The paper presents a vision based approach and neural network techniques in surface defects inspection and categorization. Depending on part design and processing techniques, castings may develop surface discontinuities such as cracks and pores that greatly influence the material’s properties Since the human visual inspection for the surface is slow and expensive, a computer vision system is an alternative solution for the online inspection. The authors present the developed vision system uses an advanced image processing algorithm based on modified Laplacian of Gaussian edge detection method and advanced lighting system. The defect inspection algorithm consists of several parameters that allow the user to specify the sensitivity level at which he can accept the defects in the casting. In addition to the developed image processing algorithm and vision system apparatus, an advanced learning process has been developed, based on neural network techniques. Finally, as an example three groups of defects were investigated demonstrates automatic selection and categorization of the measured defects, such as blowholes, shrinkage porosity and shrinkage cavity.
The paper discusses possible applications of the percolation theory in analysis of the microstructure images of polycrystalline materials. Until now, practical use of this theory in metallographic studies has been an almost unprecedented practice. Observation of structures so intricate with the help of this tool is far from the current field of its application. Due to the complexity of the problem itself, modern computer programmes related with the image processing and analysis have been used. To enable practical implementation of the task previously established, an original software has been created. Based on cluster analysis, it is used for the determination of percolation phenomena in the examined materials. For comparative testing, two two-phase materials composed of phases of the same type (ADI matrix and duplex stainless steel) were chosen. Both materials have an austenitic - ferritic structure. The result of metallographic image analysis using a proprietary PERKOLACJA.EXE computer programme was the determination of the content of individual phases within the examined area and of the number of clusters formed by these phases. The outcome of the study is statistical information, which explains and helps in better understanding of the planar images and real spatial arrangement of the examined material structure. The results obtained are expected to assist future determination of the effect that the internal structure of two-phase materials may have on a relationship between the spatial structure and mechanical properties.
The paper presents an experimental evaluation of deformation of flat samples during uniaxial tensile testing, including uniform deformation and post-necking phases. The authors recommend a specially designed vision extensometer and simplified image processing method for analytical correction of triaxial test results for extended stress–strain curve estimation. A modified correction model is proposed, based on the application of Gaussian functions, to determine the neck geometry of the tested sample. The vision extensometer can monitor a specimen’s elongation using two fibre-optic gauges inserted into the material. Measurements taken from the vision extensometer are compared with readings from analogue gauges within the range of uniform deformation. The analytical correction model’s ability to correctly assess the extended true stress–strain curve in the post-necking phase was investigated. Image processing forms the basis of an efficient method for identifying the contour of the specimen’s neck. Digital image correlation (DIC) was used to verify the proposed solutions and assess the results obtained for the uniform and post-neck deformation phases. The change in thickness of the sample was experimentally measured throughout the tensile test with a digital gauge sensor and compared with the results of the digital image correlation.
Purpose: In the development of ideas for Industry 4.0, information about the element production cycle has become more and more important. Knowledge of the subsequent forming processes, determination of the machine on which the process has been carried out and of the type and wear of the tool, leads to smart production management, which plays an increasingly important role in the metal forming industry. To meet the current expectations for these challenges, an advanced technology needs to be introduced for monitoring the manufacturing processes by deploying flexible solutions. This technology must include, but not be limited to, identifying and tracking the product using laser marking. Design/methodology/approach: Laser marking allows a permanent mark in the form of a barcode to be applied to the sheet metal surface. Commonly used marking methods and the condition of the sheet metal surface can affect the marking contrast. This paper presents a concept for recording individual stages of sheet metal forming and determination of the impact of the laser marking technology on the contrast of the applied barcode. To ensure accurate control of the deformation stages, the bulging process of the spherical dome has been used as an example. Findings: Analysis of the influence of laser marking method on the barcode recognition accuracy can contribute to the development of smart management of the production process according to the idea of Industry 4.0. Research limitations/implications: A large plastic deformation has been applied to the sheet metal surface and no limitation in a barcode reading process (using vision scanning technology) was indicated. Also, the geometry deformation (different angle view of the CCD camera) of the barcode image has introduced no additional problems with a barcode reading. Originality/value: The optimal parameters of a laser marking technique for barcode marking, which are critical for the material that is subjected to metal forming operations that deform it, have been studied. The results shows that traceability is an attractive solution for tracking technological data in the production chain for a single-shaped product. Keywords: manufacturing control, laser marking, vision analysis, sheet metal forming, barcode quality. Category of the paper: Technical paper.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.