Yunyong Punsawad

Hybrid EEG-EOG brain-computer interface system for practical machine control

Practical issues such as accuracy with various subjects, number of sensors, and time for training are important problems of existing brain-computer interface (BCI) systems. In this paper, we propose a hybrid framework for the BCI system that can make machine control more practical. The electrooculogram (EOG) is employed to control the machine in the left and right directions while the electroencephalogram (EEG) is employed to control the forword, no action, and complete stop motions of the machine. By using only 2-channel biosignals, the average classification accuracy of more than 95% can be achieved.

Palmprint image enhancement using phase congruency

This paper proposes the use of the phase congruency to enhance the palmprint image used for palmprint identification. By using phase congruency, the palmprint lines which act like the edges in the palmprint image are detected. The resulting phase image is shown as the enhanced palmprint image. Comparing with the previous palmprint enhancement method that uses the phase symmetry proposed by Kovesi et al. (1997), the proposed method is significantly less sensitive to the textures which are not the palmprint lines in the palmprint image.

Motion visual stimulus for SSVEP-based BCI system

Steady-state visual evoked potential (SSVEP)- based brain-computer interface (BCI) system is one of the most accurate assistive technologies for the persons with severe disabilities. However, the existing visual stimulation patterns still lead to the eyes fatigue. Therefore, in this paper, we propose a novel visual stimulator using the idea of the motion visual stimulus to reduce the eyes fatigue while maintaining the merit of the SSVEP phenomena. Two corresponding feature extractions, i.e. 1) attention detection and 2) SSVEP detection, are also proposed to capture the phenomena of the proposed motion visual stimulus. Two-class classification accuracy of both features is approximately 80%, where the maximum accuracy using the attention detection is 90%, and the maximum accuracy using the SSVEP detection is 100%.

Hybrid SSVEP-motion visual stimulus based BCI system for intelligent wheelchair

This paper proposes the hybrid BCI modalities for wheelchair control by taking into account weakness of the current BCI systems. The idea is to combine two hybrid BCI systems with the intelligent wheelchair for three states, i.e. normal, fatigue, and emergency states. First system is the hybrid steady state visual evoked potential (SSVEP) and alpha rhythm BCI which is designed to use in the normal state. Second system is the hybrid motion visual stimulus and alpha rhythm which can be employed during the fatigue state (after using the first system). For the experiment, subjects are asked to perform SSVEP system for 30 minutes (until the fatigue states occur). Then, the subjects will be asked to perform the hybrid motion visual stimulus and alpha rhythm testing. The accuracy of the proposed system during fatigue state is approximately 85.62%. With this idea, BCI controlled wheelchair can be efficiently employed in reality.

Wireless sensor network-based smart room system for healthcare monitoring

In this paper, the utilization of Zigbee as wireless sensor network (WSN) for medical application is demonstrated. The combination of various topologies is used to configure wireless sensors network to achieve high efficiency network architecture in medicine. The network consists of center coordinator, routers and sensor nodes. Mesh network is used for the connection between coordinator and router for range expansion. A performance of the proposed modality is tested in the normal situation. Besides, the architecture of the smart room systems is also proposed for healthcare monitoring. Physiological data and signal are transmitted using Xbee which is a wireless device operated in unlicensed radio frequency bands.

Bci-based assistive robot arm

People who lost their limbs by injury or congenital missing need prosthesis to replace the missing body part to assist or enhance the motor ability or for cosmetic purpose. In this paper, brain-computer interface (BCI) technology is proposed to assist the person with disability who has no arm. The proposed system includes two BCI algorithms, i.e. ERD/ERS algorithm and hybrid EEG-EOG algorithm. The designed assistive robot arm is light weight, low power consumption, user friendly and pleasing aesthetic. The ERD/ERS algorithm can achieve the accuracy of approximately 66% with 3 commands. Moreover, the accuracy of the hybrid EEG-EOG algorithm yields nearly 96%.

On the Performance Comparison of Using Checkerboard and Flash Ball Visual Stimulators for SSVEP-based BCI system

Searching for the visual stimulators that can reduce eye-fatigue and enhance the performance of the steady state visual evoked potential (SSVEP)-based brain computer interface (BCI) system is one of the challenging problem in BCI research. In this paper, we compare the multi-command selection performance of the two visual stimulators, i.e. the proposed flash ball and the conventional checkerboard patterns, on the liquid crystal display (LCD) screen. Each visual stimulator consists of 12 commands with different flickering frequencies. These visual stimulators are evaluated with four subjects. Fast Fourier transform FFT is used to simply extract the features of interest. The results show that, by employing the flash ball pattern, the average accuracy is 76.67% which is higher than 60% accuracy of the checkerboard pattern.

On the Selection of Biosignals for the Mental Fatigue Alarm System

Mental fatigue is one of the main causes on the traffic accidents. Seeking for the biosignal that can predict this human phenomenon is a challenging problem. In this paper, we setup the experiment to investigate the possible biosignal that can distinguish the metal fatigue from the normal condition. Simulated driving condition is created. The experimental results show that the average respiratory interval and the average eye blinking interval can efficiently distinguish between the normal and fatigue conditions.

On the enhancement of training session performance via attention for single-frequency/multi-commands based steady state auditory evoked potential BCI

To solve the eye fatigue problem on using the well known steady state visual evoked potential (SSVEP)-based brain-computer interface (BCI) system, the steady state auditory evoked potential (SSAEP) becomes one of the promising BCI modalities. However, SSAEP-based BCI system still suffers from the low accuracy. To increase the accuracy, in this paper, we propose the new training method to enhance the SSAEP training session. The training process is enhanced by making the users control their attention levels simultaneously with the detected auditory stimulus frequency. Furthermore, with the proposed training method, we also propose the corresponding single-frequency/multi-commands BCI paradigm. With the proposed paradigm, four commands can be detected by using only one auditory stimulus frequency. The proposed training system yields approximately 81% accuracy compared with 66% of the session without performing the proposed training.

User performance evaluation with visual stimulator regulation of SSVEP-based BCI system

This paper proposes the relationship between user performance and luminance of visual stimulator in steady state visual evoked potential (SSVEP) based brain computer interface BCI system. The luminance conditions that relate to the environment used to study an eye or visual fatigue. The highlight of the work is the level of eye fatigue detection model in real time by using electroencephalogram (EEG) and skin temperature (SKT). We would like to invent a protocol and guideline of SSVEP based BCI system design. The results present the model that a number of trials leads to a low efficiency of user performance of SSVEP based BCI system. Visual stimulator is also a factor of an occurring of eye fatigue. Clearly, SSVEP is easily activated by high luminance, but it can also rapidly activate eye fatigue. This paper is useful for designing and enhancing the SSVEP based BCI system.