Aleksandar Milchevski

Ss. Cyril and Methodius University in Skopje · Faculty of Electrical Engineering and Information Techologies

When a wearable ECG sensor transmits signals to a mobile device, the mobile applications needs to be very efficient and save the limited mobile phone resources. This motivates us to find an algorithm implementation that is not computationally intensive, but still very efficient in denoising the ECG signal. The use of a window-based design Finite Im...

We present a novel algorithm for digital filtering of an electrocardiogram (ECG) signal received by both stationary and non-stationary sensors. The basic idea of digital ECG signal processing is to extract heartbeat frequencies, which are found to be normal in the range between 50 and 200 beats per minute. The extracted heartbeat frequency is found...

In this work we tackle the problem of face de-identification in an image. The first step towards a solution to this problem is the design of a successful generic face detection algorithm, which will detect all of the faces in the image or video, regardless of the pose. If the face detection algorithm fails to detect even one face, the effect of the...

In this paper we present a multimodal approach for affective analysis that exploits features from video, audio, Electrocardiogram (ECG), and Electrodermal Activity (EDA) combining two regression techniques, namely Boosted Regression Trees and Linear Regression. Moreover, we propose a novel regularization approach for the Linear Regression in order...

The wavelet transform has been successfully used in the area of power quality analysis. There are many published papers with methods for power quality disturbance classification or harmonics measurement, which use wavelet transform. However, the properties of the wavelet transform can drastically vary from the choice of the wavelet. In this paper w...

In this paper we present a new method for detection and classification of power quality disturbances. Two discrete wavelet transforms with different wavelet filters are used in the feature extraction process. In this way we eliminate the problem of the selection of the most adequate wavelets in the current methods for classification of power qualit...

The wavelet transform has been successfully used in the area of power quality analysis. There are many published papers with methods for power quality disturbances classification or harmonic measurement, which use wavelet transform. However, the properties of the wavelet transform can drastically vary from the choice of the wavelet. In this paper w...

In this paper we present a new method for detection and classification of power quality disturbances. For the feature extraction process we use wavelet analysis. However, the feature vector is extended with three additional features which make the classification more accurate. For the classification of the power disturbances we use SVM (Support Vec...

In this work we propose a subjective no reference ringing metric using machine learning techniques. For every block in a JPEG compressed image the algorithm outputs a value which corresponds to the annoyance of the ringing artifacts. The extracted feature vector is designed bearing in mind the properties of the HVS (Human Visual System) and the rin...

In this work, a novel two phase approach is proposed for robust super-resolution in the presence of registration errors and outliers. In the first phase machine learning method is used to create a weight matrix for every LR image indicating the presence of registration errors. In the second phase, super-resolution is performed using all of the LR i...

In this paper an improvement of wavelet based methods for detection and classification of power quality disturbances is presented. In the feature extraction process wavelet analysis is also used as in the comparing methods. However, the feature vector is extended with three other coefficients in order to improve the accuracy of the algorithm. In or...