“To account for the results, the researchers noted that diabetes can affect the eyes in a number of ways that may not be obvious, causing retinal damage, cataracts, and glaucoma.”
A recent study published in Biomedical Engineering examined the relationship between iris recognition and diabetes, finding that iris recognition is less accurate for people with the condition. The study was conducted with a dataset of 1,300 iris images, which included healthy irises and the irises of people with diabetes, though it excluded any images with obvious optical impairments.
The researchers conducted the same test with four different iris recognition systems to make sure they were testing the eyes, and not the quality of different algorithms. In each case, the results were the same. All four systems had an easier time identifying healthy irises, and were less accurate when scanning diabetic eyes. To account for the results, the researchers noted that diabetes can affect the eyes in a number of ways that may not be obvious, causing retinal damage, cataracts, and glaucoma.
Thanks to the unique nature of the human eye, iris recognition is becoming an increasingly popular form of identification for government programs, law enforcement, and access control. However, the study highlights the limits of this kind of biometric tech, and cautions against the widespread use of modalities that only perform well in optimal circumstances and do not account for relatively common conditions like diabetes.
Source: Science Discoveries
August 20, 2019 – by Eric Weiss