Synthetic Data Generation for Surface Defect Detection
Communication avec acte
Auteur
Résumé
Ensuring continued quality is challenging, especially when customer satisfaction is the provided service. It seems to become easier with new technologies like Artificial Intelligence. However, field data are necessary to design an intelligent assistant but are not always available. Synthetic data are used mainly to replace real data. Made with a Generative Adversarial Network or a rendering engine, they aim to be as efficient as real ones in training a Neural Network. When synthetic data generation meets the challenge of object detection, its capacity to deal with the defect detection challenge is unknown. Here we demonstrate how to generate these synthetic data to detect defects. Through iterations, we apply different methods from literature to generate synthetic data for object detection, from how to extract a defect from the few data we have to how to organize the scene before data synthesis. Our study suggests that defect detection may be performed by training an object detector neural network with synthetic data and gives a protocol to do so even if at this point, no field experiments have been conducted to verify our detector performances under real conditions. This experiment is the starting point for developing a mobile and automatic defect detector that might be adapted to ensure new product quality.
Fichier(s) constituant cette publication
Cette publication figure dans le(s) laboratoire(s) suivant(s)
Documents liés
Visualiser des documents liés par titre, auteur, créateur et sujet.
-
Communication avec acteThis paper adresses the way machine learning techniques based on neural networks can be used to predict the impact of simplification processes on CAD model for heat transfer FEA purposes.
-
Communication avec acteEssential when adapting CAD model for finite element analysis, the defeaturing ensures the feasibility of the simulation and reduces the computation time. Processes for CAD model preparation and defeaturing tools exist but ...
-
Article dans une revue avec comité de lecturePERNOT, Jean-Philippe; DANGLADE, Florence; VERON, Philippe (CAD Solutions LLC (imprimé) and Taylor & Francis Online (en ligne), 2013)Numerical simulations play more and more important role in product development cycles and are increasingly complex, realistic and varied. CAD models must be adapted to each simulation case to ensure the quality and reliability ...
-
Communication avec acteBeing able to estimate a priori the impact of DMU preparation scenarios for a dedicated activity would help identifying the best scenario from the beginning. Machine learning techniques are a means to a priori evaluate a ...
-
Article dans une revue avec comité de lectureControlling the well-known triptych costs, quality and time during the different phases of the Product Development Process (PDP) is an everlasting challenge for the industry. Among the numerous issues that are to be ...