The Annals of “Dunarea de Jos” University of Galati. Fascicle IX, Metallurgy and Materials Science

Computer Vision-Based Detection of Patterns in Textile Microscopy Images

Mihaela MARIN — 2025-06-15

The analysis and detection of patterns in textile materials are a critical challenge in modern manufacturing, quality control, and automation processes. Traditional inspection methods, often reliant on manual observation, are not only time-consuming but also prone to inconsistencies. The need for a more precise, efficient, and scalable approach has driven interest in leveraging computer vision for this task. Computer vision systems, powered by advanced algorithms and machine learning models, offer the ability to process and interpret visual data from textile images at high speeds and with greater accuracy.

Static Analysis of a Converter Vessel Using the Finite Element Method

Ionel PETREA — 2025-06-15

The paper presents static and fatigue strength calculations for a converter. Following common practice with liquid steel vessels, the calculations were evaluated in accordance with the 2007 edition of the ASME Boiler and Pressure Vessel Code, Section VIII.

Assessment Using Numerical Analysis of the Behaviour of a Two Fingers Gripping Mechanism

Ionuț LAMBRESCU — 2025-06-15

An important issue in robotics is evaluating the force necessary to firmly grip objects. This can be done experimentally, or, as a more cost-effective approach, through numerical analysis. The results obtained through this method, although less precise, can offer a good assessment of the way a gripping mechanism behaves. The authors propose a comparative numerical analysis of a two-finger gripper, using a motion analysis performed in Siemens NX and a transient structural analysis conducted in Ansys. Both methods aim to capture the moment when the grasped object starts to slip. This moment marks the point at which the gripping is no longer reliable and provides insights into the reliability of the gripping mechanism. To assess the influence of the geometry of the gripping area, as well as the impact of the shape of the grasped object or the pair of materials (object and gripper), several analyses were performed in both Siemens NX and Ansys Workbench.

On the Optimization of Machined VSD Position for Repaired Pipes

Ionuţ LAMBRESCU — 2025-06-15

To be kept in use, pipelines with Volumetric Surface Defects (VSDs) must be evaluated and, if necessary, repaired. The first step in the repair process is machining the defect area. In all cases, the machined defect has a rectangular shape, with edges parallel/perpendicular to the pipe axis. We propose a novel approach: the VSD is scanned, the image is then processed using segmentation techniques, the VSD contour is detected, and finally, a family of rectangles with minimal area enclosing the VSD is defined. After that, a Finite Element Analysis is performed for each machined VSD, and an optimal case is selected.

From Mechanical Testing to Decision Making: Selecting a Structural Adhesive Using the Weighted Sum Method

Andreea ȚÎNTATU — 2025-06-15

The selection of structural adhesives in engineering applications demands a robust, transparent, and replicable methodology, especially when performance and reliability are critical. This study introduces a practical, scalable approach based on the Weighted Sum Method (WSM), a form of Multi-Criteria Decision Analysis (MCDA), to assist in the objective selection of adhesives. The novelty of this research lies in demonstrating that WSM can serve as an industrially applicable decision-making tool that translates mechanical testing data into clear, replicable selection criteria. Using experimental data from modified Arcan tests, the method provides a structured framework that can be extended to other types of tests and applications. This facilitates transparent, criterion-driven adhesive selection aligned with real-world constraints, offering engineers a powerful tool that supports consistent, informed, and reproducible decisions.

Studies and Research on Glass Thermoforming

Beatrice Daniela TUDOR — 2025-06-15

The paper presents a study on the properties that can influence the glass thermoforming process, and an experimental research on glass plate samples, with the aim of observing how glass deforms through thermoforming, using concave and convex support shapes and exposure to temperature, in different areas inside the oven.

Development of a Real-Time Eye Blink Monitoring System Using an IR Sensor and Arduino Microcontroller

Mihaela MARIN — 2025-06-15

This paper presents the design and implementation of an intelligent eye blink monitoring system aimed at detecting drowsiness in drivers with disabilities. The system integrates an infrared sensor connected to an Arduino Pro Mini microcontroller to detect blink frequency. When abnormal blink patterns associated with fatigue are detected, an audible alert is triggered, enhancing road safety.