Zoran Milivojević
Zoran Milivojević has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
- Ratko Ivković
Department of Technical Science, State University of Novi Pazar, Novi Pazar, Serbia
4 shared publications - Dijana Kostić
Šargan – inženjering d.o.o
3 shared publications - Bojan Prlinčević
Kosovo and Metohija Academy of Appl. Studies , Leposavić , Serbia
2 shared publications - Bojan Prlinčević
Kosovo and Metohija Academy of Applied Studies, Leposavic, Serbia
2 shared publications - Dijana Kostić
Academy of Applied Technical and Preschool Studies , Niš , Serbia
2 shared publications - Ivana Milošević
Department of Audio and Video Technologies, School of Electrical and Computer Engineering, Academy of Technical and Art Applied Studies, Vojvode Stepe 283, 11000 Belgrade, Serbia
2 shared publications
Latest Publications
- 16 November 2025
- Published by Springer Nature
- p. 131-140
Comparative Analysis of the Regression Error in STI Estimation for a Lecture Hall
- 16 November 2025
- Published by Springer Nature
- p. 131-140
Comparative Analysis of the Regression Error in STI Estimation for a Lecture Hall
- 12 September 2025
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 84 (41) , 49883-49903
Image quality assessment: the integral CTS index for holistic similarity evaluation of images (ISII)
- 12 September 2025
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 84 (41) , 49883-49903
Image quality assessment: the integral CTS index for holistic similarity evaluation of images (ISII)
- Open Access
- 20 January 2025
- Published by University of Nis in Facta Universitatis, Series: Working and Living Environmental Protection
- p. 293-303
THE POLAR CHARACTERISTIC OF AN ACOUSTIC PARABOLIC REFLECTOR
The first part of the paper presents the geometry of the acoustic parabolic reflector. After that, analytical formulas for calculating the amplification and polar characteristics of the reflector are shown. The second part of the paper describes the experiment in which measurements were made and the acoustic characteristics of the parabolic reflector located in the yard of the Academy of Applied Technical and Preschool Studies, Department of Niš, in Niš were calculated. First, acoustic impulse responses for acoustic excitation at angles from -90° to 90° were measured. After that, the polar characteristic of the radiation was calculated. Finally, the beamwidth of the radiation, beam solid angle of the radiation, and directivity of the acoustic parabolic reflector were determined. The results of the experiment are presented numerically and graphically.
- 20 January 2025
- Published by University of Nis in Facta Universitatis, Series: Working and Living Environmental Protection
- p. 293-303
Open AccessTHE POLAR CHARACTERISTIC OF AN ACOUSTIC PARABOLIC REFLECTOR
The first part of the paper presents the geometry of the acoustic parabolic reflector. After that, analytical formulas for calculating the amplification and polar characteristics of the reflector are shown. The second part of the paper describes the experiment in which measurements were made and the acoustic characteristics of the parabolic reflector located in the yard of the Academy of Applied Technical and Preschool Studies, Department of Niš, in Niš were calculated. First, acoustic impulse responses for acoustic excitation at angles from -90° to 90° were measured. After that, the polar characteristic of the radiation was calculated. Finally, the beamwidth of the radiation, beam solid angle of the radiation, and directivity of the acoustic parabolic reflector were determined. The results of the experiment are presented numerically and graphically.
- 8 January 2025
- Published by Springer Nature in Acta Universitatis Sapientiae, Informatica
- Vol. 16 (1) , 198-218
Optimization of the Seventh-order Polynomial Interpolation 1P Kernel in the Time Domain
This paper presents the optimization of the convolutional, seventh-order polynomial, one-parameter, interpolation kernel. In the first part of the paper, the seventh-order kernel is defined, and, after that, the process of the kernel optimization is described. The optimization criterion was the minimization of the interpolation error e. The optimization involved the selection of the optimal value of the kernel parameter 𝛼, and it was carried out in the time domain. In the second part of this paper, the experiment, which was realized with the aim of determining the precision of interpolation of the third-order, fifth-order, and the seventh-order interpolation kernels, is described. After that a comparative analysis of the interpolation precision is described. As a measure of the interpolation precision, the mean square error (MSE) was used. The results of the experiment are presented graphically and tabularly. Finally, using a comparative analysis, the precision of interpolation with the kernel, whose parameters were optimized in the time domain, in relation to the kernel, whose parameters were optimized in the spectral do-main, was analyzed. Based on the comparative analysis, a recommendation for the optimal parameter for the seventh-order kernel is given.
- 8 January 2025
- Published by Springer Nature in Acta Universitatis Sapientiae, Informatica
- Vol. 16 (1) , 198-218
Optimization of the Seventh-order Polynomial Interpolation 1P Kernel in the Time Domain
This paper presents the optimization of the convolutional, seventh-order polynomial, one-parameter, interpolation kernel. In the first part of the paper, the seventh-order kernel is defined, and, after that, the process of the kernel optimization is described. The optimization criterion was the minimization of the interpolation error e. The optimization involved the selection of the optimal value of the kernel parameter 𝛼, and it was carried out in the time domain. In the second part of this paper, the experiment, which was realized with the aim of determining the precision of interpolation of the third-order, fifth-order, and the seventh-order interpolation kernels, is described. After that a comparative analysis of the interpolation precision is described. As a measure of the interpolation precision, the mean square error (MSE) was used. The results of the experiment are presented graphically and tabularly. Finally, using a comparative analysis, the precision of interpolation with the kernel, whose parameters were optimized in the time domain, in relation to the kernel, whose parameters were optimized in the spectral do-main, was analyzed. Based on the comparative analysis, a recommendation for the optimal parameter for the seventh-order kernel is given.
- Open Access
Regeneration Filter: Enhancing Mosaic Algorithm for Near Salt & Pepper Noise Reduction
This paper presents a Regeneration filter for reducing near Salt-and-Pepper (nS&P) noise in images, designed for selective noise removal while simultaneously preserving structural details. Unlike conventional methods, the proposed filter eliminates the need for median or other filters, focusing exclusively on restoring noise-affected pixels through localized contextual analysis in the immediate surroundings. Our approach employs an iterative processing method, where additional iterations do not degrade the image quality achieved after the first filtration, even with high noise densities up to 97% spatial distribution. To ensure the results are measurable and comparable with other methods, the filter’s performance was evaluated using standard image quality assessment metrics. Experimental evaluations across various image databases confirm that our filter consistently provides high-quality results. The code is implemented in the R programming language, and both data and code used for the experiments are available in a public repository, allowing for replication and verification of the findings.
Open AccessRegeneration Filter: Enhancing Mosaic Algorithm for Near Salt & Pepper Noise Reduction
This paper presents a Regeneration filter for reducing near Salt-and-Pepper (nS&P) noise in images, designed for selective noise removal while simultaneously preserving structural details. Unlike conventional methods, the proposed filter eliminates the need for median or other filters, focusing exclusively on restoring noise-affected pixels through localized contextual analysis in the immediate surroundings. Our approach employs an iterative processing method, where additional iterations do not degrade the image quality achieved after the first filtration, even with high noise densities up to 97% spatial distribution. To ensure the results are measurable and comparable with other methods, the filter’s performance was evaluated using standard image quality assessment metrics. Experimental evaluations across various image databases confirm that our filter consistently provides high-quality results. The code is implemented in the R programming language, and both data and code used for the experiments are available in a public repository, allowing for replication and verification of the findings.
- 1 January 2025
- Published by Toplica Academy of Applied Studies
- p. 543-549
Comparative analysis of LDA, QDA, SVM, algorithms and neural networks for boredom and fear detection in speech
This paper presents an analysis of speech emotion recognition focusing on distinguishing between boredom and fear emotional states. The methodology employs the fundamental frequency trajectory (F₀) and its derivatives to extract characteristics. Emotional states are classified using four machine learning algorithms: Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Support Vector Machines (SVM), and Artificial Neural Networks (ANN). The experiment utilizes the Berlin Emotional Speech Database for training and evaluation. Results demonstrate that all classifiers effectively separate boredom and fear emotions, with ANN achieving the highest classification accuracy. Performance analysis are derived using confusion matrix analysis confirm. This study provides an efficient framework for detecting subtle emotional differences in speech signals, particularly between boredom and fear manifestations.
- 1 January 2025
- Published by Toplica Academy of Applied Studies
- p. 543-549
Comparative analysis of LDA, QDA, SVM, algorithms and neural networks for boredom and fear detection in speech
This paper presents an analysis of speech emotion recognition focusing on distinguishing between boredom and fear emotional states. The methodology employs the fundamental frequency trajectory (F₀) and its derivatives to extract characteristics. Emotional states are classified using four machine learning algorithms: Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA), Support Vector Machines (SVM), and Artificial Neural Networks (ANN). The experiment utilizes the Berlin Emotional Speech Database for training and evaluation. Results demonstrate that all classifiers effectively separate boredom and fear emotions, with ANN achieving the highest classification accuracy. Performance analysis are derived using confusion matrix analysis confirm. This study provides an efficient framework for detecting subtle emotional differences in speech signals, particularly between boredom and fear manifestations.
- Open Access
- 11 November 2024
- Published by MDPI AG in Preprints.org
Regeneration Filter: Enhancing Mosaic Algorithm for Near Salt & Pepper Noise Reduction
This paper presents a Regeneration filter for reducing near Salt-and-Pepper (nS\&P) noise in images, designed for selective noise removal while simultaneously preserving structural details. Unlike conventional methods, the proposed filter eliminates the need for median or other filters, focusing exclusively on restoring noise-affected pixels through localized contextual analysis in the immediate surroundings. Our approach employs an iterative processing method, where additional iterations do not degrade the image quality achieved after the first filtration, even with high noise densities up to 97\% spatial distribution. To ensure the results are measurable and comparable with other methods, the filter’s performance was evaluated using standard image quality assessment metrics. Experimental evaluations across various image databases confirm that our filter consistently provides high-quality results. The code is implemented in the R programming language, and both data and code used for the experiments are available in a public repository, allowing for replication and verification of the findings.
- 11 November 2024
- Published by MDPI AG in Preprints.org
Open AccessRegeneration Filter: Enhancing Mosaic Algorithm for Near Salt & Pepper Noise Reduction
This paper presents a Regeneration filter for reducing near Salt-and-Pepper (nS\&P) noise in images, designed for selective noise removal while simultaneously preserving structural details. Unlike conventional methods, the proposed filter eliminates the need for median or other filters, focusing exclusively on restoring noise-affected pixels through localized contextual analysis in the immediate surroundings. Our approach employs an iterative processing method, where additional iterations do not degrade the image quality achieved after the first filtration, even with high noise densities up to 97\% spatial distribution. To ensure the results are measurable and comparable with other methods, the filter’s performance was evaluated using standard image quality assessment metrics. Experimental evaluations across various image databases confirm that our filter consistently provides high-quality results. The code is implemented in the R programming language, and both data and code used for the experiments are available in a public repository, allowing for replication and verification of the findings.
- Open Access
- 1 March 2024
- Published by Walter de Gruyter GmbH in Cybernetics and Information Technologies
- Vol. 24 (1) , 137-156
Degradation Recoloring Deutan CVD Image from Block SVD Watermark
People with Color Vision Deficiency (CVD), which arises as a deformation of the M cones in the eye, cannot detect the color green in the image (deutan anomaly). In the first part of the paper, deutan anomalous is described. After that, the image recoloring algorithm, which enables Deutan CVD people to see a wider spectrum in images, is described. Then, the effect of the Recoloring algorithm on images with inserted watermark is analyzed. An experiment has been carried out, in which the effect of the Recoloring algorithm on the quality of extracted watermark and Recoloring image is studied. In addition, the robustness of the inserted watermark in relation to spatial transformations (rotation, scaling) and compression algorithms has been tested. By applying objective measures and visual inspection of the quality of extracted watermark and recoloring image, the optimal insertion factor α is determined. All results are presented in the form of pictures, tables and graphics.
- 1 March 2024
- Published by Walter de Gruyter GmbH in Cybernetics and Information Technologies
- Vol. 24 (1) , 137-156
Open AccessDegradation Recoloring Deutan CVD Image from Block SVD Watermark
People with Color Vision Deficiency (CVD), which arises as a deformation of the M cones in the eye, cannot detect the color green in the image (deutan anomaly). In the first part of the paper, deutan anomalous is described. After that, the image recoloring algorithm, which enables Deutan CVD people to see a wider spectrum in images, is described. Then, the effect of the Recoloring algorithm on images with inserted watermark is analyzed. An experiment has been carried out, in which the effect of the Recoloring algorithm on the quality of extracted watermark and Recoloring image is studied. In addition, the robustness of the inserted watermark in relation to spatial transformations (rotation, scaling) and compression algorithms has been tested. By applying objective measures and visual inspection of the quality of extracted watermark and recoloring image, the optimal insertion factor α is determined. All results are presented in the form of pictures, tables and graphics.
- Open Access
- 2 January 2024
- Published by Brazilian Journals in South Florida Journal of Development
- Vol. 5 (1) , 109-118
The Optimization of the STOI Algorithm Parameters in Presence of the White Gaussian Noise (WGN)
The first part of the paper describes the Short Time Objective Intelligibility - STOI algorithm, which makes an objective evaluation of intelligibility, as well as an algorithm for estimating the optimal parameters of the STOI algorithm, N and . Second part of the paper, described an experiment that examines the intelligibility of the sentences formed from the Serbian Matrix Sentence Test - SMST base using: a). subjective MOS (Mean Opinion Score) test and b). objective test STOI algorithm. Subsequently, a comparative analysis of the results were made and it was determinate an analytical formula which connecting optimal pairs (N, ). Using the results of the MOS test as a reference, a mean absolute error of estimation of the STOI algorithm was determined.
- 2 January 2024
- Published by Brazilian Journals in South Florida Journal of Development
- Vol. 5 (1) , 109-118
Open AccessThe Optimization of the STOI Algorithm Parameters in Presence of the White Gaussian Noise (WGN)
The first part of the paper describes the Short Time Objective Intelligibility - STOI algorithm, which makes an objective evaluation of intelligibility, as well as an algorithm for estimating the optimal parameters of the STOI algorithm, N and . Second part of the paper, described an experiment that examines the intelligibility of the sentences formed from the Serbian Matrix Sentence Test - SMST base using: a). subjective MOS (Mean Opinion Score) test and b). objective test STOI algorithm. Subsequently, a comparative analysis of the results were made and it was determinate an analytical formula which connecting optimal pairs (N, ). Using the results of the MOS test as a reference, a mean absolute error of estimation of the STOI algorithm was determined.
- Open Access
- 31 December 2022
- Published by ISRES Publishing in The Eurasia Proceedings of Science Technology Engineering and Mathematics
- Vol. 21, 116-124
Effect of Protan Recoloring Algorithm on Inserted Watermark in Presence of the Superimposed AWGN
In this paper, the effect of the protan recoloring algorithm on an image with an inserted watermark is analyzed. In addition, the effect of degradation of the quality of the recolored image due to the insertion of watermarks and superimposed AWGN was also analyzed. In the first part of the paper, the effect of color vision deficiency, as a consequence of anomalies in L cones, which is designated as protanomaly (protanopia), is described. People with protanopia do not see a wide range of colors because the colors from the L channel are missing. After that, the recoloring algorithm, which performs recoloring of the image, so that, in protan persons, the visible spectrum of colors is expanded, is described. Then, the algorithm for inserting a watermark, in order to protect copyrights, is described. In the second part of the work, the Experiment, in which the quality of the recoloring image was tested, as well as the quality of the extracted watermark, was described. The results of the experiment are shown with pictures and tables. By applying a comparative analysis of the results of the experiment, using objective (MSE, PSNR, NC and SSIM) and subjective (visual inspection) methods, the insertion factor of the watermark, α, which preserves the quality of the recoloring image as well as the quality of the extracted watermark, was determined. Visual inspection determined the SNR at which visible image degradation and watermarks occur.
- 31 December 2022
- Published by ISRES Publishing in The Eurasia Proceedings of Science Technology Engineering and Mathematics
- Vol. 21, 116-124
Open AccessEffect of Protan Recoloring Algorithm on Inserted Watermark in Presence of the Superimposed AWGN
In this paper, the effect of the protan recoloring algorithm on an image with an inserted watermark is analyzed. In addition, the effect of degradation of the quality of the recolored image due to the insertion of watermarks and superimposed AWGN was also analyzed. In the first part of the paper, the effect of color vision deficiency, as a consequence of anomalies in L cones, which is designated as protanomaly (protanopia), is described. People with protanopia do not see a wide range of colors because the colors from the L channel are missing. After that, the recoloring algorithm, which performs recoloring of the image, so that, in protan persons, the visible spectrum of colors is expanded, is described. Then, the algorithm for inserting a watermark, in order to protect copyrights, is described. In the second part of the work, the Experiment, in which the quality of the recoloring image was tested, as well as the quality of the extracted watermark, was described. The results of the experiment are shown with pictures and tables. By applying a comparative analysis of the results of the experiment, using objective (MSE, PSNR, NC and SSIM) and subjective (visual inspection) methods, the insertion factor of the watermark, α, which preserves the quality of the recoloring image as well as the quality of the extracted watermark, was determined. Visual inspection determined the SNR at which visible image degradation and watermarks occur.