Eye Movement Detection Systems Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

This paper starts with a review of various eye movement detection methods. The study involves comparing them and deciding upon a suitable method. The eye movements have been successfully used in developing assistive technologies. The aim of this study is to identify a convenient method to use for obtaining the eye movements and thereupon find its usefulness in identifying normal and dyslexic children from their eye movements. The method decided upon should not cause any discomfort to the subject, at the same time should be simple, economical, and convenient. Electrooculograms are used to measure the resting potential of the eyes. Electro-oculograms have been proposed in literature for several applications. They are widely used in ophthalmic research and clinical laboratories because they provide a non-invasive method for recording full range of eye movements. Besides the clinical research and laboratories usage, EOG is also broadly used in developing assistive technologies. This paper presents the study conducted in order to obtain certain details from the subjects' eye movements which will further be used to develop an optimal classifier for the application considered.

Key words: Electoroculogram, eye movements, dyslexia, horizontal and vertical movements.

1. Introduction

Eye movement detection has attracted a large number of researchers from many fields in many countries. From as early as the 1950's, many researchers began to investigate eye movements. Magnetic field search coil technique had become the most commonly used methods for quantitative studies of eye and head movements in man and in experimental animals. The technique is based on phase-locked amplitude detection of the voltage induced in a search coil in the external ac magnetic field. This method can be used to detect eye movement in 2-D and 3-D dimensions and it has large linear range and high resolution. However, the magnetic search coil has to be put inside the eye [1]. The subject is not comfortable with this and it cannot be used for long period. Image processing is also widely used to detect horizontal and vertical eye positions. Monitoring the eye movements using video camera has the advantages of being non intrusive. There are many algorithms which have been developed to process the image to get the position of the eye, such as the trigonometry method, and Kalman filter method [2]. Those methods can achieve good resolution. However, the detection system is too bulky and complex. It is not suitable for size constraint projects. There are also many other methods, such as fiber optic eye position sensing method and limbus infrared reflection method. These methods are either too complex, or too large. The video-oculogram [3], which detects eye movements from pictorial images of the eyeball, is expensive because it requires a video camera to film eye movement in real time. Eye movement detection using infrared reflectance of the cornea is difficult to use over a long period because the eyes tend to become dry and fatigued. The sclera reflection method detects eye movements using the differential reflectivity of eyes, but its accuracy is not sufficient for practical application [4]. In these methods, a part of looking is prevented by the devices. Electric wheelchairs controlled by eye movements using Electro-oculograms (EOG) have been developed as a movement support device [5]. EOG is widely used in ophthalmic research and clinical laboratories because it provides a non-invasive method for recording full range of eye movements. Besides the clinical research and laboratories usage, EOG is also broadly used in developing assistive technologies, such as HMI (Human Machine Interface) [9]. Several practical devices have used eye movements as communication support. For example, camera mouse for visual tracking of body features to provide computer access for people with severe disabilities [10], Infrared system for determining ocular position, human-computer interaction using eye-gaze input, Electro ocular switch for communication of the speechless etc. are some of the examples we find in the literatures. In the past few years, the focus on the development of assistive devices for people with severe disability has increased by improving the traditional systems.

There have been much efforts to develop EOG based assistive devices, inspired by the fact that the physical energy drained in moving eyes is much lesser when compared to other gestures such as nodding head (dumb people), speaking or writing etc. Electrooculograms (EOGs) occurring as a result of eye movements, are found to be most advantageous and convenient [7-9]. It causes least discomfort to the patient. The advantages of this technique include recording with minimal interference with subject activities and minimal discomfort. Today the recording of the EOG is a routinely applied diagnostic method in investigating the human oculomotor system. From above studies, it is found out that EOG based method is easy to apply, with total minimized discomfort to the subject. EOG can be used for long term monitoring. These characteristics are required for the purpose of continuing our studies. Therefore, the method based on human EOG is preferred over all other methods.

The paper is organized as follows: Section 2 explains the experimental components including the challenges, electrode configuration and EOG signal acquisition, Section 3 explains the significance of the proposed work and section 4 briefs the conclusion and future considerations.

2 Experimental components

2.1 Challenges

The frequency content of the EOG signal is very close to DC and hence the separation of DC drifts from the useful signal content is a difficult task. The shift in the DC level during EOG signal recording is much observed using any EOG measuring system. Another factor affecting the response of the system is saccadic eyeball motion. It describes quick jumps of the eye from one fixation point to another. The speed may be 20 - 700 â-¦ /s. The artifacts are caused by eyeball rotation and movement, eyelid movement, the EMG signal produced by the muscle of the eye, eye blinks, electrode placement, head movement, influence of luminance and others.

2.2 Electrode Configuration for EOG Based Method

The human eye can be described as a fixed dipole with positive pole at the cornea and negative pole at the retina. The magnitude of this corneoretinal potential is in the range 0.4-1.0 mV. Emil du Bois-Reymond (1848) observed that the cornea of the eye is electrically positive relative to the back of the eye. Electrooculography is a technique for measuring the resting potential of the retina. The main applications of EOG are in ophthalmological diagnosis and in recording eye movements. The Electrooculogram is essentially a record of the difference in electrical voltage between the front and back of the eye that is correlated with eyeball movement and obtained by electrodes placed on the skin near the eye.Horizontal EOG is measured as a voltage by means of electrodes strategically placed as close as possible to each eye. Similarly, vertical EOG is measured as a voltage by means of electrodes placed just above and below the eye. The reference electrode is placed on the forehead. The electrodes can easily be fixed in place without causing any discomfort to a subject. The subject can wear eyeglasses, spectacles and EOG apparatus docs not interfere significantly with the subject's visual field. The direct inspection of eye movement traces by a skilled professional does not provide any reliable method for critical diagnosis of disabilities. At the same time, manual evaluation misses the important quantitative information about the oculomotoric behavior.

EOG is a non imaging mode of eye movement detection. It detects eye movements based on the potential differences around the eyes with electrodes attached around a subject's eye. The recording of the EOG signal has traditionally been associated with several problems. The signal is seldom deterministic, even for the same person in different experiments. It is a result of a number of factors, including eyeball rotation and movements, eyelid movement. The EMG signal produced by the muscle of the eye, eyelid movement, the eye blinks, electrode placement, head movements, influence of luminance, etc.

2.3 EOG Signal Acquisition

BIOPAC MP35/30 is used for EOG data acquisition. Considering the, cost and reliability, Silver (Ag)-Silver Chloride (AgCl) electrodes have been used. After cleaning the surface of the subject's skin, additionally an electrolytic gel is applied to the skin. A total of six electrodes have been used, considering the reference electrode separately for the horizontal and vertical measurements. Channel 1 was used for horizontal measurements and channel 2 for vertical measurements. In both cases EOG 0.05-35 Hz range was fixed. The EOG signals have been obtained for the subject in vertical and horizontal channel. EOG has been recorded for different cases like right and left eye movements, by following a specific object, and also by reading a simple paragraph. The EOG signals were pre-amplified and filtered initially, the noise including power line interference is suppressed through a 50 Hz notch filter. To further remove the noise, a band pass filter with a cut-off frequency of 0.05Hz and 30Hz, and a moving average filter were used. The signals were then amplified.

The sample EOG obtained for a single subject is shown below:

Fig 1. Horizontal and Vertical movements of the eyes

3. Significance of the proposed work

The rotation of the eye to the right results in a difference of potential, with the electrode in the direction of movement, becoming positive with respect to the second electrode. The opposite effect results from the rotation to the left. Literature studies give an insight that the eye movements can be effectively and advantageously used for the social well-being of human beings which fall into the strata of physically disabled, learning disabled and paralytic people. The main aim of the work is to exploit the significance of eye movements in learning disability known as dyslexia. An estimated more than 30 million children are known to be dyslexic in India. Specific reading disability is a common, cognitively and behaviorally heterogeneous developmental condition, characterized primary by severe difficulty in the mastery of reading despite average intelligence and adequate education. It is traditionally defined as an enduring and heavy impairment of reading ability in spite of normal intelligence and adequate educational opportunities. Dyslexics have a specific disorder of written language and can have some associated deficits like: attention deficit, visuo-attentional deficit, auditory and memory deficits. A major difference between dyslexia and other reading disabilities is that, unlike dyslexia, other categories of reading failure can be predicted on the basis of neurological, intelligence, socio-economic, educational, and psychological (motivation, emotional) factors known to adversely effect the reading process. In contrast, if a child has none of these mentioned problems, he is expected to be a normal reader. Children can be classified as dyslexic when their failure to learn to read cannot be predicted by deficiencies in any of the known causes of poor reading. Dyslexia is sometimes called the "invisible disability" because no one knows a person has it unless that person is asked to read or write. The main type of eye movements used during visual scanning and reading are called saccadic eye movements or saccades [11]. Saccades are preceded and followed by pauses or fixations. The brain obtains its information through the eyes only during fixations. Eye movements can be used to pinpoint the specific problems of the reader. The study of eye movements during reading has provided many important insights about the process of reading [11]. The current work is mainly inspired by the G. Pavlidis, who in his work brings out the diagnostic significance of Eye movements in dyslexia and how the eye movements are different for dyslexics and non-dyslexics [11-12].

4. Conclusion and future considerations

In this paper a review of various eye movement detection methods is done. The eye movements have been successfully used in developing assistive technologies. The aim of this study is to identify a convenient method to use for obtaining the eye movements of subjects. The method decided upon should not cause any discomfort to the subject, at the same time should be simple, economical, and convenient. Electrooculogram is preferred over other methods of eye movement detection methods due its simplicity, cost effectiveness and minimal discomfort to the subject. These eye movements will be explored. Features will be extracted from the subjects' eye movements and they will be used in identifying normal and dyslexics. The research work aims at developing a optimal classifier for the application considered.