Taiwanese doctoral student invents unique iris recognition technology to prevent hackers from invading

Taiwan National Sun Yat-sen University’s doctoral student in the Department of Electrical Engineering and Institute of Communications Engineering, Lin Guo-Jun, has made a breakthrough in iris recognition cyber security technology after 14 years of global bottleneck, eliminating hacker intrusion vulnerabilities and increasing security levels by thousands of times. His research papers have been published in the prestigious international journal “IEEE Transactions on Dependable and Secure Computing”.

According to a press release from Taiwan National Sun Yat-sen University, high-tech iris recognition, as depicted in the movie “Minority Report”, where Tom Cruise’s character can unlock doors, make payments, and even receive personalized advertisements with just a glance, is omnipresent. Lin Guo-Jun’s interdisciplinary research on iris recognition cyber security technology has led to the acceptance of two high-quality papers in journals, with one paper being included in the prestigious “IEEE Transactions on Dependable and Secure Computing” and earning him the 2024 Foxconn Technology Award.

Lin Guo-Jun stated that in the era of AI advancements, biometric recognition technology provides users with “portable” and unique features compared to traditional methods like passwords and ID cards. The biometric feature is one that is not easily lost or duplicated, thereby reducing the risk of data theft and offering a multitude of applications in daily life. The iris, with its uniqueness, permanence, and high density characteristics, provides a higher level of identification even among identical twins whose irises are unique to each individual. Additionally, iris recognition offers over 240 unique feature points, far superior to the 40 in fingerprints and 68 in frontal facial features, providing an absolute advantage in security performance.

However, Lin Guo-Jun emphasized that once biometric information falls into the hands of hackers, the user risks permanent loss of access rights. Therefore, research on the security of biometric resources is imperative. His study on iris recognition technology revealed that hackers could exploit core concepts and theorems of information theory and coding theory, using error correction codes to aid attacks on encryption systems. This vulnerability had been raised by scholars 14 years ago but had remained unresolved until Lin’s intervention, leaving the advancement of iris security technology in limbo.

Guided by Associate Professor Chen Yan-Ming, Lin Guo-Jun focused on designing error correction code structures, feature extraction, and decoding algorithms in the early stages of his research. This significantly enhanced the security level and recognition efficiency of biometric password systems, making his team the first globally to develop a system that improves biometric recognition efficiency without changing the code rate.

Subsequently, Lin Guo-Jun proposed a method for extracting iris feature code reliability values, implemented template mapping technology based on coding design concepts, and used error correction codes with higher decoding complexity, adjusting error correction codes according to different iris databases to maximize error correction capabilities.

Furthermore, Lin Guo-Jun introduced multiscale dominant feature points, leveraging Gaussian blur filters in image recognition tools to augment the iris feature code by a hundredfold, filtering out a minimal number of feature points for recognition. In summary, the system’s recognition accuracy has significantly improved, with security bits reaching 100 in stringent environments, and security coefficients increasing by a hundredfold, and potentially over a thousandfold in ideal conditions, successfully preventing attacks exploiting error correction codes.

While most people are still unfamiliar with iris recognition, Lin Guo-Jun highlighted that apart from India’s Dubai Airport, the Fast Entry Nexus Program in the United States and Canada also employs iris recognition systems. Previously, major international mobile phone manufacturers had launched iris recognition devices. To enhance iris recognition security in the future, the latest defense technologies must be implemented.

Associate Professor Chen Yan-Ming of the Department of Communications Engineering at Taiwan National Sun Yat-sen University currently serves as the head of the Digital Forensics Unit in the Information Security Center. He praised Lin Guo-Jun for his outstanding presentation skills and in-depth analysis in each research topic to support his arguments. The groundbreaking research outcomes in information security, particularly in biometric recognition and information security, have made significant contributions.