Slobodan Marko Beroš

The Multiresonant Converter Steady-State Analysis Based on Dominant Resonant Process

The steady-state analysis of the multiresonant converter is presented. The multiresonant converter utilizes multielement resonant network that consists of resonant inductor and switch capacitances. The resonant process between resonant inductor and capacitances across rectifiers has been found as dominant resonant process. All key circuit parameters are normalized in terms of the characteristic impedance and resonant frequency given by resonant inductor and capacitances across rectifiers. The effects of the transformer magnetizing and leakage inductances have been analyzed. It is shown that dc output values and resonant current rms value strongly depend on load and switching frequency to resonant frequency ratio. It is shown that the resonant process of the resonant inductor and MOSFET output capacitances have no significant influence on overall dc conversion ratio at heavy loads. The simplified analysis is presented and resonant tank order is reduced. The circuit utilizes four different modes of operation. In each mode of operation, the 2-D state-plane diagram is derived. The dc conversion ratio, normalized rms MOSFET, and rectifier capacitance currents have been calculated. A 2.5-kW prototype has been built and tested.

Pre-Processing for Image Sequence Visualization Robust to Illumination Variations

Several images (a sequence) may be used to obtain better image quality. This method is perfect for super-resolution algorithms, which improve sub-pixel clarity of the image and allow a more detailed view. It is possible that illumination variations, e.g. those caused by a light source, lessen the benefits of super-resolution algorithms. The reduction of the quantity of such occurrences by stabilizing variations is important. An enhanced stabilization algorithm is proposed for purposes of reduction of variations in illumination. It is based on the energy contained in wavelet coefficients. In the proposed algorithm, energy plays a role of the memory buffer in memory-based techniques of illumination variation reduction. The benefits of the proposed image stabilization are the higher quality of images and better visualization. Possible applications are in surveillance, product quality control, engine monitoring, corrosion monitoring, micro/nano microscopy, etc.

Wavelet Compression/Reconstruction and Visualization of Pulmonary X-Ray Images for Achieving of Asbestosis Infected Patients Data

An algorithm for reliable wavelet compression/reconstruction and visualization of pulmonary X-ray is presented in this chapter. Pulmonary X-rays are obtained by real patients from an asbestos factory. The aim is to make job easier to occupational medicine specialists and radiologists. Algorithm is primarily concerned for correct compression of the images to save space (digital memory space as well as space for storing X-ray films). Specialists must, according to law, save all X-ray images over 40 years. Instead of archiving X-ray films this algorithm allows saving of wavelet coefficients vectors on magnetic or optical storage. Independent radiologists confirmed that medical data is unchanged. Secondary concern is to emphasize possible asbestos-infected areas, which covers for visualization part of the work. Benefits are in monitoring of health condition, prevention of disease, early diagnostics, more reliable diagnostics, and saving space for achieving medical data.

Discontinuity Detection in the Vibration Signal of Turning Machines

The chapter deals with the detection of discontinuities in the vibration signal created by a turning machine in soft and hard processes. This chapter presents an experiment with two embedded piezo-electrical sensors on the turning machine tool tip. AISI 4140 steel workpieces are used in the experiments to obtain the analyzed signal. The purpose of the experiment is to identify the nature and position of rapid changes as a measure of compliance with surface roughness. A new algorithm for wavelet selection is developed and the procedure proved the Daubechies wavelet of the 6th order to be the best choice. The proposed algorithm is based on a new technique of energy matching for wavelets. Due to the algorithm’s efficiency, it is well suited for real time monitoring. Furthermore, it is possible to build a real time monitoring system based on the digital signal system.

Visualization of Global Illumination Variations in Motion Segmentation

Motion segmentation is the most important part in many applications, such as surveillance, security, monitoring, recognition, etc. The presented research deals with short-term illumination variations in video streams. Illumination variations influence values of pixels and greatly impact the segmentation mask obtained as a part of a motion detection algorithm. In order to subjectively visualize the extent of variations, these must be emphasized. The chapter presents a wavelet energy model based algorithm which detects and emphasizes illumination variations.