Hironobu Takano

Highly reliable liveness detection method for iris recognition

The importance of personal authentication is increasing with the development of the information society. The accuracy of personal authentication by identifying the iris is higher than that by using other biometric traits such as faces or fingerprints. However, the iris authentication system is vulnerable to deception by a fake iris even though the recognition accuracy is high. In this study, we developed a liveness detection method by using a variation in the brightness of an iris pattern induced by a pupillary reflex. The live and artificial irises were classified by a decision threshold of 7% brightness variation rate.

Pupil Diameter Variation in Positive and Negative Emotions with Visual Stimulus

The severely disabled people, e.g. ALS (Amyotrophic Lateral Sclerosis), cannot move their head and hands and desire to communicate with other people. They often feel discomfort because they cannot indicate their intention. To respond their request, the communication devices for disabled people have been developed. However, a contact-type of the communication device using EEG or EMG is burdensome for disabled people because these devices require to put electrodes on their body. Thus, the simplified non-contact type of the communication device is needed. To develop the communication system for severely disabled people, we focused on the characteristics of the pupil diameter variation due to the emotional change and the degree of attention and interest. In this paper, we investigate the pupil diameter variation by using the visual stimuli with positive or negative affective images. From experimental results, the pupil diameter in the positive affective stimulus is significantly smaller than that in the negative one during 2 to 3 second after the visual stimulus. This experimental result indicates that the time variation of the pupil diameter can estimate the emotional states in the positive and negative affections.

Development of the eye input device using eye movement obtained by measuring the center position of the pupil

In this study, we propose a novel eye input device using only eye movement without the calibration for correcting the gaze direction. The traditional eye input device using eye positional information requires head movement. However, the severely disabled people who are unable to move their head, e.g. amyotrophic lateral sclerosis, are difficult to use this device. Although an eye-gaze input device does not require the head movement, the calibration for correction of the gaze direction induces the degradation of usability. In this paper, we propose the eye input device using the eye movement derived from the relative position of the pupil. The proposed system allows users to control a cursor of a personal computer using only eye movement. The cursor control method using the length and direction of the eye movement vector connecting from the reference point to the center position of pupil eliminates the calibration for the gaze direction correction and the head movement control. The experiments were performed to evaluate the operational performance and usability of the proposed eye input device. From experimental results, the operation time of the proposed eye input device became shorter with the increase of trial numbers. Furthermore, the parameter setting for individual improved the operational performance of the eye input device.

Suitable checkpoints of features surrounding the eye for eye tracking using template matching

In order to investigate whether or not checkpoints of features surrounding the eye are suitable for eye tracking, we evaluated the checkpoints using statistical hypothesis testing. By combining the checkpoints of features surrounding the eye, the eye tracking system using the template matching was effective for the eye detection. The parameters of the checkpoints were empirically determined to detect the eye in any individual. The statistical hypothesis testing revealed that the empirical setting values of the checkpoints were significantly different from actual facial features obtained using face images. We investigated suitable parameters of the checkpoints derived from actual facial features obtained with facial images. It was revealed that new checkpoint settings were effective in drastically reducing the possibility of incorrect eye detection, although the correct detection rate was slightly decreased.

Iris Recognition Independent of Rotation and Ambient Lighting Variations

We previously proposed a rotation spreading neural network (R-SAN net). This neural net can recognize the orientation of an object irrespective of its shape, and its shape irrespective of its orientation. The characteristics of the R-SAN net are to spread information about the axis orientation of the object in relation to surrounding space by spreading weight with similar tuning characteristics to axis orientation neurons in the parietal cortex. The R-SAN net is suitable for orientation recognition of a concentric circular pattern such as iris image, because it uses polar conversion. Previously, variations of ambient lighting conditions affected iris detection. In this study, we introduced brightness normalization for accuracy improvement of iris detection in various lighting conditions. Brightness normalization provided high accuracy iris extraction in severe lighting conditions. We also attempted real-time individual identification with the iris patterns. In the recognition experiment, the R-SAN net combined with the minimum distance was able to accurately recognize the orientation and shape of iris images in arbitrary orientation. We suggest application of the R-SAN net as an iris recognition system.

Rotation independent iris recognition by the rotation spreading neural network

We proposed a iris recognition system using the rotation spreading neural network (R-SAN net). The R-SAN net correctly recognized the orientation of iris images using the iris pattern alone, not the positional arrangement of respective face parts. The orientation recognition performance of R-SAN net allows the accurate compensation of the orientation variation. For the characteristics of the iris pattern recognition, the equal error rate was 0.79%, which was investigated with iris images acquired from 19 subjects. On the other hand, the liveness detection method using a variation in the brightness of an iris pattern induced by a pupillary reflex was developed. The live and artificial irises were classified by a decision threshold of 3.7% brightness variation rate.

Rotation Invariant Iris Recognition Method Adaptive to Ambient Lighting Variation

We previously proposed a rotation-spreading neural network (R-SAN net). This neural net can recognize the orientation of an object irrespective of its shape, and its shape irrespective of its orientation. The R-SAN net is suitable for orientation recognition of a concentric circular pattern such as an iris image. Previously, variations of ambient lighting conditions affected iris detection. In this study, we introduce brightness normalization for accuracy improvement of iris detection in various lighting conditions. Brightness normalization provides high accuracy iris extraction in severe lighting conditions. A recognition experiment investigated the characteristics of rotation and shape recognition for both learned and un-learned iris images in various plane rotations. The R-SAN net recognized the rotation angle for the learned iris images in arbitrary orientation, but not for un-learned iris images. Thus, the variation of the rotation angle was corrected only for learned irises, but not un-learned irises. Although the R-SAN net rightly recognized the learned irises, it could not completely reject the un-learned irises as unregistered irises. Using the specific orientation recognition characteristics of the R-SAN net, a minimum distance was introduced as a new shape recognition criterion for the R-SAN net. In consequence, the R-SAN net combined with the minimum distance rightly recognized both learned (registered) and un-learned irises; the unregistered irises were correctly rejected.

Iris recognition by a rotation spreading neural network

We previously proposed a rotation spreading neural network (R-SAN net). This neural net can recognize the orientation of an object irrespective of its shape, and its shape irrespective of its orientation. The characteristic of the R-SAN net is to spread the information of axis orientation of the object on surrounding space by spreading weight that has similar tuning characteristics to axis orientation neurons in the parietal cortex. The R-SAN net is suitable for the shape and orientation recognition of a concentric circular pattern, because it uses polar conversion. Previously, the validity of this neural net for learning and recollection of human faces had been demonstrated by off-line processing. In the present study, we used iris images instead of face images for the input patterns of the neural network. We attempted real-time individual identification by the iris patterns. In the recognition experiment, the R-SAN net was able to simultaneously recognize the orientation and shape of iris images of learned persons. We suggest application of the R-SAN net as an iris recognition system.

Pupil diameter measurement in visible-light environment using separability filter

The various systems using the information of pupil diameter variation have been studied and developed. These systems are presupposed to be operated on a general-purpose PC or a portable terminal. To obtain the information of pupil diameter, a near-infrared camera and illuminations are required. However, the conventional PC or portable terminal does not have a near-infrared camera and illumination. If the pupil diameter can be measured by using the visible-light camera equipped in the general-purpose PC or the portable terminal, such systems using the information of pupil diameter variation can be widely used. It is difficult to discriminate between the iris and pupil regions in an image captured by the visible-light camera because the iris and pupil colors of Asian people are similar. In this paper, for the generalization of the systems using the information of pupil diameter variation, we proposed the measurement method of the pupil diameter from the image taken by the visible-light camera. From the experimental results, the accuracy of proposed method was equivalent to that of the conventional method of pupil diameter measurement using the near-infrared camera and the illumination.

Real-time Eye Detection Method Robust to Facial Pose Variations Using Gradient Directional Features and Particle Filter

In this paper, we propose an eye detection method that is robust to facial pose changes using gradient directional features of brightness and a particle filter. The rejection function of incorrect eye detection in the proposed method allows for the eye detection again even if the eye is incorrectly detected. In this method, to estimate the boundary between the iris and sclera or eyelid, the gradient intensities are calculated by four directional Prewitt filters in four regions. The likelihood used in the particle filter is obtained by averaging the gradient intensities for the specific direction in the four regions and the upper eyelid area. From experimental results, the average detection rates of both eyes for roll, yaw, and pitch angles of the face are more than 90% by using rejection function for incorrect eye detection. The rejection function produces the 4.4%, 4.5%, and 4.9% increases in average detection rates of both eyes for roll, yaw, and pitch facial angles, respectively. The proposed eye detection method can track both eye in real-time (about 20 ms) and is robust to the facial pose changes.