Norrima Mokhtar

HMM based automated wheelchair navigation using EOG traces in EEG

This paper presents a wheelchair navigation system based on a hidden Markov model (HMM), which we developed to assist those with restricted mobility. The semi-autonomous system is equipped with obstacle/collision avoidance sensors and it takes the electrooculography (EOG) signal traces from the user as commands to maneuver the wheelchair. The EOG traces originate from eyeball and eyelid movements and they are embedded in EEG signals collected from the scalp of the user at three different locations. Features extracted from the EOG traces are used to determine whether the eyes are open or closed, and whether the eyes are gazing to the right, center, or left. These features are utilized as inputs to a few support vector machine (SVM) classifiers, whose outputs are regarded as observations to an HMM. The HMM determines the state of the system and generates commands for navigating the wheelchair accordingly. The use of simple features and the implementation of a sliding window that captures important signatures in the EOG traces result in a fast execution time and high classification rates. The wheelchair is equipped with a proximity sensor and it can move forward and backward in three directions. The asynchronous system achieved an average classification rate of 98% when tested with online data while its average execution time was less than 1 s. It was also tested in a navigation experiment where all of the participants managed to complete the tasks successfully without collisions.

Improving Vector Evaluated Particle Swarm Optimisation by Incorporating Nondominated Solutions

The Vector Evaluated Particle Swarm Optimisation algorithm is widely used to solve multiobjective optimisation problems. This algorithm optimises one objective using a swarm of particles where their movements are guided by the best solution found by another swarm. However, the best solution of a swarm is only updated when a newly generated solution has better fitness than the best solution at the objective function optimised by that swarm, yielding poor solutions for the multiobjective optimisation problems. Thus, an improved Vector Evaluated Particle Swarm Optimisation algorithm is introduced by incorporating the nondominated solutions as the guidance for a swarm rather than using the best solution from another swarm. In this paper, the performance of improved Vector Evaluated Particle Swarm Optimisation algorithm is investigated using performance measures such as the number of nondominated solutions found, the generational distance, the spread, and the hypervolume. The results suggest that the improved Vector Evaluated Particle Swarm Optimisation algorithm has impressive performance compared with the conventional Vector Evaluated Particle Swarm Optimisation algorithm.

Development of Automated Image Stitching System for Radiographic Images

Standard X-ray images using conventional screen-film technique have a limited field of view that is insufficient to show the full bone structure of large hands on a single frame. To produce images containing the whole hand structure, digitized images from the X-ray films can be assembled using image stitching. This paper presents a new medical image stitching method that utilizes minimum average correlation energy filters to identify and merge pairs of hand X-ray medical images. The effectiveness of the proposed method is demonstrated in the experiments involving two databases which contain a total of 40 pairs of overlapping and non-overlapping hand images. The experimental results are compared with that of the normalized cross-correlation (NCC) method. It is found that the proposed method outperforms the NCC method in classifying and merging the overlapping and non-overlapping medical images. The efficacy of the proposed method is further indicated by its average execution time, which is about five times shorter than that of the other method.

Feature selection using angle modulated simulated Kalman filter for peak classification of EEG signals

Abstract In the existing electroencephalogram (EEG) signals peak classification research, the existing models, such as Dumpala, Acir, Liu, and Dingle peak models, employ different set of features. However, all these models may not be able to offer good performance for various applications and it is found to be problem dependent. Therefore, the objective of this study is to combine all the associated features from the existing models before selecting the best combination of features. A new optimization algorithm, namely as angle modulated simulated Kalman filter (AMSKF) will be employed as feature selector. Also, the neural network random weight method is utilized in the proposed AMSKF technique as a classifier. In the conducted experiment, 11,781 samples of peak candidate are employed in this study for the validation purpose. The samples are collected from three different peak event-related EEG signals of 30 healthy subjects; (1) single eye blink, (2) double eye blink, and (3) eye movement signals. The experimental results have shown that the proposed AMSKF feature selector is able to find the best combination of features and performs at par with the existing related studies of epileptic EEG events classification.

Bit-depth scalable lossless coding for high dynamic range images

In this paper, we propose a bit-depth scalable lossless coding method for high dynamic range (HDR) images based on a reversible logarithmic mapping. HDR images are generally expressed as floating-point data, such as in the OpenEXR or RGBE formats. Our bit-depth scalable coding approach outputs base layer data and enhancement layer data. It can reconstruct the low dynamic range (LDR) image from the base layer data and reconstructs the HDR image by adding the enhancement layer data. Most previous two-layer methods have focused on the lossy coding of HDR images. Unfortunately, the extension of previous lossy methods to lossless coding does not significantly compress the enhancement layer data. This is because the bit depth becomes very large, especially for HDR images in floating-point data format. To tackle this problem, we apply a reversible logarithmic mapping to the input HDR data. Moreover, we introduce a format conversion to avoid any degradation in the quality of the reconstructed LDR image. The proposed method is effective for both OpenEXR and RGBE formats. Through a series of experiments, we confirm that the proposed method decreases the volume of compressed data while maintaining the visual quality of the reconstructed LDR images.

Tree species classification based on image analysis using Improved-Basic Gray Level Aura Matrix

The study focuses on classifying wood species based on macroscopic image of wood texture.Use dataset of 52 tropical wood species where 100 images are taken from each wood species.I-BGLAM feature extractor is proposed to extract 136 features from the wood texture.Significant improvement compared to previous system that used GLCM feature extractor. Classifying wood species accurately is crucial since incorrect labelling of wood species may incur huge loss to timber industries. An automated wood species recognition system is designed based on image analysis of the wood texture which consists of image acquisition, feature extraction, and classification. There are 100 images captured from each wood sample which are divided into training samples and testing samples. An effective feature extractor is important to extract most discriminant features from the wood texture in order to distinguish the wood species accurately. Therefore, in this paper, a novel feature extractor based on Improved-Basic Gray Level Aura Matrix (I-BGLAM) technique is proposed to extract 136 features from each wood image. Fundamentally, the proposed I-BGLAM feature extractor which focuses on the gray level of the wood images is rotational invariant and has smaller feature dimension since only discriminative features are considered. Then, the proposed system automatically classifies 52 wood species by using backpropagation neural network classifier. The proposed I-BGLAM feature extractor had shown to overcome the limitations of Gray Level Co-occurrence Matrix (GLCM) and conventional BGLAM feature extractors in wood species recognition system. Experiments were performed to determine which dataset would be the most ideal when dividing the 100 wood images into training samples and testing samples. Results showed that the most ideal dataset that should be used is dataset that consists of 80 training samples and 20 test samples. The proposed method showed marked improvement of 97.01% accuracy to the work done previously.

Flashover voltage of insulator string under various conditions

Porcelain insulator has been slowly replaced by composite insulator in power grid nowadays. However, flashover fault still occurs on this insulator type. Under certain conditions, partial discharges may occur on the insulator and lead to a complete flashover. The flashover can cause the breakdown and damage the whole power system. Therefore, this paper reports on the flashover voltage of 11kV composite insulator under different conditions. They are pollution methods, insulator surface condition, water content, operating orientations and the ambient temperature. It was found that the flashover voltage can be improved by using 0° operating orientation angle of the insulator, at lower temperature, dry condition, less insulator surface roughness and no pollution on the insulator. Therefore, a better understanding of flashover voltage under different pollution conditions may be attained from this work.

Automatic Cryptosporidium and Giardia viability detection in treated water

In the automatic detection of Cryptosporidium and Giardia (oo)cysts in water samples, low contrast and noise in the microscopic images can adversely affect the accuracy of the segmentation results. An improved partial differential equation (PDE) filtering that achieves a better trade-off between noise removal and edge preservation is introduced where the compass operator is utilized to attenuate noise while retaining edge information at the cytoplasm wall and around the nuclei of the (oo)cysts. Then the anatomically important information is separated from the unwanted background noise using the Otsu method to improve the detection accuracy. Once the (oo)cysts are located, a simple technique to classify the two types of protozoans using area, roundness metric and eccentricity is implemented. Finally, the number of nuclei in the cytoplasm of each (oo)cyst is counted to check the viability of individual parasite. The proposed system is tested on 40 microscopic images obtained from treated water samples, and it gives excellent detection and viability rates of 97% and 98%, respectively.

Edge sharpening for diabetic retinopathy detection

People with diabetes may face eye problem as a complication of diabetes. These eye problems can cause vision loss and even blindness. There are several lesions that appear such microaneurysms, hemorrhages, cotton wool spots and exudates. Exudates tend to form ring, around area of diseased vessel and appeared as yellowish-white deposits with well-defined edges meanwhile cotton wool spots are grayish-white with poorly defined fluffy edges. Exudates can be highlighted from the background easier rather than cotton wool spots since it has well defined edge. In order to detect these lesions, a proper technique is needed to segment the cotton wool spots and exudates from the background. Therefore, this paper is proposed to sharpen the edge to simplify the segmentation process for cotton wool spots and exudates through ramp width reduction.

Evaluation Of Different Peak Models Of Eye Blink Eeg For Signal Peak Detection Using Artificial Neural Network

There is a growing interest of research being conducted on detecting eye blink to assist physically impaired people for verbal communication and controlling devices using electroencephalogram (EEG) signal. One particular eye blink can be determined from use of peak points. Therefore, the purpose of peak detection algorithm is to distinguish an actual peak location from a list of peak candidates. The need of a good peak model is important in ensuring a satisfy classification performance. In general, there are various peak models available in literature, which have been tested in several peak detection algorithms. In this study, performance evaluation of the existing peak models is conducted based on Artificial Neural Network (ANN) with particle swarm optimization (PSO) as learning algorithm. This study evaluates the performance of eye blink EEG signal peak detection algorithm for four different peak models which are Dumpalas, Acirs, Lius, and Dingles peak models. To generalize the performance evaluation, two case studies of eye blink EEG signal are considered, which are single and double eye blink signals. It has been observed that the best test performance, in average, is 91.94% and 87.47% for single and double eye blink signals, respectively. These results indicate that the Acirs peak model offers high accuracy of peak detection for the two eye blink EEG signals as compared to other peak models. The result of statistical analysis also indicates that the Acirs peak model is better than Dingles and Dumpalas peak models.