A delegation from the University of Donja Gorica visited the Republic of China from March 22 to April 1, with the primary goal of strengthening cooperation with Chinese universities in the field of artificial intelligence. During the visit, the delegation engaged with different universities and institutions: Henan University, SIAS University, Beijing University of Technology (BJUT), Beijing Institute of Technology (BIT), Beijing Union University (BUU), and the Chinese Academy of Sciences (ANSO).
UDG visits Beijing, China
Mr Dejan Babic (PhD student) and mr Stevan Cakic (PhD candidate), both members of the HPC4S3ME project team, were responsible for discussion related to AI and applied computer science. They were actively involved in engagement of their Chinese peers to explore the oppotunities for collaboration.
UDG continues exploring oppotrunities for collaborations
Each visit included structured meetings focused on developing joint research initiatives, student and staff exchange programs, as well as opportunities for participation in summer schools and international collaborations. The Chinese side demonstrated exceptional hospitality, professionalism, and a strong interest in building long-term partnerships. In addition to the official program, the delegation had the opportunity to explore Chinese history and culture through visits to key historical landmarks and institutions.
Mr Aleksandar vesovic defended his BSc thesis on the use of AI and HPC to develop a solution for attendance records in schools or universities. His mentors were Stevan Cakic and Tomo Popovic. He defended his theses on Friday, 28 March.
The theses and presentation discussed the integration of AI models into a web application and HPC integration
ABSTRACT – This thesis addresses the challenge of tracking student attendance in lectures through facial recognition. The aim of the research is to develop and implement a system that allows for automatic and accurate attendance tracking, thereby eliminating traditional methods that are often prone to errors and manipulation. The study analyzes the latest technologies in artificial intelligence, machine learning, and high- performance computing ( HPC) to achieve optimal accuracy and system efficiency. The implementation was tested on a sample of students and demonstrated high accuracy in facial recognition and attendance recording. This work also considers ethical aspects and p r ivacy concerns, given the sensitivity of the data collected and processed. The results suggest that applying facial recognition technology in an educational setting can significantly improve administrative processes while maintaining student security and privacy. Finally, possible future applications and recommendations for further system discussed.
HPC4S3ME and EUROCC2/EUROCC4SEE team members mentored and supported this research
A new scientific publication by researchers from the University of Donja Gorica and DunavNET explores the innovative use of generative AI in digital agriculture. Titled “Evaluating the FLUX.1 Synthetic Data on YOLOv9 for AI-Powered Poultry Farming”, the study demonstrates how synthetic data, generated using FLUX.1, can effectively enhance deep learning models for chicken detection in smart farms. The paper was published in the Journal of Applied Sciences, a special issue dedicated to the application of computer vision in industry and agriculture [link].
Using generative AI to create sytnhetic data used to train computer vision models for agriculture sector
By combining real and AI-generated images and streamlining annotation with Grounding DINO and SAM2 models, the team achieved impressive detection accuracy—proving that generative AI can bridge the data gap in precision farming. This research is a part of broader effortsincluding PhD research of mr. Stevan Cakic, as well as collaboration with company that produces smart agriculture platform. This was also supported through EuroCC Montenegro initiatives, showcasing how high-performance computing and AI can drive sustainable innovation in agriculture.
High-level architecture used for experiment execution
ABSTRACT – This research explores the role of synthetic data in enhancing the accuracy of deep learning models for automated poultry farm management. A hybrid dataset was created by combining real images of chickens with 400 FLUX.1 [dev] generated synthetic images, aiming to reduce reliance on extensive manual data collection. The YOLOv9 model was trained on various dataset compositions to assess the impact of synthetic data on detection performance. Additionally, automated annotation techniques utilizing Grounding DINO and SAM2 streamlined dataset labeling, significantly reducing manual effort. Experimental results demonstrate that models trained on a balanced combination of real and synthetic images performed comparably to those trained on larger, augmented datasets, confirming the effectiveness of synthetic data in improving model generalization. The best-performing model trained on 300 real and 100 synthetic images achieved mAP = 0.829, while models trained on 100 real and 300 synthetic images reached mAP = 0.820, highlighting the potential of generative AI to bridge data scarcity gaps in precision poultry farming. This study demonstrates that synthetic data can enhance AI-driven poultry monitoring and reduce the importance of collecting real data.
On 27 December, University of Donja Gorica hosted a discussion on active project results and progress. Ms Zoja Scekic, a young researcher on HPC4S3ME project gave a presentation on the project and its final results. Most notably she shared the experences from the HPC/AI Workshop and Students Conference that was the final event withing the HPC4S3ME project.
Ms Zoja Scekic presented the project at UDG This was a good opportunity to discuss the sustainability and cross-project collaborationWe shared experiences from the final event – HPC/AI Workshop and Students Conference
The University of Donja Gorica hosted the highly anticipated HPC/AI Workshop and Student Conference on Saturday, December 21st. This event served as a key platform to highlight the outcomes of the HPC4S3ME IPA project and the AIFusion training program supported by the Innovation Fund of Montenegro. Both projects were developed within the framework of EuroCC2/EuroCC4SEE, with support from the NCC Montenegro team. The events showcased significant advancements in high-performance computing (HPC) and artificial intelligence (AI) in Montenegro.
Ms Zoja Scekic presenting HPC4S3ME project resultMr Stevan Cakic presenting AIFustion training that was supported by the Innovation Fund of Montenegro
The workshop commenced with presentations on the results and achievements of the HPC4S3ME and AIFusion projects. Participants were also introduced to the activities and contributions of NCC Montenegro and the broader EuroCC2/EuroCC4SEE initiatives. These sessions underscored the critical role of collaboration and innovation in advancing HPC and AI capabilities.
Dr Luka Filipovic discussing the EuroCC2/EuroCC4SEE projects and HPC initiatives/opportunities
A highlight of the event was the student conference, which featured 19 project presentations by student teams, primarily from MSc and BSc programs. These projects demonstrated the creativity, technical acumen, and forward-thinking approaches of the next generation of researchers and innovators. The diversity of ideas presented reflected the growing interest and expertise in leveraging HPC and AI for real-world applications. The project discussed HPC and AI applications in digital transformation of education, medicine, fashion, mathematcis, and tourism.
A variety of interesting project implemented by student teams were presentedThis was a great opportunity to exchange experiences and lessons learnedComputer vision and robotics applicationsThe presentation covered applications of HPC/AI in all domains in S3 Montenegro
The day concluded with a dynamic panel discussion focusing on the potential of HPC and AI, emphasizing the importance of training, skills development, and success stories from academia and industry. Experts and participants engaged in vibrant discussions, exchanging insights and exploring the future possibilities of these transformative technologies.
This event was a great way to conclude activities on the HPC4S3ME project
With over 50 attendees, the event fostered a collaborative and engaging atmosphere, encouraging knowledge sharing and networking. The conference provided participants with an inspiring experience, leaving them motivated to explore further advancements in HPC and AI.
Over 50 people attended the event
The event concluded on a high note with a cocktail reception, offering attendees the opportunity to network in an informal setting. This gathering reaffirmed the University of Donja Gorica’s commitment to fostering innovation and excellence in HPC and AI, further establishing its role as a regional leader in these critical domains. This was another successful cross-project collaboration.
The event finished with a coctail and networking in the club
The Book of abstracts can be accessed by clicking on the image below (link).
On Saturday, December 21st, the University of Donja Gorica will host an HPC/AI Workshop and Student Conference, where participants from the AIFusion and HPC4S3ME projects will present their results. This will be a great way of finalizing efforts in both projects!
HPC/AI Workshop and student conference are organized in context of HPC4S3ME and AI Fusion projects
The event will include:
Presentation of key results and achievements of both projects,
NCC Montenegro and EuroCC2/EuroCC4SEE presentation,
Presentation of student projects,
Panel discussion,
Coctail and networking.
Location: AP Amphitheatre, University of Donja Gorica Time: 10:00am – 16:00pm
After the program, the socializing will continue with a cocktail. Join us to celebrate the results and exchange ideas in the field of HPC and AI.
HPC4S3ME team presented a scientific paper called “Optimizing solar energy management using AI based systems” at the 8th International Symposium on Innovative Approaches in Smart Technologies that took place 6-7 December 2024 in Istanbul, Turkiye. The authors of the paper were Arnad Lekic, Anesa Abazovic, Dejan Babic, Ivan Jovovic, and Tomo Popovic. This research was a direct result of capacity building activity supported by the HPC4S3ME project.
Abstract— The research presented in this paper explores the possibilities of optimizing solar energy usage and managing solar panels in hybrid energy systems that use both solar energy and the electrical grid. A model has been created to track time series data and solar radiation in order to predict the expected solar energy yield. The goal of this research is to enable users to plan their energy consumption and maximize savings through optimal use of solar panels. This process involves collecting weather data, optimizing energy usage, and proposing a system implementation that can be used by both companies and end users.
The University of Donja Gorica (UDG) hosted a captivating lecture by Dr. Yang Jingxin, Associate Professor at the Robotics College (Artificial Intelligence College) of Beijing Union University (BUU). A distinguished expert in artificial intelligence, Dr. Yang has an impressive academic journey, including a Doctorate in Materials Science and Engineering from Tsinghua University and a National Visiting Scholarship at Rice University in the USA. Her extensive research bridges intelligent manufacturing and artificial intelligence, focusing on innovation and sustainability.
Prof Yang gave an inspiring lecture on AI research and applicationsThe presentation included both background on AI and current research activities at BUU
During her lecture, Dr. Yang provided an in-depth exploration of AI’s transformative potential, its cutting-edge applications, and the future trends shaping this dynamic field. She highlighted AI’s role at BUU and shared insights into the intersection of intelligent manufacturing and AI.
During November, two UDG researchers implemented a month long mobility at DunavNET as part of the TRACEWINDU MSCA H2020 project. Mr Ivan Jovovic and mr Dejan Babic, PhD students, engaged in productive discussions with DunavNET’s expert team members. The focus was on evaluating blockchain frameworks to identify the most suitable structure for ensuring transparent and secure wine traceability. They also analyzed data model requirements, highlighting the importance of data integrity and accessibility for tracking wine from production to consumption. During their stay, they went through a training on the utilisation of various AI tools and resource management on Azure platforms. An idea that will be pursued in their future research would be creation and use of chatbots for engagement of stakeholders. More on the TRACEWINDU project can be seen at the following link.
Two young researcher mobilities implemented through cross-pollination with TRACEWINDU project
The paper “Detection of Livestock Using Edge Devices” by E. Taruh, M. Raicevic, I. Jovovic, D. Babic, and T. Popovic, was presented at the 5th International Conference on Communications, Information, Electronic and Energy Systems, 2024 IEEE CIEES conference that takes place in Tarnovo, Bulgaria. The paper is result of research and experimenting in the agricultural domain throgh mentoring and cross-project collaboration.
Mr Elvis Taruh, MSc candidate, presenting the paper at the session on AI
ABSTRACT – This paper presents the development of a real-time system for detecting and monitoring large livestock using cameras connected to a Jetson Nano device and the YOLO v8 model for animal recognition (specifically cattle). The implementation of this system enables precise tracking of livestock movements and conditions, facilitating supervision and management on farms. The system provides real-time notifications and information crucial for efficient resource management. Model evaluation showed an accuracy of 65% after 35 epochs of training. The developed platform demonstrates potential for improving efficiency and safety in agricultural enterprises with possibilities for further enhancement and scaling.
