Page 55 - tech fest 2025 ב״ש חוברת תקצירים

Classification of Drones, UAVs and Birds Using CNN
CS-E-31
Ben Levintan; benlevintan@gmail.com Omri Buhbut; omribuhbut787@gmail.com
Advisor: Prof. Shlomo Greenberg
SCE - Shamoon College of Engineering, Be’er-Sheva
This project designed a machine learning-based system for drone and UAV detection, with the ability to distinguish them from sightings of birds or planes. A deep learning model, which was trained on diverse aerial datasets and multiple convolutional neural network (CNN) modules, managed to achieve high accuracy in real-time classification. Optimized for security applications, the system ensures reliable performance under varying conditions, with minimal false positives.
Keywords: deep learning, drone detection, RGB imagery, UAV detection
AI-Based Mobile Application for Animal Identification in Zoos
CS-E-32
Noor Biadsa; noorbi@ac.sce.ac.il Karam Abu Siam; karamab4@ac.sce.ac.il
Advisors: Dr. Moshe Davidian1, Prof. Shlomo Greenberg1 1SCE - Shamoon College of Engineering, Be’er-Sheva
Zoos offer a unique opportunity to observe wildlife up close, but visitors often have limited access to detailed information about the animals they encounter. This project developed a mobile application that enhances the zoo experience by enabling real-time animal identification. Using a convolutional neural network (CNN), the app allows users to take a photo of an animal, and instantly receive relevant information about its species, habitat, and characteristics. A significant challenge in this process was the absence of a comprehensive dataset for training the model, requiring extensive data collection and refinement. By integrating deep learning and image recognition, this project created an engaging educational tool that improves accessibility and visitor interaction in zoos.
Keywords: animal recognition, CNN, convolutional neural networks, deep learning, enhanced zoo experience
Book of Abstracts | 2025
55