An identification system includes means for scanning an identity pattern in a predetermined manner and correlating previously-obtained recognition data with information derived therefrom to determine a positive identification. In obtaining recognition data, the identity pattern is scanned and digitized in a predetermined manner, and then analyzed and sorted to provide prominent information most likely to be recognized in subsequent identity verifications with the original recognition pattern. In this manner, unique recognition data for each identity pattern is stored and is available for rapid retrieval and correlation when the original identity pattern is presented.
When verification is performed using a plurality of fingers in accordance with a positional relationship of the joints of the fingers, an image of the plurality of fingers is separated into the images of the respective fingers, and comparison and collation are individually performed for the fingers. Phase data of each finger can be individually extracted, and a collation result of the fingers and phase shift data between the fingers based on phase data are generally determined. A guide member for guiding the placement of the fingers is mounted on a finger table to define positions at which the plurality of fingers are placed.
A method for analyzing stored image details for identification purposes is disclosed in which slopes are abstracted from an image to provide three-dimensional recognition information as abstractions. Data representing light levels of an image are stored in a picture memory device, which is analyzed in a predetermined manner to select absolute illumination magnitudes between fixed locations of the image. This information is directly related to the slope between the locations. Steeper slopes and their corresponding locations are stored as recognition data in a learn mode. These abstractions are compared and several are permanently stored. In an access mode, the previously-obtained information is utilized to verify with current data, and depending upon the degree of correlation therebetween, an indication of recognition is either verified or rejected. If current data is better than any of the permanently-stored data, the permanently-stored data is replaced by the current data to update the permanently-stored data.
A computer vision apparatus for automatically inspecting printed labels is disclosed. A television type camera views a label under inspection and inputs an image thereof to a pattern recognition and analysis circuit for verification. A monitor with viewing screen, joystick and keypad is connected to the pattern recognition and analysis circuit for operator supervision of the system.
There is disclosed a method of registering palm imprints of individuals to be identified to authenticate the individuals. Each palm imprint is divided into parts having regions with predetermined positions and sizes. Feature points are extracted from the parts of the imprint. The obtained feature information is registered as information for identity of individuals.
Improved method and apparatus are provided for hand dimension verification. A two dimensional scanner provides image data from top and angled scans to an image processor which processes this data to develop invariable hand measurements using reference points located from contour sequences generated from the image scan data. The reference points are located so as to be immune from variations introduced by fingernail length, hand pressure and finger spread. Finger width measurements are made using certain reference points and finger width variations are used to determine the presence of abnormalities which could be natural or introduced by wearing of jewelry. The finger width variations are in effect removed by locating finger contour points at the beginning and ending of the variations and using finger slope data for trimming and thereby provide invariable finger width measurements together with finger thickness, finger length and hand area measurements for hand dimension verification.