Abnormality detection apparatus, abnormality detection method, and recording medium storing abnormality detection program

Patent metadata

Field	Value
Publication number	US-9875409-B2
Application number	US-201514815786-A
Country	US
Kind code	B2
Filing date	Jul 31, 2015
Priority date	Aug 1, 2014
Publication date	Jan 23, 2018
Grant date	Jan 23, 2018

Abstract

An abnormality detection apparatus, an abnormality detection method, and an abnormality detection program are provided. Each of the abnormality detection apparatus, an abnormality detection method, and an abnormality detection program extracts a target image to be monitored and a reference image, respectively, from target video to be monitored, detects an abnormality based on a difference between the target image to be monitored and the reference image, and displays an image indicating a difference between the target image to be monitored and the reference image on a monitor. Moreover, an abnormality detection system is provided including the abnormality detection apparatus, a video that captures an image of a target to be monitored, and a monitor.

First claim

What is claimed is:

1. An abnormality detection apparatus comprising:
a still image extraction unit configured to process video data of a reference video that records an object for a first period of time, to determine a first interval within the first period of time for which the object is stationary in the reference video, to extract a reference image that illustrates the object during the first interval, to process video data of a target video that records the object for a second period of time that is different from the first period of time, to determine a second interval within the second period of time for which the object is stationary in the target video, and to extract a target image that illustrates the object during the second interval;
an abnormality detector configured to detect an abnormality based on a difference between the target image and the reference image; and
a display controller configured to, in response to detection of the abnormality, display the target image and the reference image on a monitor.

2. The abnormality detection apparatus according to claim 1 , wherein the apparatus further comprises a unit configured to classify images from the first interval into one of an object image to be used for the abnormality detection and a nontarget image not to be used for the abnormality detection.

3. The abnormality detection apparatus according to claim 2 , wherein the abnormality detector performs abnormality detection based on the object image and the target image only when a difference between the nontarget image and the target image is equal to or greater than a prescribed threshold.

4. The abnormality detection apparatus according to claim 1 , further comprising: a masked image generator configured to generate a masked image where masking processes are performed on an area other than an object area to be detected in the reference image, and to combine the masked image with each one of the reference image and the target image.

5. The abnormality detection apparatus according to claim 3 , wherein the abnormality detector replaces the reference image with the target image when a difference between the nontarget image and the target image is equal to or greater than the prescribed threshold.

6. The abnormality detection apparatus according to claim 1 , wherein the abnormality detector detects an abnormality in the target image or replaces the reference image according to a level of a difference between the target image and the reference image.

7. A method of detecting an abnormality, the method comprising:
processing video data of a reference video that records an object for a first period of time;
determining a first interval within the first period of time for which the object is stationary in the reference video;
extracting a reference image that illustrates the object during the first interval;
processing video data of a target video that records the object for a second period of time that is different from the first period of time;
determining a second interval within the second period of time for which the object is stationary in the target video;
extracting a target image that illustrates the object during the second interval;
detecting an abnormality based on a difference between the target image and the reference image; and
in response to detecting the abnormality, displaying the target image and the reference image on a monitor.

8. The method according to claim 7 , wherein the method further comprises classifying images from the first interval into one of an object image to be used for the abnormality detection and a nontarget image not to be used for the abnormality detection.

9. The method according to claim 8 , wherein the detecting includes performing abnormality detection based on the object image and the target image only when a difference between the nontarget image and the target image is equal to or greater than a prescribed threshold.

10. The method according to claim 7 , further comprising: generating a masked image where masking processes are performed on an area other than an object area to be detected in the reference image, and combining the masked image with each one of the reference image and the target image.

11. The method according to claim 9 , wherein the detecting includes replacing the reference image with the target image when a difference between the nontarget image and the target image is equal to or greater than the prescribed threshold.

12. The method according to claim 7 , wherein the abnormality detector detects an abnormality in the target image or replaces the reference image according to a level of a difference between the target image and the reference image.

13. A computer-readable non-transitory recording medium storing a program for causing a computer to execute an image processing method, the method comprising:
processing video data of a reference video that records an object for a first period of time;
determining a first interval within the first period of time for which the object is stationary in the reference video;
extracting a reference image that illustrates the object during the first interval;
processing video data of a target video that records the object for a second period of time that is different from the first period of time;
determining a second interval within the second period of time for which the object is stationary in the target video;
extracting a target image that illustrates the object during the second interval;
detecting an abnormality based on a difference between the target image and the reference image; and
in response to detecting the abnormality, displaying the target image and the reference image on a monitor.

14. An abnormality detection system comprising:
the abnormality detection apparatus according to claim 1 ; and
a monitor.

15. The abnormality detection apparatus according to claim 1 , wherein
the display controller displays the target image and the reference image in a row.

16. The abnormality detection apparatus according to claim 1 , wherein
the display controller displays an image indicating a difference between the target image and the reference image.

Assignees

• Ohmura Keiji
• Ohtsuka Aiko
• Takahashi Kazuya
• Ricoh Co Ltd

Inventors

• Ohmura Keiji
• Ohtsuka Aiko
• Takahashi Kazuya

Classifications

• G06V10/7788

  the supervisor being a human, e.g. interactive learning with a human teacher · CPC title

• G06F18/41

  Interactive pattern learning with a human teacher · CPC title

• G06T2207/30232

  Surveillance · CPC title

• G06K9/00771 Primary

  Physics · mapped topic

• G08B13/19613

  Recognition of a predetermined image pattern or behaviour pattern indicating theft or intrusion · CPC title