A trainable security system emthod for the same

• US 5,091,780 A
• Filed: 05/09/1990
• Issued: 02/25/1992
• Est. Priority Date: 05/09/1990
• Status: Expired due to Fees

• Current Assignee: Carnegie Mellon University
• Original Assignee: Carnegie Mellon University
• Inventors: Pomerleau, Dean A.
• Primary Examiner(s): Kostak, Victor R.
• Application Number:
  US07/521,237
• Time in Patent Office: 657 Days
• Field of Search: 358/108, 358/105, 382/14, 382/15, 364/513
• US Class Current: 348/152
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G08B 13/19602 Image analysis to detect motion of the intruder, e.g. by frame subtraction
• CPC Class Codes: G08B 13/19656 Network used to communicate with a camera, e.g. WAN, LAN, Internet
• CPC Class Codes: G08B 13/19697 Arrangements wherein non-video detectors generate an alarm themselves
• CPC Class Codes: Y10S 706/902 Application using ai with detail of the ai system
• CPC Class Codes: Y10S 706/933 Law, law enforcement, or government

Abstract

A security system comprised of a device for monitoring an area under surveillance. The monitoring device produces images of the area. The security system is also comprised of a device for processing the images to determine whether the area is in a desired state or an undesired state. The processing device is trainable to learn the difference between the desired state and the undesired state. In a preferred embodiment, the monitoring device includes a video camera which produces video images of the area and the processing device includes a computer simulating a neural network. A method for determining whether an area under surveillance is in a desired state or an undesired state. The method comprises the steps of collecting data in a computer about the area which defines when the area is in the desired state or the undesired state. Next, training the computer from the collected data to essentially correctly identify when the area is in the desired state or in the undesired state while the area is under surveillance. Next, performing surveillance of the area with a computer such that the computer determines whether the area is in a desired state or the undesired state.

Claims

1. A security system comprised of:

means for monitoring an area under surveillance, said monitoring means producing difference images of the area;

means for processing the difference images to determine whether the area is in a desired state or an undesired state, said processing means includes a trainable neural network to learn the difference between the desired state and the undesired state.

2. A system as described in claim 1 wherein the monitoring means includes a video camera which produces video difference images of the area.

3. A system as described in claim 2 wherein the processing means includes a computer simulating the neural network.

4. A system as described in claim 3 including an alarm which is triggered by the computer when an undesired state is determined to exist in the area.

5. A system as described in claim 4 wherein the image of the area is comprised of pixels, each pixel having a value corresponding to the amount of light associated with the pixel;

and wherein said computer compares the values of pixels of a most recent image of the area with the values of pixles of an earlier in time image of the area to produce a difference image comprised of pixels, each of which have a value corresponding to the difference in values between corresponding pixels of the most recent image and the earlier in time image; and

wherein said neural network simulated by said computer having weights for each pixel network simulated by said computer having weights for each pixel which are multiplied by the respective pixel value of the difference image and then added together to form a sum, which if greater than a predetermined amount, results in the computer triggering the alarm.

6. A system as described in claim 5 wherein the monitoring means includes additional sensors, that together with the video camera produces the difference image of the area, said value of each of the pixels of the image having a component corresponding to the additional sensors as well as a component corresponding to the amount of light associated with the pixel.

7. A system as described in claim 6 wherein the additional sensors include at least a second video camera.

8. A system as described in claim 7 wherein the additional sensors include infrared motion detectors.

9. A system as described in claim 6 wherein the additional sensors include microwave detectors.

10. A system as described in claim 6 wherein the additional sensors include infrared motion detectors.

11. A system as described in claim 8 wherein the additional sensors include microwave detectors.

12. An apparatus as described in claim 1 wherein the trainable neural network uses back propagation techniques.

13. A method for determining whether an area under surveillance is in a desired state or the undesired state;

collecting difference images in a computer about the area which defines when the area is in the desired state or the undesired state;

training the computer from the difference images to essentially correcting identify when the area is in the desired state or in the undesired state while the area is under surveillance; and

performing surveillance of the area with the computer such that the computer determines whether the area is in the desired state or the undesired state.

14. A method as described in claim 13 including after the performing step, the step of activating an alarm when the computer determines the area is the undesired state.

15. A method as described in claim 13 including after the performing step, the step of repeating the training step if the area'"'"'s state determined by the computer is incorrect for a given scene of the area such that the computer does not make the same mistake about the state of the area for the given scene.