29-04-2017, 01:19 PM
With the pronounced need for reliable personal identification, iris recognition has become an important enabling technology in our society. Although an iris pattern is, of course, an ideal identifier, the development of a high performance iris recognition algorithm and its transfer from the research lab to practical applications remains a difficult task. Automatic iris recognition has to deal with unpredictable variations of iris images in real-world applications. For example, recognition of poor-quality iris images, non-linearly distorted iris images, remote iris images, moving iris images and false iris images are open problems in iris recognition. A basic work to solve problems is to design and develop a database of high quality iris images that includes all these variations. In addition, a new iris image database can help identify some border problems in iris recognition and leads to a new generation of iris recognition technology.
CASIA-Iris has been released to the international biometric community and updated from CASIA-IrisV1 to CASIA-IrisV3 since 2002. More than 3,000 users from 70 countries or regions have downloaded CASIA-Iris and A lot of excellent work on iris recognition has been Made based on these iris image databases. While great progress has been made in iris recognition since the 1990s, the rapid growth of iris recognition applications has clearly highlighted two challenges: usability and scalability.
Usability is the largest bottleneck of current iris recognition. It is a trend to develop long-range iris image acquisition systems for user-friendly authentication. However, iris images captured at a distance are more challenging than traditional close-up iris images. The lack of long-range iris image data in the public domain has hampered the research and development of next-generation iris recognition systems.
Most current iris recognition methods have typically been evaluated in medium-sized iris image databases with a few hundred subjects. However, more and more large-scale iris recognition systems are deployed in real-world applications. Many new problems are fulfilled in the classification and indexing of large-scale iris image databases. Therefore, scalability is another challenging problem in iris recognition.
CASIA-Iris has been released to the international biometric community and updated from CASIA-IrisV1 to CASIA-IrisV3 since 2002. More than 3,000 users from 70 countries or regions have downloaded CASIA-Iris and A lot of excellent work on iris recognition has been Made based on these iris image databases. While great progress has been made in iris recognition since the 1990s, the rapid growth of iris recognition applications has clearly highlighted two challenges: usability and scalability.
Usability is the largest bottleneck of current iris recognition. It is a trend to develop long-range iris image acquisition systems for user-friendly authentication. However, iris images captured at a distance are more challenging than traditional close-up iris images. The lack of long-range iris image data in the public domain has hampered the research and development of next-generation iris recognition systems.
Most current iris recognition methods have typically been evaluated in medium-sized iris image databases with a few hundred subjects. However, more and more large-scale iris recognition systems are deployed in real-world applications. Many new problems are fulfilled in the classification and indexing of large-scale iris image databases. Therefore, scalability is another challenging problem in iris recognition.
