Unmanned Aircraft Systems (UASs) have been recognized as an important resource in search-and-rescue (SAR) missions and, as such, have been used by the Croatian Mountain Search and Rescue (CMRS) service for over seven years. The UAS scans and photographs the terrain. The high-resolution images are afterwards analyzed by SAR members to detect missing persons or to find some usable trace. It is a drawn out, tiresome process prone to human error. To facilitate and speed up mission image processing and increase detection accuracy, we have developed several image-processing algorithms. The latest are convolutional neural network (CNN)-based. CNNs were trained on a specially developed image database, named HERIDAL. Although these algorithms achieve excellent recall, the efficiency of the algorithm in actual SAR missions and its comparison with expert detection must be investigated. A series of mission simulations are planned and recorded for this purpose. They are processed and labelled by a developed algorithm. A web application was developed by which experts analyzed raw and processed mission images. The algorithm achieved better recall compared to an expert, but the experts achieved better accuracy when they analyzed images that were already processed and labelled.

The bag of words (BoW) model is an efficient image representation technique for image categorization and annotation tasks. Building good feature vocabularies from automatically extracted image feature vectors produces discriminative feature words, which can improve the accuracy of image categorization tasks. In this paper we use feature vocabularies based biometric characteristic for identification on student ID and classification of students’ papers and various exam documents used at the University of Mostar. We demonstrated an experiment in which we used OpenCV as an image processing tool and tool for feature extraction. As regards to classification method, we used Neural Network for Recognition of Handwritten Digits (student ID). We tested out proposed method on MNIST test database and achieved recognition rate of 94,76% accuracy. The model is tested on digits which are extracted from the handwritten student exams and the accuracy of 82% is achieved (92% correctly classified digits)

With increasing urbanization and the growing number of vehicles in cities, parking challenges are becoming more significant. Traditional parking systems often fail to provide sufficient information about available parking spaces, resulting in drivers wasting valuable time searching for a spot. This results in higher fuel consumption, increased emissions, and additional strain on traffic networks. Some large cities have implemented smart solutions that use sensor systems to display real-time parking availability. However, poor signage and unclear regulations can still make it difficult for drivers to find legal parking. This paper examines the potential for enhancing parking assistance systems by detecting and recognizing parking signs. Parking signs contain essential information about parking rules and space availability, but their diversity and complexity pose challenges for efficient analysis. The primary objective of this study was to conduct a preliminary investigation into the feasibility of developing an automated system for detecting and recognizing parking signs. To support this aim, a dedicated dataset of parking sign images was compiled and used to train a YOLO-based object detection model, enabling accurate localization and classification of parking signs within urban scenes. We then applied text recognition techniques to extract regulatory information from the detected signs and convert it into explicit, machine-readable parking rules. This approach lays the foundation for practical applications in automated parking management systems. Parking Assistance, Smart City, Sign Detection, Deep Learning, Image Recognition

The development of mobile health (mHealth) applications has revolutionized health and fitness management by providing users with innovative digital tools to monitor and improve their well-being. These apps integrate features such as step counting, health tracking, and activity monitoring, making them valuable for personalized health interventions. However, ensuring effective user engagement remains a critical challenge. Persuasive System Design (PSD) strategies are essential for influencing user behaviour, improving adherence to health interventions, and enhancing app usability. Despite extensive research on PSD strategies, a structured framework that prioritizes the most effective strategies for mHealth apps is lacking. This study aims to address this gap by utilizing the Fuzzy Analytic Hierarchy Process (Fuzzy AHP) to identify and rank the most influential PSD strategies for health and fitness mobile applications. By gathering expert opinions from product managers, this research will develop a structured Top 10 PSD Strategies Model that offers product designers a prioritized framework for implementing persuasive techniques. The findings will help improve user retention, engagement, and overall app success.