摘要:
A polarized image acquisition section shoots a subject through a polarizing element operable to set principal axes of which directions are different from each other. An incident plane specifying section specifies an incident plane of each pixel, and an incident angle computation section computes an incident angle of each pixel. A classifying section clusters pixels similar to each other in both incident plane and incident angel. A reflection component separation section performs reflection component separation on each clustered pixel set on the assumption of probabilistic independence between the diffuse reflection component and the specular reflection component.
摘要:
A polarized image acquisition section shoots a subject through a polarizing element operable to set principal axes of which directions are different from each other. An incident plane specifying section specifies an incident plane of each pixel, and an incident angle computation section computes an incident angle of each pixel. A classifying section clusters pixels similar to each other in both incident plane and incident angle. A reflection component separation section performs reflection component separation on each clustered pixel set on the assumption of probabilistic independence between the diffuse reflection component and the specular reflection component.
摘要:
An eye (EY) of a person to be shot is illuminated coaxially with the optical axis of a camera (11), and then, is shot. At this time, a retinal reflex is caused and the luminance in the pupil region becomes high if the eye (EY) is a living eye. A living eye judgment section (17) judges whether an eye included in the image is a living eye is performed based on the luminance in the pupil region of the image captured by the camera (11).
摘要:
An eye (EY) of a person to be shot is illuminated coaxially with the optical axis of a camera (11), and then, is shot. At this time, a retinal reflex is caused and the luminance in the pupil region becomes high if the eye (EY) is a living eye. A living eye judgment section (17) judges whether an eye included in the image is a living eye is performed based on the luminance in the pupil region of the image captured by the camera (11).
摘要:
A plurality of iris images are acquired (SA0), and aggregation of iris images of which distribution of pupil openings is uniform is acquired from the plurality of iris images by duplication and/or deletion (SA1). Features are generated from the respective iris images that belong to the aggregation (SA2), and a predetermined number of registration features are selected from the features, using authentication performance as an evaluation index.
摘要:
A plurality of iris codes are registered for each registrant in an iris database (12) together with pupil diameter-iris diameter ratio R. At the time of authentication, an iris code is obtained from a captured iris image by feature extraction while pupil diameter-iris diameter ratio R is obtained. Ratio R obtained at the time of registration and ratio R obtained at the time of authentication are compared to specify an appropriate iris code from the iris database (12) as an item to be collated before authentication.
摘要:
A plurality of iris images are acquired (SA0), and aggregation of iris images of which distribution of pupil openings is uniform is acquired from the plurality of iris images by duplication and/or deletion (SA1). Features are generated from the respective iris images that belong to the aggregation (SA2), and a predetermined number of registration features are selected from the features, using authentication performance as an evaluation index.
摘要:
An eye (EY) of a person to be shot is illuminated coaxially with the optical axis of a camera (11), and then, is shot. At this time, a retinal reflex is caused and the luminance in the pupil region becomes high if the eye (EY) is a living eye. A living eye judgment section (17) judges whether an eye included in the image is a living eye is performed based on the luminance in the pupil region of the image captured by the camera (11).
摘要:
In iris authentication, iris regions are extracted from a plurality of images (2a, 2b). The extracted iris regions are subjected to polar coordinate transformation (3a, 3b) and band limitation (4a, 4b). Thereafter, an iris code is generated from the coordinate-transformed, band-limited iris regions (5a, 5b). The iris code is subjected to rotational compensation (6a, 6b). A plurality of resultant iris codes are synthesized into a single iris code by determining the bit value of each bit based on a majority rule (7a, 7b).
摘要:
A plurality of iris images are acquired (SA0), and aggregation of iris images of which distribution of pupil openings is uniform is acquired from the plurality of iris images by duplication and/or deletion (SA1). Features are generated from the respective iris images that belong to the aggregation (SA2), and a predetermined number of registration features are selected from the features, using authentication performance as an evaluation index.