Tailgating and reverse entry detection, alarm, recording and prevention using machine vision
First Claim
1. A method for detecting a violation of security of a controlled access area, comprising:
- (a) acquiring one or more stereoscopic images of an area of observation, wherein objects in the area of observation are characterized by (1) an object type and (2) an authorization status indicating whether the object is authorized with respect to the controlled access area;
(b) analyzing the one or more images using a machine vision processing system to identify a first object in the area of observation and to classify the first object in a first object type among a plurality of object types that are pre-defined in the machine vision processing system, said first object having supplied an authorization with respect to the controlled access area, and further using the machine vision processing system to identify a second object in the area of observation and to classify the second object in a second object type among the plurality of pre-defined object types, wherein the authorization status of each of the objects is determined separately from classifying the object in an object type; and
(c) applying one or more access control rules to the information obtained from the image analysis to determine whether the second object is attempting to breach the controlled access area by utilizing the authorization supplied by the first object in violation thereof,wherein the controlled access area is limited to objects that are classified in a defined object type and have a status that is authorized with respect to the controlled access area, andwherein the one or more access control rules determine whether the second object is attempting to breach the controlled access area by separately determining (1) whether the classification of the second object is in the defined object type and (2) whether the second object has a status that is authorized with respect to the controlled access area.
1 Assignment
0 Petitions
Accused Products
Abstract
Unauthorized entry into controlled access areas using tailgating or reverse entry methods is detected using machine vision methods. Camera images of the controlled area are processed to identify and track objects in the controlled area. In a preferred embodiment, this processing includes 3D surface analysis to distinguish and classify objects in the field of view. Feature extraction, color analysis, and pattern recognition may also be used for identification and tracking of objects. Integration with security monitoring and control systems provides notification when a tailgating or reverse entry event has occurred. More reliable operation in practical circumstances is thus obtained, such as when multiple people are using an entrance or exit under variable light and shadow conditions. Electronic access control systems may further be combined with the machine vision methods of the invention to more effectively prevent tailgating or reverse entry.
106 Citations
54 Claims
-
1. A method for detecting a violation of security of a controlled access area, comprising:
-
(a) acquiring one or more stereoscopic images of an area of observation, wherein objects in the area of observation are characterized by (1) an object type and (2) an authorization status indicating whether the object is authorized with respect to the controlled access area; (b) analyzing the one or more images using a machine vision processing system to identify a first object in the area of observation and to classify the first object in a first object type among a plurality of object types that are pre-defined in the machine vision processing system, said first object having supplied an authorization with respect to the controlled access area, and further using the machine vision processing system to identify a second object in the area of observation and to classify the second object in a second object type among the plurality of pre-defined object types, wherein the authorization status of each of the objects is determined separately from classifying the object in an object type; and (c) applying one or more access control rules to the information obtained from the image analysis to determine whether the second object is attempting to breach the controlled access area by utilizing the authorization supplied by the first object in violation thereof, wherein the controlled access area is limited to objects that are classified in a defined object type and have a status that is authorized with respect to the controlled access area, and wherein the one or more access control rules determine whether the second object is attempting to breach the controlled access area by separately determining (1) whether the classification of the second object is in the defined object type and (2) whether the second object has a status that is authorized with respect to the controlled access area. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 49, 50, 51, 52, 53, 54)
-
-
34. An apparatus for detecting a security violation with respect to a controlled access area, comprising:
-
(a) a stereoscopic imaging device for acquiring one or more stereoscopic images of an area of observation, wherein objects in the area of observation are characterized by (1) an object type and (2) an authorization status indicating whether the object is authorized with respect to the controlled access area; and (b) a machine vision processing system in communication with the stereoscopic imaging device to receive the one or more images, the machine vision processing system being further configured to analyze the one or more images to identify an object in the area of observation and to classify the object in an object type among a plurality of object types that are pre-defined in the machine vision processing system, and apply one or more access control rules to determine whether the object is attempting to breach the controlled access area in violation of security of the controlled access area, wherein the controlled access area is limited to objects that are classified in a defined object type and have a status that is authorized with respect to the controlled access area, and wherein the one or more access control rules applied by the machine vision processing system determine whether the object is attempting to breach the controlled access area by separately determining (1) whether the classification of the object is in the defined object type and (2) whether the second object has a status that is authorized with respect to the controlled access area. - View Dependent Claims (35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48)
-
Specification