Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
Review
. 2022 Nov 15;22(22):8810.
doi: 10.3390/s22228810.

Review and Evaluation of Eye Movement Event Detection Algorithms

Birtukan Birawo  1 , Pawel Kasprowski  1
Affiliations
Review

Review and Evaluation of Eye Movement Event Detection Algorithms

Birtukan Birawo et al. Sensors (Basel). .

Abstract

Eye tracking is a technology aimed at understanding the direction of the human gaze. Event detection is a process of detecting and classifying eye movements that are divided into several types. Nowadays, event detection is almost exclusively done by applying a detection algorithm to the raw recorded eye-tracking data. However, due to the lack of a standard procedure for how to perform evaluations, evaluating and comparing various detection algorithms in eye-tracking signals is very challenging. In this paper, we used data from a high-speed eye-tracker SMI HiSpeed 1250 system and compared event detection performance. The evaluation focused on fixations, saccades and post-saccadic oscillation classification. It used sample-by-sample comparisons to compare the algorithms and inter-agreement between algorithms and human coders. The impact of varying threshold values on threshold-based algorithms was examined and the optimum threshold values were determined. This evaluation differed from previous evaluations by using the same dataset to evaluate the event detection algorithms and human coders. We evaluated and compared the different algorithms from threshold-based, machine learning-based and deep learning event detection algorithms. The evaluation results show that all methods perform well for fixation and saccade detection; however, there are substantial differences in classification results. Generally, CNN (Convolutional Neural Network) and RF (Random Forest) algorithms outperform threshold-based methods.

Keywords: event detection algorithms; eye movement events; eye tracking; fixations; saccades.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure 1
Figure 1
Graphical presentation of eye movement events for the horizontal axis.
Figure 2
Figure 2
Example from one data file from the dataset. The (left) chart shows the raw gaze data, while the (right) one shows a sequence of fixations as annotated by one of the manual coders.
Figure 3
Figure 3
The accuracy for fixations and saccades of the I-DT algorithm for different dispersion thresholds.
Figure 4
Figure 4
The accuracy for fixations and saccades of the I-VT algorithm for different velocity thresholds.
Figure 5
Figure 5
The RF confusion matrix.
Figure 6
Figure 6
The architecture of the CNN used in the experiment.
Figure 7
Figure 7
Confusion matrix for the CNN Classifier.
Figure 8
Figure 8
Eye fixations obtained from the I-VT algorithm at optimum threshold value of 3.5 px/ms. It is visible that many fixations occur nearby and could probably be combined together.
Figure 9
Figure 9
Eye fixations obtained from the RF algorithm. Compared to Figure 8, there are far fewer fixations.
See this image and copyright information in PMC

References

    1. Klaib A.F., Alsrehin N.O., Melhem W.Y., Bashtawi H.O., Magableh A.A. Eye tracking algorithms, techniques, tools and applications with an emphasis on machine learning and Internet of Things technologies. Expert Syst. Appl. 2021;166:114037. doi: 10.1016/j.eswa.2020.114037. - DOI
    1. Punde P.A., Jadhav M.E., Manza R.R. A study of eye tracking technology and its applications; Proceedings of the 2017 1st International Conference on Intelligent Systems and Information Management (ICISIM); Aurangabad, India. 5–6 October 2017; Piscataway, NJ, USA: IEEE; 2017. pp. 86–90.
    1. Naqvi R.A., Arsalan M., Park K.R. Fuzzy system-based target selection for a NIR camera-based gaze tracker. Sensors. 2017;17:862. doi: 10.3390/s17040862. - DOI - PMC - PubMed
    1. Naqvi R.A., Arsalan M., Batchuluun G., Yoon H.S., Park K.R. Deep learning-based gaze detection system for automobile drivers using a NIR camera sensor. Sensors. 2018;18:456. doi: 10.3390/s18020456. - DOI - PMC - PubMed
    1. Braunagel C., Geisler D., Stolzmann W., Rosenstiel W., Kasneci E. On the necessity of adaptive eye movement classification in conditionally automated driving scenarios; Proceedings of the Ninth Biennial ACM Symposium on Eye Tracking Research & Applications; Charleston, SC, USA. 14–17 March 2016.

LinkOut - more resources