Atiqul Islam

PHYSIOLOGICAL SIGNAL ANALYSIS FOR COGNITIVE STATE ESTIMATION

The purpose of this paper is to identify inconsistency in human physiological signals based on cognitive states by measuring and analyzing bio-signals. In this paper, the cognitive states are estimated using physiological signal analysis. The parameters are electrocardiogram (ECG), electromyogram (EMG), electroencephalogram (EEG) and blood pressure (BP). The signals have been collected using BIOPAC system in which the subjects were induced to undergo the specific sequence of the cognitive state. For getting physiological signals during different conditions, we utilized power point slide show, video clips and question answer method which elicits mental reactions from the subjects. Data is taken before and after four tasks that encompassed the motor action (MA), thought (TH), memory related (MR) and emotion (EM). These measured values are analyzed using BIOPAC Acknowledge software. It was found that the motor action and thought states have effects on BP while MR and EM state mainly affect the ECG measurement. The decibel value and frequency found for EM state in ECG are minimum compared to relaxed state (RS) condition. Similarly, the maximum frequency and dB value is found for MR state. No significant variation was seen for MA and TH states. Thus it was decided that the MR and EM states mainly affect the ECG measurement. For BP the value increases in MA state and decreases in TH state. The MA state mainly affects the EMG signal while other states have no significant changes. The EEG mainly detects the signal of task performed by the specific brain region where the electrodes are placed. In EEG analysis, the electrodes are placed in occipital lobe region which gives mainly the variation in alpha amplitude of EEG with eyes closed and eyes opened. Alpha wave amplitudes vary with the subjects attention to mental tasks performed with eyes closed.

An efficient Optical Character Recognition algorithm using artificial neural network by curvature properties of characters

Optical Character Recognition (OCR) has gained so much importance among the researchers now a day as it is an eminent sector for a Human Computer Interaction (HCI) System. For an efficient recognition system the primary need is a reliable feature extraction process. So far the feature extraction systems used are mainly based on the character pattern, enclosure or symmetry. Still another property which is based on the angular properties of the several predetermined positions can be used for the purpose of feature extraction process that is the main motivation of this work. The effectiveness of the algorithm has been discussed in the experimental result section where the performance has been compared for different number of feature used.

Mental states estimation with the variation of physiological signals

In this paper, we describe the consequences of mental conditions due to the variation of electrocardiogram, electroencephalogram and blood pressure using BIOPAC system. The different data sets are collected using BIOPAC system in which subjects were induced to undergo the specific sequence of mental condition or cognitive state. For getting physiological signals during different affective condition, we utilized power point slide show, video clips and question-answer method which elicit mental reactions from the subjects. Data was taken before and after four tasks that encompassed the Motor Action (MA), Thought (TH), Memory Related (MR), Emotion (EM). These measured values were analyzed using BIOPAC AcqKnowledge software. It was found that the effect of motor action and thought states are mainly on blood pressure while memory related and emotion state mainly affect the ECG measurement. The EEG mainly detects the signal of task performed by the specific brain region where the electrodes are placed. Here the electrodes are placed in occipital lobe region which gives mainly the variation in alpha amplitude of EEG with eyes closed and eyes opened. Alpha wave amplitudes vary with the subjects attention to mental tasks performed with eyes closed.

3D kinematic measurement of human movement using low cost fish-eye cameras

3D motion capture is difficult when the capturing is performed in an outdoor environment without controlled surroundings. In this paper, we propose a new approach of using two ordinary cameras arranged in a special stereoscopic configuration and passive markers on a subject’s body to reconstruct the motion of the subject. Firstly for each frame of the video, an adaptive thresholding algorithm is applied for extracting the markers on the subject’s body. Once the markers are extracted, an algorithm for matching corresponding markers in each frame is applied. Zhang’s planar calibration method is used to calibrate the two cameras. As the cameras use the fisheye lens, they cannot be well estimated using a pinhole camera model which makes it difficult to estimate the depth information. In this work, to restore the 3D coordinates we use a unique calibration method for fisheye lenses. The accuracy of the 3D coordinate reconstruction is evaluated by comparing with results from a commercially available Vicon motion capture system.

Single Electron Transistors (SET) substituting MOSFETs to reduce power consumption of an inverter circuit

Instead of MOSFET Single Electron Transistor (SET) can be used as a voltage controlled switch. The current flowing through a SET is the result of electron tunneling through tunnel junctions of its source and drain. So the supply voltage requirement is also very low. As a result, the power consumption across a Single Electron Transistor is ultra-low in comparison to that of a MOSFET. In this research work simulations have been done with PSPICE for an inverter built with both SETs and MOSFETs. Supply voltage requirement for a SET based inverter was 35mV whereas the supply voltage requirement for a CMOS inverter was 3.5V. SET based inverter consumes almost 5 million times less power than CMOS inverter.

A strategic numerical approach to estimate actual demand in case of systematic data scarcity

Electricity, the most convenient form of energy is one of the most important issues in Bangladesh perspectives. It is very important to know the actual demand (i.e. if there were no load-shed) of electricity for proper planning and managing electricity generation. It will help the decision-makers to provide directions on cost-effective investment and on scheduling the operation of the power plants. But it is very difficult to know the actual demand of Bangladesh due to lack of computerized and systematic recording of data. In this research, a method, supported by statistical test, has been proposed to determine the total demand of Bangladesh in a typical peak hour in the year 2011. This method is suitable for a power system like that in Bangladesh where proper information is not available.

Computational complexity analysis of H.264/AVC video compression by DC mode prediction

The latest video coding standard H.264/AVC achieves better video compression rates since it adopts many new features. The improvement in compression efficiency of H.264 has brought about the dramatic increase in computational complexity. In this paper an algorithm for DC mode prediction employing a new mathematical expression has been developed and analyzed computational complexity against existing algorithm for DC mode prediction. Using the proposed algorithms the encoding time decreased while slightly increasing the PSNR values with no significant rate distortion. Choosing some specific parameters and strategies the experiment has been performed and experimental result have shown that only encoding time has reduced in case of QCIF format but PSNR and bit rate are same. Using CIF format the experimental result have shown that the encoding time was increased.

The effect of the variation of the concentration factor and interdot distance of In x Ga 1−x N/GaN QD—IBSC

Power conversion efficiency of conventional solar devices are low because low energy photons cannot excite carriers to the conduction band, therefore do not contribute to the devices current and high energy photons are not efficiently used due to a poor match to the energy gap. Efficiency of solar cell may be increased in various ways; now-a-days quantum dot intermediate band solar cell (QDIBSC) is most promising approach among them. Introducing intermediate levels into the energy gap of a conventional solar cell; low energy photons can be used to promote charge carriers in a stepwise manner to the conduction band thereby enhancing the current while maintaining a large open-circuit voltage. In this thesis InXGa1-XN/InN quantum-dot intermediate band solar cell is calculated by means of solving Schrödinger equation according to the kronig-penney model. On the basis of particular assumptions, the power conversion efficiency is calculated. The results reveal that the InxGa1-xN/InN quantum dot intermediate-band solar cell gives much larger power conversion efficiency than that of conventional solar cells and the power conversion efficiency strongly depends on the size of the quantum dot and the inter dot distance.