Abstract
This paper leverages image processing techniques, including edge detection and feature extraction, to identify and perform pose tracking of cylindrical-like structures within underwater scenes. Examples of these cylindrical-shaped objects are pipes typically used in offshore oil and gas, pillars of bridge structures, or mooring line cables when their sag angle is near zero, making them approximately flat and thus can be approximated as rectilinear. In addition to the pipe's contour identification, this algorithm provides relative distance and bearing to the vision sensor to enable a convergence framework between the structure and any vehicle equipped with this sensor. Furthermore, the algorithm proposed was tested in a pollsimulated scene with a digital twin of the actual vision sensor, onboard an in-house developed ROV prototype. Additionally, the effects of common underwater challenges, such as lighting variability, shadows, turbidity, and visual noise from the pool's geometric structure, were all analyzed to describe the algorithm's performance and robustness fully. Performance was evaluated for distances, bearing angles, FOV, turbidity, camera resolutions, and algorithm processing complexity. © 2025 Marine Technology Society.

Abstract
Advanced rider assistance systems (ARAS) play a crucial role in enhancing motorcycle safety through features such as collision avoidance, blind-spot detection, and adaptive cruise control, which rely heavily on sensors like radar, cameras, and LiDAR. However, their performance is often compromised under adverse weather conditions, leading to sensor interference, reduced visibility, and inconsistent reliability. This study evaluates the effectiveness and limitations of ARAS technologies in rain, fog, and snow, focusing on how sensor performance, algorithms, techniques, and dataset suitability influence system reliability. A thematic analysis was conducted, selecting studies focused on ARAS in adverse weather conditions based on specific selection criteria. The analysis shows that while ARAS offers substantial safety benefits, its accuracy declines in challenging environments. Existing datasets, algorithms, and techniques were reviewed to identify the most effective options for ARAS applications. However, more comprehensive weather-resilient datasets and adaptive multi-sensor fusion approaches are still needed. Advancing in these areas will be critical to improving the robustness of ARAS and ensuring safer riding experiences across diverse environmental conditions.

Abstract
The textile industry faces economic and environmental challenges due to low recycling rates and contamination from fasteners like buttons, rivets, and zippers. This paper proposes an Red, Green, Blue (RGB) vision system using You Only Look Once version 11 (YOLOv11) with a sliding window technique for automated fastener detection. The system addresses small object detection, occlusion, and fabric variability, incorporating Grounding DINO for garment localization and U2-Net for segmentation. Experiments show the sliding window method outperforms full-image detection for buttons and rivets (precision 0.874, recall 0.923), while zipper detection is less effective due to dataset limitations. This work advances scalable AI-driven solutions for textile recycling, supporting circular economy goals. Future work will target hidden fasteners, dataset expansion and fastener removal.

Abstract
Riding a motorcycle involves risks that can be minimized through advanced sensing and response systems to assist the rider. The use of camera-collected images to monitor road conditions can aid in the development of tools designed to enhance rider safety and prevent accidents. This paper proposes a method for developing deep learning models designed to operate efficiently on embedded systems like the Raspberry Pi, facilitating real-time decisions that consider the road condition. Our research tests and compares several state-of-the-art convolutional neural network architectures, including EfficientNet and Inception, to determine which offers the best balance between inference time and accuracy. Specifically, we measured top-1 accuracy and inference time on a Raspberry Pi, identifying EfficientNetV2 as the most suitable model due to its optimal trade-off between performance and computational demand. The model's top-1 accuracy significantly outperformed other models while maintaining competitive inference speeds, making it ideal for real-time applications in traffic-dense urban settings.

Abstract
The increasing demand for automated quality inspection in modern industry, particularly for transparent and reflective parts, has driven significant interest in vision-based technologies. These components pose unique challenges due to their optical properties, which often hinder conventional inspection techniques. This systematic review analyzes 24 peer-reviewed studies published between 2015 and 2025, aiming to assess the current state of the art in computer vision-based inspection systems tailored to such materials. The review synthesizes recent advancements in imaging setups, illumination strategies, and deep learning-based defect detection methods. It also identifies key limitations in current approaches, particularly regarding robustness under variable industrial conditions and the lack of standardized benchmarks. By highlighting technological trends and research gaps, this work offers valuable insights and directions for future research-emphasizing the need for adaptive, scalable, and industry-ready solutions to enhance the reliability and effectiveness of inspection systems for transparent and reflective parts.

Abstract
In an era where robots are becoming an integral part of human quotidian activities, understanding how they function is crucial. Among the inherent building complexities, from electronics to mechanics, path planning emerges as a universal aspect of robotics. The primary contribution of this work is to provide an overview of the current state of robot path planning topics and a comparison between those same algorithms and its inherent characteristics. The path planning concept relies on the process by which an algorithm determines a collision-free path between a start and an end point, optimizing parameters such as energy consumption and distance. The quest for the most effective path planning method has been a long-standing discussion, as the choice of method is highly dependent on the specific application. This review consolidates and elucidates the categories of path planning methods, specifically classical or analytical methods, and computer intelligence methods. In addition, the operational principles of these categories will be explored, discussing their respective advantages and disadvantages, and reinforcing these discussions with relevant studies in the field. This work will focus on the most prevalent and recognized methods within the robotics path planning problem, being mobile robotics or manipulator arms, including Cell Decomposition, A*, Probabilistic Roadmaps, Rapidly-exploring Random Trees, Genetic Algorithms, Particle Swarm Optimization, Ant Colony Optimization, Artificial Potential Fields, Fuzzy, and Neural Networks. Following the detailed explanation of these methods, a comparative analysis of their advantages and drawbacks is organized in a comprehensive table. This comparison will be based on various quality metrics, such as the type of trajectory provided (global or local), the scenario implementation type (real or simulated scenarios), testing environments (static or dynamic), hybrid implementation possibilities, real-time implementation, completeness of the method, consideration of the robot's kinodynamic constraints, use of smoothing techniques, and whether the implementation is online or offline.