초록 ▼

A machine is disclosed that is capable of locating human faces in video scenes with random content within two minutes, and capable of recognizing the faces that it locates. The machine uses images obtained from a video camera and is insensitive to variations in brightness, scale, focus, and operates

A machine is disclosed that is capable of locating human faces in video scenes with random content within two minutes, and capable of recognizing the faces that it locates. The machine uses images obtained from a video camera and is insensitive to variations in brightness, scale, focus, and operates without any human intervention or inputs. When a motion detection feature is included, (one of its options), the location and recognition events occur in less than 1 minute. One embodiment of the system uses: a camera, a Micro-Vax computer, an analog-go-digital A/D converter, and a hard copy print out to process video scenes with random content using an original computer program to locate human faces and identify them. In operation, the camera converts the video scenes into an analog electrical signal, which is converted into digital and forwarded to the computer. The computer performs an original pattern recognition algorithm to search for facial components, identify a gestalt face, and compare the gestalt-face\s detected facial characteristics with a stored set of facial characteristics of known human faces, to identify the face thereby.

대표청구항 ▼

A face recognition video system which locates and recognizes a human face in a video scene which has an input scene size of about 512 by 480 pixels and a random background, wherein said face recognition video system identifies said human face with a set of stored characteristics which are correlated

A face recognition video system which locates and recognizes a human face in a video scene which has an input scene size of about 512 by 480 pixels and a random background, wherein said face recognition video system identifies said human face with a set of stored characteristics which are correlated with known human faces, said face recognition video system comprising: a camera which receives said video scene and outputs it in an analog electrical signal; a means for converting analog signals into digital signals, said converting means receiving and converting said analog electrical signal from said camera into a digital electrical signal; a means for applying an elliptical mask to said video scene around a location of said human face for recognizing and identifying human faces, said applying means receiving said digital electrical signal from said converting means and locating said human face within that video scene by searching for and extracting detected facial characteristics within said video scene, said applying means then applying said elliptical mask around the location of the human face and defining a detected gestalt-face within said video scene with a minimum facial size of 64 by 64 pixels by identifying said detected facial characteristics using the detected facial characteristics of whole human faces by comparing said detected facial characteristics of said detected gestalt face with said set of stored face characteristics and outputting a name associated with the stored face characteristics which successfully correlates with said detected gestalt face, wherein said applying means comprises a Micro Vax computer which has a memory which has said set of stored face characteristics stored therein along with a set of names of said known human faces, said Micro Vax computer being electrically connected with said converting means to receive said digital electrical signal thereby, said Micro Vax computer searching said video scene for detected facial characteristics beginning with a search for a pair of eyes in said video scene, said Micro Vax computer outputting a signal to indicate that there are no human faces in said video scene when no pair of eyes are detected, said Micro Vax computer continuing to look for said facial characteristics once a human face is detected until it has sufficient detected facial characteristics to form said detected gestalt face which may be compared with said set of stored face characteristics and to identify said human face in said video scene thereby; and and a means for displaying said name produced by said applying means, said displaying means being electrically connected with said applying means to receive said name therefrom.