Željko Eremić
Željko Eremić has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Dragan Halas
Visoka tehnička škola strukovnih studija u Zrenjaninu, Zrenjanin, Srbija
2 shared publications - Aleksandar Dedić
Univerzitet u Beogradu, Šumarski fakultet, Beograd
1 shared publication - Ana Petrović
University of Belgrade, Faculty of Mechanical Engineering, Belgrade, Serbia
1 shared publication Beograd Akademija Strukovnih Studija Politehnika
1 shared publication- Dražan Kozak
Mechanical Engineering Faculty in Slavonski Brod, University of Slavonski Brod, Trg I. B. Mažuranić 2, 35000 Slavonski Brod, Croatia
1 shared publication Fakultet Tehničkih Nauka Univerzitet U Novom Sadu
1 shared publication
Latest Publications
- 1 January 2026
- Published by Centre for Evaluation in Education and Science (CEON/CEES)
- No. Session 6_,p. 447-454
Assessment of the possibilities of artificial intelligence for agrochemical soil analysis with fertilization recommendations
Various artificial intelligence models exist today. This paper examines the capabilities these models offer in the field of agrochemical soil analysis and fertilizer recommendations. We leveraged internet technologies to evaluate the potential of several artificial intelligence models within agricultural technology. The outputs obtained were compared against official recommendations. The paper presents a comparison between soil fertilization recommendations based on agrochemical analysis results provided by professional services and those generated by various artificial intelligence models. The results of this comparison and the associated conclusions are presented and subsequently discussed in terms of the reliability of the tested models.
- 1 January 2026
- Published by Centre for Evaluation in Education and Science (CEON/CEES)
- No. Session 6_,p. 447-454
Assessment of the possibilities of artificial intelligence for agrochemical soil analysis with fertilization recommendations
Various artificial intelligence models exist today. This paper examines the capabilities these models offer in the field of agrochemical soil analysis and fertilizer recommendations. We leveraged internet technologies to evaluate the potential of several artificial intelligence models within agricultural technology. The outputs obtained were compared against official recommendations. The paper presents a comparison between soil fertilization recommendations based on agrochemical analysis results provided by professional services and those generated by various artificial intelligence models. The results of this comparison and the associated conclusions are presented and subsequently discussed in terms of the reliability of the tested models.
- Open Access
- 1 January 2026
- Published by Centre for Evaluation in Education and Science (CEON/CEES)
- No. Part four,p. 187-192
AI supported solutions for improvement of air quality and health risk evaluations
At Technical College of Applied Studies in Zrenjanin indoor air-quality was monitored using View Plus Radon Detector. Measured results were transferred via API to the Web application, which then send these data to the Artificial intelligence for evaluation. Artificial intelligence estimates health risks and proposes methods for indoor air quality improvement.
- 1 January 2026
- Published by Centre for Evaluation in Education and Science (CEON/CEES)
- No. Part four,p. 187-192
Open AccessAI supported solutions for improvement of air quality and health risk evaluations
At Technical College of Applied Studies in Zrenjanin indoor air-quality was monitored using View Plus Radon Detector. Measured results were transferred via API to the Web application, which then send these data to the Artificial intelligence for evaluation. Artificial intelligence estimates health risks and proposes methods for indoor air quality improvement.
- Open Access
- 1 January 2026
- Published by Centre for Evaluation in Education and Science (CEON/CEES)
- No. Part two,p. 85-92
Using multimodal AI to analyze video content
In our paper, we present the basic characteristics of LLM as well as their division into Unimodal and Multimodal. Multimodal AI models today can analyze input data in different formats. We use Google Gemini to analyze video content. The model "Gemini 2.0. Flash" is used. The experiment is conducted on two samples, which are the presentations of work at the conference so that the structure of the samples is comparable. The content we analyze is video, but it contains audio of spoken words as well as text in the background. Questions are asked about each of these types of information. The potential of AI in extracting information from these media is examined. During the experiment, identical questions are asked about each of the samples. After receiving the answers, their analysis and results are tabularly and descriptively presented. The answers received were discussed in order to observe certain regularities in the view of the possibilities of artificial intelligence. At the end, conclusions and directions for further development are presented.
- 1 January 2026
- Published by Centre for Evaluation in Education and Science (CEON/CEES)
- No. Part two,p. 85-92
Open AccessUsing multimodal AI to analyze video content
In our paper, we present the basic characteristics of LLM as well as their division into Unimodal and Multimodal. Multimodal AI models today can analyze input data in different formats. We use Google Gemini to analyze video content. The model "Gemini 2.0. Flash" is used. The experiment is conducted on two samples, which are the presentations of work at the conference so that the structure of the samples is comparable. The content we analyze is video, but it contains audio of spoken words as well as text in the background. Questions are asked about each of these types of information. The potential of AI in extracting information from these media is examined. During the experiment, identical questions are asked about each of the samples. After receiving the answers, their analysis and results are tabularly and descriptively presented. The answers received were discussed in order to observe certain regularities in the view of the possibilities of artificial intelligence. At the end, conclusions and directions for further development are presented.
- Open Access
- 16 July 2025
- Published by Union of Mechanical and Electrotechnical Engineers and Technicians of Serbia (SMEITS)
- Vol. 38 (1) , 219-226
Enhanced osmotic vacuum dryer for fruits and vegetable
This paper presents an innovative osmotic vacuum drying system designed to improve the efficiency and quality of fruit and vegetable dehydration. Traditional atmospheric pressure dryers using heated air streams are often inefficient due to high energy consumption and limited operational periods during harvest seasons. Additionally, inadequate temperature and humidity control in conventional dryers can lead to product defects such as surface hardening, incomplete dehydration, or microbial spoilage. The proposed osmotic vacuum dryer addresses these challenges by integrating controlled osmotic dehydration with vacuum drying, ensuring better moisture removal while preserving organoleptic properties. This approach enhances energy efficiency, reduces drying time, and minimizes product deterioration. The study highlights key design features, operational advantages, and potential applications of this technology in the food industry.
- 16 July 2025
- Published by Union of Mechanical and Electrotechnical Engineers and Technicians of Serbia (SMEITS)
- Vol. 38 (1) , 219-226
Open AccessEnhanced osmotic vacuum dryer for fruits and vegetable
This paper presents an innovative osmotic vacuum drying system designed to improve the efficiency and quality of fruit and vegetable dehydration. Traditional atmospheric pressure dryers using heated air streams are often inefficient due to high energy consumption and limited operational periods during harvest seasons. Additionally, inadequate temperature and humidity control in conventional dryers can lead to product defects such as surface hardening, incomplete dehydration, or microbial spoilage. The proposed osmotic vacuum dryer addresses these challenges by integrating controlled osmotic dehydration with vacuum drying, ensuring better moisture removal while preserving organoleptic properties. This approach enhances energy efficiency, reduces drying time, and minimizes product deterioration. The study highlights key design features, operational advantages, and potential applications of this technology in the food industry.
- Open Access
- 16 July 2025
- Published by Union of Mechanical and Electrotechnical Engineers and Technicians of Serbia (SMEITS)
- Vol. 38 (1) , 209-218
The cloud-based system for monitoring and management of the drying process based on microcontroller and web application
This paper introduces a drying system integrated with a web application for the remote monitoring and management of the drying process in a small laboratory dryer. The application enables users to set and adjust critical drying parameters and store data in an appropriate format for further analysis. An overview of the relevant literature and previous works of the author is provided. Key aspects of the drying process, along with its critical parameters, are discussed in detail, followed by an overview of the system's hardware and software components.The interface and functionalities of the web application are explained, and the results of the conducted experiments are presented. The paper concludes with a summary of the findings, demonstrating that the main goal, utilizing modern technologies such as React.js, JavaScript, CSS, and HTML to develop a user-friendly interface for remote monitoring and process management, has been successfully achieved.
- 16 July 2025
- Published by Union of Mechanical and Electrotechnical Engineers and Technicians of Serbia (SMEITS)
- Vol. 38 (1) , 209-218
Open AccessThe cloud-based system for monitoring and management of the drying process based on microcontroller and web application
This paper introduces a drying system integrated with a web application for the remote monitoring and management of the drying process in a small laboratory dryer. The application enables users to set and adjust critical drying parameters and store data in an appropriate format for further analysis. An overview of the relevant literature and previous works of the author is provided. Key aspects of the drying process, along with its critical parameters, are discussed in detail, followed by an overview of the system's hardware and software components.The interface and functionalities of the web application are explained, and the results of the conducted experiments are presented. The paper concludes with a summary of the findings, demonstrating that the main goal, utilizing modern technologies such as React.js, JavaScript, CSS, and HTML to develop a user-friendly interface for remote monitoring and process management, has been successfully achieved.