Object monitoring system, object monitoring method, and monitoring target extraction program

• US 9,811,755 B2
• Filed: 02/25/2014
• Issued: 11/07/2017
• Est. Priority Date: 03/29/2013
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Miyano, Hiroyoshi
• Primary Examiner(s): Tabatabai, Abolfazl
• Application Number:
  US14/780,884
• Publication Number:
  US 20160042243A1
• Time in Patent Office: 1,351 Days
• Field of Search: 382103, 382104, 382107, 382115-118, 382236, 348149, 348164, 348169-172, 348352, 348180, 348184, 700 91, 700108, 700144, 700174
• 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: G06V 40/166 using acquisition arrangements
• CPC Class Codes: G06V 40/23 Recognition of whole body movements, e.g. for sport training
• CPC Class Codes: H04N 7/18 Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
• CPC Class Codes: H04N 7/181 for receiving images from a plurality of remote sources

Abstract

First imaging means 60 creates a state image by capturing a state occurring in a first range. Second imaging means 70 creates a monitoring target image by capturing a monitoring target present in a second range. Specific state detection means 81 detects a specific state from the state image. Monitoring target analysis means 82 analyzes a status of a monitoring target, using the state image from which the specific state is detected. Monitoring target feature extraction means 85 extracts, from the monitoring target image, a feature used for specifying the monitoring target, based on the status of the monitoring target.

Claims

1. An object monitoring system comprising:

a memory storing instructions; and

at least one processor configured to process the instructions to;

create a state image by capturing a state occurring in a first range,

create a monitoring target image by capturing a monitoring target present in a second range,

detect a specific state from the state image,

extract, from the monitoring target image created after the specific state has been detected, a feature used for specifying the monitoring target, based on the status of the monitoring target, and

specify, for each type of specific state, as an object, a monitoring target which appears with a high frequency in a plurality of monitoring target images.

2. The object monitoring system according to claim 1, wherein the at least one processor is further configured to process the instructions to:

predict a time at which the monitoring target appears in the second range, from the status of the monitoring target,

create the monitoring target image in a period determined based on the predicted time, and

extract the feature used for specifying the monitoring target, from the monitoring target image.

3. The object monitoring system according to claim 2, wherein the at least one processor is further configured to process the instructions to:

analyze a movement status of the monitoring target, and

predict the time at which the monitoring target appears in the second range, based on the movement status.

4. The object monitoring system according to claim 3, wherein the at least one processor is further configured to process the instructions to:

store a past state image created in the past, and

analyze the movement status of the monitoring target, using the state image from which the specific state is detected and the past state image captured within a predetermined period before the state image is captured.

5. The object monitoring system according to claim 1, wherein the at least one processor is further configured to process the instructions to:

analyze an appearance of the monitoring target, and

narrow down the monitoring target whose feature is to be extracted, based on the analyzed appearance.

6. The object monitoring system according to claim 1, wherein the at least one processor is further configured to process the instructions to:

create the monitoring target image by capturing a person as the monitoring target, and

extract a face feature of the person from the monitoring target image.

7. The object monitoring system according to claim 1, wherein the at least one processor is further configured to process the instructions to:

compare, for each type of specific state, features of monitoring targets extracted from the plurality of monitoring target images to estimate the monitoring target which appears with a high frequency in the plurality of monitoring target images.

8. The object monitoring system according to claim 7, wherein the at least one processor is further configured to process the instructions to:

specify the monitoring target for which a feature estimated to be identical is extracted at least a predetermined number of times, as the object.

9. The object monitoring system according to claim 7, wherein the at least one processor is further configured to process the instructions to:

calculate a likelihood indicating a degree of probability of the monitoring target, together with the feature of the monitoring target, and

specify the monitoring target using the likelihood.

10. A monitoring target extraction device comprising:

a memory storing instructions; and

at least one processor configured to process the instructions to;

detect a specific state from a state image captured by a first imaging device which captures a state occurring in a first range,

extract, from a monitoring target image captured by a second imaging device which captures a monitoring target present in a second range after the specific state has been detected, a feature used for specifying the monitoring target, based on the status of the monitoring target, and

specify, for each type of specific state, as an object, a monitoring target which appears with a high frequency in a plurality of monitoring target images.

11. The monitoring target extraction device according to claim 10, wherein the at least one processor is further configured to process the instructions to:

predict a time at which the monitoring target appears in the second range, from the status of the monitoring target, and

extract the feature used for specifying the monitoring target, from the monitoring target image in a period determined based on the predicted time.

12. An object monitoring method, implemented by at least one processor, the object monitoring method comprising:

creating a state image by capturing a state occurring in a first range;

detecting a specific state from the state image;

creating a monitoring target image by capturing a monitoring target present in a second range, based on the status of the monitoring target;

extracting, from the monitoring target image created after the specific state has been detected, a feature used for specifying the monitoring target; and

specifying, for each type of specific state, as an object, a monitoring target which appears with a high frequency in a plurality of monitoring target images.

13. The object monitoring method according to claim 12, comprising:

predicting a time at which the monitoring target appears in the second range, from the status of the monitoring target;

creating the monitoring target image by capturing the monitoring target present in the second range in a period determined based on the predicted time; and

extracting the feature used for specifying the monitoring target, from the monitoring target image.

14. A monitoring target extraction method, implemented by at least one processor, the monitoring target extraction method comprising:

detecting a specific state from a state image capturing a state occurring in a first range;

extracting, from a monitoring target image capturing a monitoring target present in a second range, after the specific state has been detected, a feature used for specifying the monitoring target, based on the status of the monitoring target; and

specifying, for each type of specific state, as an object, a monitoring target which appears with a high frequency in a plurality of monitoring target images.

15. The monitoring target extraction method according to claim 14, comprising:

predicting a time at which the monitoring target appears in the second range, from the status of the monitoring target; and

extracting the feature used for specifying the monitoring target, from the monitoring target image in a period determined based on the predicted time.

16. A non-transitory computer readable information recording medium storing a monitoring target extraction program, when executed by a processor, performs a method for:

detecting a specific state from a state image capturing a state occurring in a first range;

extracting, from a monitoring target image capturing a monitoring target present in a second range, after the specific state has been detected, a feature used for specifying the monitoring target, based on the status of the monitoring target; and

specifying, for each type of specific state, as an object, a monitoring target which appears with a high frequency in a plurality of monitoring target images.

17. The non-transitory computer readable information recording medium according to claim 16, wherein the method is further for:

predicting a time at which the monitoring target appears in the second range, from the status of the monitoring target; and

extracting the feature used for specifying the monitoring target, from the monitoring target image in a period determined based on the predicted time.