Method of intruder detection and device thereof

• US 20040239761A1
• Filed: 03/30/2004
• Published: 12/02/2004
• Est. Priority Date: 05/26/2003
• Status: Active Grant

• Current Assignee: S1 Corporation (ACI Worldwide Incorporated)
• Original Assignee: S1 Corporation (ACI Worldwide Incorporated)
• Inventors: Kuznetsov, Victor D., Murynin, Alexander B., Jeong, Il-Jun, Ivanov, Peter A., Jin, Mun Wang, Jin, Lee Yong
• Granted Patent: US 7,088,243 B2
• Time in Patent Office: Days
• US Class Current: 348/153
• CPC Class Codes: G08B 13/19604 involving reference image or background adaptation with time to compensate for changing conditions, e.g. reference image update on detection of light level change
• CPC Class Codes: G08B 13/19606 Discriminating between target movement or movement in an area of interest and other non-signicative movements, e.g. target movements induced by camera shake or movements of pets, falling leaves, rotating fan
• CPC Class Codes: G08B 13/19608 Tracking movement of a target, e.g. by detecting an object predefined as a target, using target direction and or velocity to predict its new position
• CPC Class Codes: G08B 13/19641 Multiple cameras having overlapping views on a single scene
• CPC Class Codes: G08B 13/19645 Multiple cameras, each having view on one of a plurality of scenes, e.g. multiple cameras for multi-room surveillance or for tracking an object by view hand-over
• CPC Class Codes: G08B 13/19658 Telephone systems used to communicate with a camera, e.g. PSTN, GSM, POTS
• CPC Class Codes: G08B 13/19691 Signalling events for better perception by user, e.g. indicating alarms by making display brighter, adding text, creating a sound
• CPC Class Codes: G08B 13/19693 using multiple video sources viewed on a single or compound screen
• CPC Class Codes: H04N 13/239 using two 2D image sensors having a relative position equal to or related to the interocular distance H04N13/243 takes precedence
• CPC Class Codes: H04N 2013/0074 Stereoscopic image analysis
• CPC Class Codes: H04N 7/181 for receiving images from a plurality of remote sources

Abstract

Disclosed is a method of intruder'"'"'s detection, comprising of using of more than 1 sensor, represented, for example, by video cameras that are arranged with fixed spatial orientation for stereo detector formation at that 2D distributions of light intensity that form stereo images registered by stereo detector are transmitting to blocks of processing, for example, blocks of digital processing at that mentioned blocks of digital processing perform processing of sequence of stereo images with different frequencies (high and low), to determine the presence of intruder, it'"'"'s position and speed of transference.

Claims

1. A method of intruder detection, comprising of usage of more than 1 sensor, represented by cameras that are arranged with fixed spatial orientation to form a stereo detector of 2D distributions of light intensity of a surveillance zone, that form stereo images registered by said stereo detector, transmitting the stereo images to processing blocks, for performing processing of a sequence of stereo images with different frequencies (high and low), for determining the presence of an intruder, a position of said intruder and a speed of transference of said intruder.

2. A method as defined in claim 1 wherein the processing includes the comparison of 2 or more consecutive images that are entering the processing block with high frequency, for detection of a fast moving intruder.

3. A method as defined in claim 1 wherein the processing includes measurement of 3D relief in the view field of the stereo detector, for detection of a slow moving or static intruder.

4. A method as defined in claim 1 wherein the processing includes measurement of 3D relief by the processor of detection of a static intruder in the case of detection of possible intruder by the processor of detection of a moving intruder.

5. A method as defined in claim 1 wherein the processing includes determining a local measure of difference of elements that are visible on a stereo image, and memorizing elements of the stereo image when the measured local value is more than a predetermined one in a storage device, while comparing 2 or more consequent images in the processor of detection of a moving intruder.

6. A method as defined in claim 5 wherein the processing includes performing integration of closely set elements of stereo image into geometrical figures that are stored to the storage device.

7. A method as defined in claim 6 wherein the processing includes determining a distance from stereo camera and a geometrical size for every geometrical figure, taking into account of a fixed positional relationship of sensors.

8. A method as defined in claim 6 wherein the processing includes comparing geometrical figures as with a set of geometrical figures that were memorized in advance that describe both resolved objects and a possible intruder, and the scaling of a figure in dependence of a measured distance between the stereo detector and the object that is represented by given figure and generating an alarm signal in case of coincidence with figure of possible intruder.

9. A method as defined in claim 3 wherein the processing includes searching of correspondent points in stereo image in the processor of detection of static intruder for measurement of 3D relief as in view field of stereo camera, after that determining the distance up to the elements of relief that are presented on stereo image using the known fixed positional relationship of the sensors.

10. A method as defined in claim 9 wherein the processing includes memorization of 3D relief of scene thereto memorize a series of a predetermined number of stereo images of the surveillance zone in absence of possible intruders, performing measurement of distances up to elements that are presented on the stereo image averaging the measured distances on the series of stereo images and memorizing the obtained relief in the storage device.

11. A method as defined in claim 6 wherein the processing includes performing a comparison of 3D relief that was measured elementwise with relief that was measured in advance in absence of intruder and memorized in a storage device in the processor of the detection of a static intruder, and memorizing elements of relief for whose the distinction differs from visible elements of relief that are saved to memory on a value that is more than predetermined one, as a result of comparison.

12. A method as defined in claim 11 wherein the processing includes performing the integration of closely set elements of stereo image that are saved to memory of the device into the geometrical figures.

13. A method as defined in claim 12 wherein the processing includes performing the comparison of parameters of geometrical figures that are distinguished with corresponding geometrical parameters of human body that are known in advance, and thereby making a decision about the appearance of intruder.

14. A method as defined in claim 13 wherein the processing includes performing an initial calibration that consists of saving to memory of series of stereo images of an object with known geometrical characteristics, measuring of corresponding geometrical characteristics, comparing the measured characteristics with known ones and determining of corresponding corrections used to perform the calibration again in the case of change of positional relationship of sensors of stereo detector.

15. A device for intruder detection that includes more than one signal sensor, signal processor and executive block that differs in that they install coupled video cameras as a signal sensor, the dynamic signal preprocessor that is extra inserted into the processor whose input is connected to output of the first sensor, and output is connected to input of extra inserted moving object detector and input of extra inserted first static preprocessor, at the same time the second input of the moving objects detector is connected to output of the second dynamic signal preprocessor whose input is connected to output of the second sensor whose second output is connected to input of second static preprocessor whose second input is connected to second output of second dynamic preprocessor and the third input of moving objects detector is connected to the output of control panel whose second output is connected to the first input of static decision performer whose second input is connected to output of dynamic decision performer whose input is connected to output of moving objects detector at that outputs of first and second static preprocessors are connected to the first input of extra inserted reconstructor of 3D scene whose second input is connected to extra inserted 3D objects calibrator and output is connected to extra inserted 3D objects detector whose output is connected to the third input of static decision performer whose output is connected to executive device of alarm signal creation.