Umnozeno za odbranu. - Univerzitet u Beogradu, Saobracajni fakultet, odbranjeno 30.12.2005. - Bib... more Umnozeno za odbranu. - Univerzitet u Beogradu, Saobracajni fakultet, odbranjeno 30.12.2005. - Bibliografija: listovi 117-125. - Abstract.
This paper considers optical IoT sensing for the analysis of cardiovascular state. It is used for... more This paper considers optical IoT sensing for the analysis of cardiovascular state. It is used for skin blood pulsation, i.e. photoplethysmography (PPG) signals acquisition and analysis using DFA (detrended fluctuation analysis) approach. Visual stimulus has been performed during the acquisition process. The results showed that DFA based spectra can be useful in further monitoring of the PPG signals. Analysis was performed using signals recorded from eight young volunteers.
Local binary patterns represent very powerful feature for image classification. It evolved from t... more Local binary patterns represent very powerful feature for image classification. It evolved from the original model to many modifications and adaptations. However, almost all of derived models are based on the basic idea to code each pixel’s 3×3 neighborhood binary. In this paper we propose modification of this idea in order to increase its robustness to Gaussian noise. Instead of calculating differences on 3×3 neighborhood regarding central pixel, we apply calculation of differences regarding average pixel intensity on that neighborhood. This kind of averaging improves classification performances with respect to noise. Proposed method is further tested for classification on two publicly available texture datasets and obtained results, with and without noise addition, prove our assumptions.
Fluorescence in situ hybridization technique is often used for evaluation of HER2 gene status. Th... more Fluorescence in situ hybridization technique is often used for evaluation of HER2 gene status. This gene is amplified in more than a quarter of breast cancer cases. Accurate assessment of HER2 status is important since there is available treatment for HER2 positive breast cancers. Evaluation is done by finding the ratio of HER2 and centromere of chromosome 17 markers in each nucleus. The first step in most algorithms for FISH image evaluation is nuclei segmentation. A method for nuclei segmentation from contrast enhanced image is presented in this paper.
The explosive growth of electronics, computers and telecommunications has marked the recent past.... more The explosive growth of electronics, computers and telecommunications has marked the recent past. Moreover, different multimedia requirements became the driving force in the development of new transfer techniques; asynchronous transfer mode (ATM) being the most promising one. These rather new techniques have already changed the traffic type. According to the ubiquitous variable bit rates the traffic appeared to be fractal. As a consequence of multiplexing of different bursty traffic streams, a multifractal phenomenon, usually described by the multifractal spectrum, arises. This paper explains the complex nature of modern teletraffic. Having in mind the high variability of its performances, the new computational technique based on radial basis function seems to be a promising solution in the approximation of the multifractal
This paper advocates a method for image classification based on new multifractal descriptors. It ... more This paper advocates a method for image classification based on new multifractal descriptors. It is explained in context of existing monofractal and multifractal techniques from literature. Employed multifractal descriptors are based on monofractal blanket method from literature and represented by multifractal spectrum. Generalization of blanket method results in three new multifractal methods based on interpretation of the image. All proposed generalizations provide better classification results than monofractal blanket method. Additionally, a new adaptive technique using multifractal model is presented for low-pass image filtering with edges and textural regions preservation. Quantitative evaluation on natural images shows better filtering results regarding edge preservation than non-adaptive filters.
In this paper we further investigate multifractal models based on morphological dilation and eros... more In this paper we further investigate multifractal models based on morphological dilation and erosion. Multifractal spectrum, as a signature of multifractal behavior of the signal, is inspected regarding its most significant parameters: local dimension scales and boxes used for calculation of box-counting (global) dimensions. Extensive tests done within this research provide directions toward selection of these parameters concerning classification of textural images. It is shown that the adequate choice of parameters of the multifractal model provides significant improvement of the classification.
Radiographic images are widely used for diagnostic purposes and it is realized through visual ins... more Radiographic images are widely used for diagnostic purposes and it is realized through visual inspection. Further enhancement of details in the radiographs can improve inspection and diagnosis certainty. The scope of this paper is analysis of the fractional derivative operator for visual improvement of the chest radiographs. Two implementations of the fractional derivative operator, based on central fractional differences, are evaluated.
Radiographic images are widely used for diagnostic purposes and it is realized through visual ins... more Radiographic images are widely used for diagnostic purposes and it is realized through visual inspection. Further enhancement of details in the radiographs can improve inspection and diagnosis certainty. The scope of this paper is analysis of the fractional derivative operator for visual improvement of the chest radiographs. Two implementations of the fractional derivative operator, based on central fractional differences, are evaluated.
2020 Zooming Innovation in Consumer Technologies Conference (ZINC)
Computer vision techniques implemented in modern vehicles should be designed to distinguish diffe... more Computer vision techniques implemented in modern vehicles should be designed to distinguish different changes in a video sequence, captured by RGB and RGBD cameras mounted in or out a vehicle. Autonomous driving process could improve safety of all passengers by introducing additional sensing. In this paper, we used input data from mentioned cameras acquired with inertial sensor for road roughness as a limiter of velocity. Abrupt changes of the velocity and driver comfort affects the driver’s head position. The head position is monitored using 3D skeleton model and depth information. The results show possibility of detection of the potential risk found for unusual driver behavior. Then, the human control could be taken by safety application and system.
2018 14th Symposium on Neural Networks and Applications (NEUREL), 2018
This paper analyzes possibilities to integrate scanning devices and methods of diagnostics road c... more This paper analyzes possibilities to integrate scanning devices and methods of diagnostics road conditions, into the more global network such is the Internet of Vehicles, connecting vehicles and/or stationary access points. Moreover, initial estimation of real time processing effectivity, based on the interaction of diagnostic equipment with the appropriate road surface material, is considered.
The realization of high-quality 10 MHz crystal oscillators is under consideration. Some methods f... more The realization of high-quality 10 MHz crystal oscillators is under consideration. Some methods for improving the stability and phase noise in OCXO (oven controlled crystal oscillator) are described. High-quality 10 MHz oscillators may take a place in professional devices such as GPS and used for synchronization in digital TV.
There has been a sustained effort in the research community over the recent years to develop algo... more There has been a sustained effort in the research community over the recent years to develop algorithms that automatically analyze heart sounds. One of the major challenges is identifying primary heart sounds, S1 and S2, as they represent reference events for the analysis. The study presented in this paper analyzes the possibility of improving the structure characterization based on shape context and structure assessment using a small number of descriptors. Particularly, for the primary sound characterization, an adaptive waveform filtering is applied based on blanket fractal dimension for each preprocessed sound candidate belonging to pediatric subjects. This is followed by applying the shape based methods selected for the structure assessment of primary heart sounds. Different methods, such as the fractal ones, are used for the comparison. The analysis of heart sound patterns is performed using support vector machine classifier showing promising results (above 95% accuracy). The o...
2013 20th International Conference on Systems, Signals and Image Processing (IWSSIP), 2013
ABSTRACT This paper proposes a face recognition system based on Gabor image representation, princ... more ABSTRACT This paper proposes a face recognition system based on Gabor image representation, principal component analysis and neural networks. Gabor filter bank at 5 scales and 8 orientations is used for face representation. The dimensionality of Gabor representation of face images is reduced by the Principal component analysis. The images are classified using the Feedforward neural network with backpropagation learning algorithm. The ORL database of faces is used for testing.
In the most general case finding the shear stress distribution on the cross-section of prismatic ... more In the most general case finding the shear stress distribution on the cross-section of prismatic bar subjected to torsion presents a serious problem that can be solved in two steps. The first of them consists in finding the stress function, while the second one consists in finding the shear stress distribution by using this function. The stress function appears to be the solution of Poisson’s partial differential equation for the given conditions of unambiguity, which in the theory of elasticity describes torsion of prismatic bars in terms of stresses. The modeling by means of electric networks is one of the few possible means for finding the stress functions. This paper describes the method the cellular neural networks can successfully be applied as analogous models in finding stress functions of twisted prismatic bars with complex polygonal cross-sections. The stress functions produced in this way are straightforwardly applicable in calculation of resulting shear stress distributi...
Umnozeno za odbranu. - Univerzitet u Beogradu, Saobracajni fakultet, odbranjeno 30.12.2005. - Bib... more Umnozeno za odbranu. - Univerzitet u Beogradu, Saobracajni fakultet, odbranjeno 30.12.2005. - Bibliografija: listovi 117-125. - Abstract.
This paper considers optical IoT sensing for the analysis of cardiovascular state. It is used for... more This paper considers optical IoT sensing for the analysis of cardiovascular state. It is used for skin blood pulsation, i.e. photoplethysmography (PPG) signals acquisition and analysis using DFA (detrended fluctuation analysis) approach. Visual stimulus has been performed during the acquisition process. The results showed that DFA based spectra can be useful in further monitoring of the PPG signals. Analysis was performed using signals recorded from eight young volunteers.
Local binary patterns represent very powerful feature for image classification. It evolved from t... more Local binary patterns represent very powerful feature for image classification. It evolved from the original model to many modifications and adaptations. However, almost all of derived models are based on the basic idea to code each pixel’s 3×3 neighborhood binary. In this paper we propose modification of this idea in order to increase its robustness to Gaussian noise. Instead of calculating differences on 3×3 neighborhood regarding central pixel, we apply calculation of differences regarding average pixel intensity on that neighborhood. This kind of averaging improves classification performances with respect to noise. Proposed method is further tested for classification on two publicly available texture datasets and obtained results, with and without noise addition, prove our assumptions.
Fluorescence in situ hybridization technique is often used for evaluation of HER2 gene status. Th... more Fluorescence in situ hybridization technique is often used for evaluation of HER2 gene status. This gene is amplified in more than a quarter of breast cancer cases. Accurate assessment of HER2 status is important since there is available treatment for HER2 positive breast cancers. Evaluation is done by finding the ratio of HER2 and centromere of chromosome 17 markers in each nucleus. The first step in most algorithms for FISH image evaluation is nuclei segmentation. A method for nuclei segmentation from contrast enhanced image is presented in this paper.
The explosive growth of electronics, computers and telecommunications has marked the recent past.... more The explosive growth of electronics, computers and telecommunications has marked the recent past. Moreover, different multimedia requirements became the driving force in the development of new transfer techniques; asynchronous transfer mode (ATM) being the most promising one. These rather new techniques have already changed the traffic type. According to the ubiquitous variable bit rates the traffic appeared to be fractal. As a consequence of multiplexing of different bursty traffic streams, a multifractal phenomenon, usually described by the multifractal spectrum, arises. This paper explains the complex nature of modern teletraffic. Having in mind the high variability of its performances, the new computational technique based on radial basis function seems to be a promising solution in the approximation of the multifractal
This paper advocates a method for image classification based on new multifractal descriptors. It ... more This paper advocates a method for image classification based on new multifractal descriptors. It is explained in context of existing monofractal and multifractal techniques from literature. Employed multifractal descriptors are based on monofractal blanket method from literature and represented by multifractal spectrum. Generalization of blanket method results in three new multifractal methods based on interpretation of the image. All proposed generalizations provide better classification results than monofractal blanket method. Additionally, a new adaptive technique using multifractal model is presented for low-pass image filtering with edges and textural regions preservation. Quantitative evaluation on natural images shows better filtering results regarding edge preservation than non-adaptive filters.
In this paper we further investigate multifractal models based on morphological dilation and eros... more In this paper we further investigate multifractal models based on morphological dilation and erosion. Multifractal spectrum, as a signature of multifractal behavior of the signal, is inspected regarding its most significant parameters: local dimension scales and boxes used for calculation of box-counting (global) dimensions. Extensive tests done within this research provide directions toward selection of these parameters concerning classification of textural images. It is shown that the adequate choice of parameters of the multifractal model provides significant improvement of the classification.
Radiographic images are widely used for diagnostic purposes and it is realized through visual ins... more Radiographic images are widely used for diagnostic purposes and it is realized through visual inspection. Further enhancement of details in the radiographs can improve inspection and diagnosis certainty. The scope of this paper is analysis of the fractional derivative operator for visual improvement of the chest radiographs. Two implementations of the fractional derivative operator, based on central fractional differences, are evaluated.
Radiographic images are widely used for diagnostic purposes and it is realized through visual ins... more Radiographic images are widely used for diagnostic purposes and it is realized through visual inspection. Further enhancement of details in the radiographs can improve inspection and diagnosis certainty. The scope of this paper is analysis of the fractional derivative operator for visual improvement of the chest radiographs. Two implementations of the fractional derivative operator, based on central fractional differences, are evaluated.
2020 Zooming Innovation in Consumer Technologies Conference (ZINC)
Computer vision techniques implemented in modern vehicles should be designed to distinguish diffe... more Computer vision techniques implemented in modern vehicles should be designed to distinguish different changes in a video sequence, captured by RGB and RGBD cameras mounted in or out a vehicle. Autonomous driving process could improve safety of all passengers by introducing additional sensing. In this paper, we used input data from mentioned cameras acquired with inertial sensor for road roughness as a limiter of velocity. Abrupt changes of the velocity and driver comfort affects the driver’s head position. The head position is monitored using 3D skeleton model and depth information. The results show possibility of detection of the potential risk found for unusual driver behavior. Then, the human control could be taken by safety application and system.
2018 14th Symposium on Neural Networks and Applications (NEUREL), 2018
This paper analyzes possibilities to integrate scanning devices and methods of diagnostics road c... more This paper analyzes possibilities to integrate scanning devices and methods of diagnostics road conditions, into the more global network such is the Internet of Vehicles, connecting vehicles and/or stationary access points. Moreover, initial estimation of real time processing effectivity, based on the interaction of diagnostic equipment with the appropriate road surface material, is considered.
The realization of high-quality 10 MHz crystal oscillators is under consideration. Some methods f... more The realization of high-quality 10 MHz crystal oscillators is under consideration. Some methods for improving the stability and phase noise in OCXO (oven controlled crystal oscillator) are described. High-quality 10 MHz oscillators may take a place in professional devices such as GPS and used for synchronization in digital TV.
There has been a sustained effort in the research community over the recent years to develop algo... more There has been a sustained effort in the research community over the recent years to develop algorithms that automatically analyze heart sounds. One of the major challenges is identifying primary heart sounds, S1 and S2, as they represent reference events for the analysis. The study presented in this paper analyzes the possibility of improving the structure characterization based on shape context and structure assessment using a small number of descriptors. Particularly, for the primary sound characterization, an adaptive waveform filtering is applied based on blanket fractal dimension for each preprocessed sound candidate belonging to pediatric subjects. This is followed by applying the shape based methods selected for the structure assessment of primary heart sounds. Different methods, such as the fractal ones, are used for the comparison. The analysis of heart sound patterns is performed using support vector machine classifier showing promising results (above 95% accuracy). The o...
2013 20th International Conference on Systems, Signals and Image Processing (IWSSIP), 2013
ABSTRACT This paper proposes a face recognition system based on Gabor image representation, princ... more ABSTRACT This paper proposes a face recognition system based on Gabor image representation, principal component analysis and neural networks. Gabor filter bank at 5 scales and 8 orientations is used for face representation. The dimensionality of Gabor representation of face images is reduced by the Principal component analysis. The images are classified using the Feedforward neural network with backpropagation learning algorithm. The ORL database of faces is used for testing.
In the most general case finding the shear stress distribution on the cross-section of prismatic ... more In the most general case finding the shear stress distribution on the cross-section of prismatic bar subjected to torsion presents a serious problem that can be solved in two steps. The first of them consists in finding the stress function, while the second one consists in finding the shear stress distribution by using this function. The stress function appears to be the solution of Poisson’s partial differential equation for the given conditions of unambiguity, which in the theory of elasticity describes torsion of prismatic bars in terms of stresses. The modeling by means of electric networks is one of the few possible means for finding the stress functions. This paper describes the method the cellular neural networks can successfully be applied as analogous models in finding stress functions of twisted prismatic bars with complex polygonal cross-sections. The stress functions produced in this way are straightforwardly applicable in calculation of resulting shear stress distributi...
Uploads
Papers by Branimir Reljin