Characteristic features of images of an object eye are extracted to enable non-contact detection of eye movement. Two images of the eye are focused, and a differential image is generated to eliminate background noise and to permit feature extraction to be performed. In one feature of the invention, the illuminating light is polarized for use as a reference and the reflected light is separated to two light paths, each of which is focused to form an image of the object. In one path a polarizing plate blocks the regularly reflected light from the cornea so that only a diffused reflection component of the illuminating light from the other parts of the eye is passed, while in the other path both the regularly and diffusedly reflected light components are passed. A resulting differential image emphasizes the regular reflection component from the cornea relative to the background. In another aspect of the invention, two light sources are placed at different positions relative to the optical axis, to provide bright and dark images of the pupil without otherwise affecting the reflected image. A resulting differential image emphasizes the pupil relative to background noise.
The invention concerns a method for locating, in a digital image, a circle centre comprising the following steps: a) predefining a set of potential radii of the circle; b) dimensioning (303) two accumulators to a dimension in the form of a column matrix not larger than the size of the image in x-axis and a line matrix not larger than the size of the image in y-axis; c) sequentially, for each pixel image of the image: (i) selecting successively each potential radius; (ii) evaluating the position of the potential center of a circle of the selected radius and whereof the pixel concerned is on the periphery; and (iii) incrementing accumulators at the x-axis and the y axis of the potential center; and d) selecting (304) as coordinates of the located centre, the x-axis and the y-axis corresponding to the maximum of accumulators.
A bodily state detection apparatus comprising a CCD camera, an infrared LED device, a pickup image memory, a pupil extraction circuit and a bodily state judgment circuit. The CCD camera inputs images of a predetermined area including the subject person's face. The infrared LED device illuminates the subject person in such a way that the optical axis of the camera and the direction of the illumination coincide with each other. The pickup image memory stores temporarily the output data of the CCD camera. The pupil extraction circuit extracts the subject person's pupil position from the pickup images. The bodily state judgment circuit judges the subject person's bodily state by use of the result of pupil extraction performed by the pupil extraction circuit.
The present invention provides an eye image tracking apparatus capable of tracking not only images of the eyes but also images of their periphery to accurately track the images of the eyes without erroneous detection while recognizing missing of the eyes when the eyes have been missed. The eye image tracking apparatus comprises a face image input portion 4 for inputting an image of a face, a digitalizing portion 5 for digitalizing the input image of the face, an image retrieval portion 6 for retrieving images of the eyes and their periphery within the digitalized image, and an image tracking portion 7 for tracking the retrieved images of the eyes and their periphery.
A system for capturing an image when an amount of sclera is visible in a preview image comprises a photosensor configured to detect an image, a memory configured to store at least a sclera setting, a processor configured to determine when at least one face is present in the detected image and further configured to determine an amount of sclera present in the face so that the determined amount of sclera is compared to the sclera setting, and an actuator configured to initiate capture of the detected image such that the detected image is captured when the determined amount of sclera is at least equal to the sclera setting.
A system for rapid and automatic identification of persons, with very high reliability and confidence levels. The iris of the eye is used an optical fingerprint, having a highly detailed pattern that is unique for each individual and stable over many years. Image analysis algorithms find the iris in a live video image of a person's face, and encode its texture into a compact signature, or "iris code." Iris texture is extracted from the image at multiple scales of analysis by a self-similar set of quadrature (2-D Gabor) bandpass filters defined in a dimensionless polar coordinate system. The sign of the projection of many different parts of the iris onto these multi-scale quadrature filters, determines each bit in an abstract (256-byte) iris code. The degrees-of-freedom in this code are based on the principle forms of variation in a population of irises studied. Because of the universal mathematical format and constant length of the iris codes, comparisons between them are readily implemented by the Exclusive-OR (XOR) logical operation. Pattern recognition is achieved by combining special signal processing methods with statistical decision theory, leading to a statistical test of independence based on a similarity metric (the Hamming distance) that is computed from the XOR of any two iris codes. This measure positively establishes, confirms, or disconfirms, the identity of any individual. It also generates an objective confidence level associated with any such identification decision.
