| Autor: |
Balakrishnan, D., Mariappan, Umasree, Revuru, Sai Thanooj Kumar, Sameer, C., Yalla, Satya Praveen Kumar Reddy, Bayya, Pranav Naadh |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2024, Vol. 3180 Issue 1, p1-9, 9p |
| Abstrakt: |
Iris-based human identification detection is used in iris recognition technology to authenticate and verify the identity of individuals for security reasons. It entails recording and analyzing the particular patterns found in a person's iris, which remain steady throughout time. Deep learning approaches also play an important part in iris-based human identity recognition, harnessing the capacity of neural networks to increase accuracy and reliability of the identification process. Deep learning enables iris identification algorithms to extract high-level representations and understand complicated patterns within iris images, hence increasing their discriminatory skills. In this article, a learning technology called CNN. This adapt and capture complex patterns and subtle variations in iris texture, resulting in more robust and accurate identification. CNN can be trained on large datasets to learn the different individuals' irises, enabling them to differentiate between individuals with a high precision. Moreover, CNN enable end-to-end learning, where the entire iris recognition pipeline, from image preprocessing to feature extraction and classification, can be optimized jointly, further improving overall system performance. The flexibility and capability of CNN make it a powerful tool in advancing iris-based human identity detection systems, contributing to more reliable and efficient security applications. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
