TY - JOUR AU - Zare Bidaki, Ehsan AU - Wong, Alexander AU - Murphy, Paul PY - 2023/05/10 Y2 - 2024/03/28 TI - A Novel Computational Thermal-Visual Imaging System for Automatic Cornea Temperature Measurement and Tracking JF - Journal of Computational Vision and Imaging Systems JA - J. Comp. Vis. Imag. Sys. VL - 8 IS - 1 SE - Articles DO - 10.15353/jcvis.v8i1.5362 UR - https://openjournals.uwaterloo.ca/index.php/vsl/article/view/5362 SP - 20-23 AB - <table style="height: 416px;" width="699"><tbody><tr><td width="711">Ocular surface temperature (OST) is affected by changes in the<br>physiology of the eye caused by normal homeostasis, environmental<br>changes, or systemic and local disease. OST can help a physician<br>to diagnose eye disease with improved accuracy and provide<br>useful information for eye research. This paper presents a novel<br>system including novel hardware design and novel algorithms to<br>measure and track OST from the cornea automatically over any period<br>of time. The system uses an IR camera and a visible light camera<br>to capture synchronous thermal and visible videos, respectively,<br>from the eye surface. The frames for the two video sequences are<br>then registered and the cornea was segmented using a semantic<br>segmentation method. At the final step, the corneal area was localized<br>on the registered thermal frames to extract temperature information.<br>The mean square error for the registration was 5.03 ±1.82<br>and the mean Intersection over Union (IoU) was 94.6%, representing<br>the accuracy of corneal segmentation. A system for measuring<br>and tracking eye surface temperature over time was developed. The<br>system is able to localise the cornea on both visible and thermal images<br>and report temperature profiles of the cornea over the period<br>of measurement. Experimental results shows that the whole system<br>can work as a tool for measuring and tracking OST over time.</td></tr></tbody></table> ER -