When processing materials, there are a number of factors which influence the final quality of products. The possibility of defects occurrence, both sub-surface and surface ones, is one of the factors. In the precision engineering production, the automotive, bearing or aerospace industry in particular, no level of defects is tolerable. The thesis deals with the causes and consequences of the occurrence of such defects, and the methods for their identification. For this purpose, a new contactless optical method for the fully automatic inspection of the surface defects has been developed by Mesing, spol. s r.o. In order to verify the performance of the aforementioned method, four specific products have been tested by the author using the method. Finally, there are examples of the successful implementation of the automatic machines, for a design of which the results of the model testing were utilized.

Gas Metal Arc Welding (GMAW) is one of the main methods used to join structural members. One of the largest challenges involved in production of welds is ensuring the quality of the weld. One of the main factors attributing to weld quality is penetration depth. Automatic control of the welding process requires non-contact, non-destructive sensors that can operate in the presence of high temperatures and electrical noise found in the welding environment. Inspection using laser generation and electromagnetic acoustic transducer (EMAT) reception of ultrasound was found to satisfy these conditions. Using this technique, the time of flight of the ultrasonic wave is measured and used to calculate penetration depth. Previous works have shown that penetration depth measurement performance is drastically reduced when performed during welding. This work seeks to realize in-process penetration depth measurement by compensating for errors caused by elevated temperature. Neuro-fuzzy models are developed that predict penetration depth based on in-process time of flight measurements and the welding process input. Two scenarios are considered in which destructive penetration depth measurements are or are not available for model training. Results show the two scenarios are successful. When destructive measurements are unavailable, model error is comparable to that of offline ultrasonic measurements. When destructive measurements are available, measurement error is reduced by 50% compared to offline ultrasonic measurements. The two models can be effectively applied to permit in-process penetration depth measurements for the purpose of real-time monitoring and control. This will reduce material, production time, and labor costs and increase the quality of welded parts.

This work deals with analysis one of the production lines at ModusLink Czech Republic s.r.o. It seeks to find possibilities for automation of different processes on the line, currently performed manually. The basic focus is to find the appropriate technical solutions to the automated control of the designated production line. The work seeks to describe all meaningful solutions and compare their advantages and disadvantages. From this comparison of the final proposal is based on comprehensive solutions with the use of machine vision. Another point is the selection of a suitable contractor for the implementation of the proposal, making its advantage in economic terms and the resulting recommendations, if implemented the proposal.

Recent developments in the field of Semi-Supervised Learning are working to avoid the bottleneck of data labeling. This can be achieved by leveraging unlabeled data to limit the amount of labeled data needed for training deep learning models. Semi-supervised learning algorithms are showing promising results; however, research has been focusing on algorithm development, without proceeding to test their effectiveness in real-world applications. This research project has adapted and tested some semi-supervised learning algorithms on a dataset extracted from the manufacturing en-vironment, in the context of the surface analysis of stainless steel, in collaboration with Outokumpu Stainless Oy. In particular, a simple algorithm combining Pseudo-Labeling and Consistency Regularization has been developed, inspired by the state-of-the-art algorithm Fix match. The results show some potential, because the usage of Semi-Supervised Learning techniques has significantly reduced overfitting on the training set, while maintaining a good accuracy on the test set. However, some doubts are raised regarding the application of these techniques in a real environment, due to the imperfect nature of real datasets and the high algorithm development cost due to the increased complexity introduced with these methods. / Den senaste utvecklingen inom området Semi-Supervised Learning arbetarför att undvika flaskhalsen med datamärkning. Detta kan uppnås genom att utnyttja omärkta data för att begränsa mängden märkt data som behövs för att träna modeller för djupinlärning. Semi-övervakade inlärningsalgoritmer visarlovande resultat; forskning har dock fokuserat på algoritmutveckling, utan att testa deras effektivitet i verkliga tillämpningar. Detta forskningsprojekt har anpassat och testat några semi-övervakade in-lärningsalgoritmer på en datauppsättning extraherad från tillverkningsmiljön, i samband med ytanalys av rostfritt stål, i samarbete med Outokumpu Stainless Oy. I synnerhet har en enkel algoritm som kombinerar Pseudo-Labeling och Consistency Regularization utvecklats, inspirerad av den toppmoderna algoritmen Fixmatch .Resultaten visar en viss potential, eftersom användningen av Semi-Supervised Learning-tekniker avsevärt har minskat överanpassningen av träningssetet, samtidigt som en god noggrannhet på testsetet bibehålls. Vissa tvivel reses dock angående tillämpningen av dessa tekniker i en verklig miljö, på grund av den ofullkomliga karaktären hos riktiga datauppsättningar och den höga algoritmutvecklingskostnaden på grund av den ökade komplexiteten som introduceras med dessa metoder.

This thesis presents a case of study of the development and performance analysis of a surface grading application with real-time compliance. We address the issue of spatial and temporal uniformity in the acquisition system. In a surface grading application it is crucial to ensure the uniform response of the system through time and space. All the results presented for surface grading were obtained using real data from the ceramic tile industry. The VxC TSG database is public and can be accessed at www.disca.upv.es/vision/vxctsg. We present a method based on soft colour-texture descriptors computed in perceptually uniform colour spaces. The method is parameterized and the involved factors are studied using two statistical procedures; experimental design and logistic regression. Although it is not a new theoretical contribution, we have found and demonstrate that a simple set of global colour and texture statistics, together with well-known classifiers, are powerful enough to meet stringent factory requirements for real-time and performance. Also the method is compared with two other approaches from the scientific literature; Colour Histograms and Centile-LBP. Finally, we explore the method's capacity for on-line inspection in a study of real-time compliance and parallelization. / López García, F. (2005). Real-time surface grading of ceramic tiles [Tesis doctoral]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/1909