en este proceso consiste en una inspección visual en la superficie de la falla que revelará las características, marcas y textura que distinguen cada tipo de fractura. Esta inspección es generalmente llevada a cabo por personal que que usualmente no cuenta con el suficiente conocimiento o experiencia necesaria. Este artículo propone un método de clasificación para tres modos de fracturas en materiales cristalinos: súbita frágil, progresiva por fatiga y súbita dúctil. El método propuesto usa visión en 3D, y busca ser un apoyo en el análisis de falla. Las características usadas en este estudio fueron i) las características de Haralick y ii) la dimensión fractal. La adquisición de imágenes 3D se realizó con un microscopio confocal de escaneo laser Zeiss LSM 700. Para llevar a cabo la clasificación, dos clasificadores fueron evaluados: Redes de Neuronas Artificiales y Máquinas de Vectores de Soporte. La evaluación de desempeño se logró extrayendo cuatro relaciones marginales de la matriz de confusión: exactitud, sensibilidad, especificidad y precisión, y los siguientes tres métodos de evaluación: Característica Operativa del Receptor o espacio ROC, el iíndice individual de éxito en la clasificación ICSI y el coeficiente de Jaccard. A pesar que el porcentaje de clasificación obtenida por un experto es mejor que la obtenida por el algoritmo, este último logra obtener porcentajes de clasificación cerca o superior al 60% en exactitud para los tres modos de falla analizados. Los resultados que aquí se presentan representan un buen acercamiento para estructurar investigaciones futuras en análisis de textura usando datos 3D. Universidad Pedagógica y Tecnológica de Colombia 2016-09-01 info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion investigation investigación application/pdf text/html https://revistas.uptc.edu.co/index.php/ingenieria/article/view/5301 10.19053/01211129.v25.n43.2016.5301 Revista Facultad de Ingeniería; Vol. 25 No. 43 (2016); 83-96 Revista Facultad de Ingeniería; Vol. 25 Núm. 43 (2016); 83-96 2357-5328 0121-1129 eng https://revistas.uptc.edu.co/index.php/ingenieria/article/view/5301/4429 https://revistas.uptc.edu.co/index.php/ingenieria/article/view/5301/5064 |
| spellingShingle | Artificial Neural Network brittle fracture ductile fracture fracture due to fatigue Support Vector Machine 3D data datos 3D fractura dúctil fractura frágil fractura por fatiga Máquinas de Vectores de Soporte Red Neuronal Artificial Bastidas-Rodríguez, Maria Ximena Prieto-Ortíz, Flavio A. Espejo-Mora, Édgar Fractographic classification in metallic materials by using 3D processing and computer vision techniques |
| title | Fractographic classification in metallic materials by using 3D processing and computer vision techniques |
| title_alt | Clasificación fractográfica de materiales metálicos usando técnicas 3D de procesamiento y visualización en computador |
| title_full | Fractographic classification in metallic materials by using 3D processing and computer vision techniques |
| title_fullStr | Fractographic classification in metallic materials by using 3D processing and computer vision techniques |
| title_full_unstemmed | Fractographic classification in metallic materials by using 3D processing and computer vision techniques |
| title_short | Fractographic classification in metallic materials by using 3D processing and computer vision techniques |
| title_sort | fractographic classification in metallic materials by using 3d processing and computer vision techniques |
| topic | Artificial Neural Network brittle fracture ductile fracture fracture due to fatigue Support Vector Machine 3D data datos 3D fractura dúctil fractura frágil fractura por fatiga Máquinas de Vectores de Soporte Red Neuronal Artificial |
| topic_facet | Artificial Neural Network brittle fracture ductile fracture fracture due to fatigue Support Vector Machine 3D data datos 3D fractura dúctil fractura frágil fractura por fatiga Máquinas de Vectores de Soporte Red Neuronal Artificial |
| url | https://revistas.uptc.edu.co/index.php/ingenieria/article/view/5301 |
| work_keys_str_mv | AT bastidasrodriguezmariaximena fractographicclassificationinmetallicmaterialsbyusing3dprocessingandcomputervisiontechniques AT prietoortizflavioa fractographicclassificationinmetallicmaterialsbyusing3dprocessingandcomputervisiontechniques AT espejomoraedgar fractographicclassificationinmetallicmaterialsbyusing3dprocessingandcomputervisiontechniques AT bastidasrodriguezmariaximena clasificacionfractograficadematerialesmetalicosusandotecnicas3ddeprocesamientoyvisualizacionencomputador AT prietoortizflavioa clasificacionfractograficadematerialesmetalicosusandotecnicas3ddeprocesamientoyvisualizacionencomputador AT espejomoraedgar clasificacionfractograficadematerialesmetalicosusandotecnicas3ddeprocesamientoyvisualizacionencomputador |
