Image processing device for tracking a target object and image processing method, computer program product, and image processing system for same

• US 10,075,652 B2
• Filed: 03/21/2016
• Issued: 09/11/2018
• Est. Priority Date: 03/23/2015
• Status: Active Grant

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Okumura, Mayu, Watanabe, Tomoki, Tasaki, Tsuyoshi, Takatani, Tsuyoshi
• Primary Examiner(s): Thirugnanam, Gandhi
• Application Number:
  US15/075,309
• Publication Number:
  US 20160284098A1
• Time in Patent Office: 904 Days
• Field of Search: 382103
• CPC Class Codes: G06T 7/292 Multi-camera tracking
• CPC Class Codes: G06T 7/41 based on statistical description of texture
• CPC Class Codes: H04N 23/90 Arrangement of cameras or camera modules, e.g. multiple cameras in TV studios or sports stadiums

Abstract

According to an embodiment, an image processing device includes an image acquirer, a tracker, an extractor, a feature amount acquirer, and an association controller. The image acquirer is configured to acquire a time series of images from each imaging device. The tracker is configured to track a position of an object in the time series of images. The extractor is configured to extract, from the time series of images, an observed value corresponding to a feature value in each of regions into which the time series of images is divided depending on a distance from the imaging device. The feature amount acquirer is configured to acquire a feature amount based on a deviation value of the observed value in a feature value distribution for the region. The association controller is configured to associate the tracked objects among multiple time series of images acquired from the imaging devices by using the feature amount.

Claims

1. An image processing device comprising:

storage; and

a processor configured to at least;

acquire a time series of images from each of a plurality of imaging devices;

track position of a target object in each time series of images;

control storing, in the storage, distributions of feature values for respective regions into which each image of a respective time series of images is divided based on an imaging range of an imaging device corresponding to the respective time series of images;

extract observed values corresponding to feature values in the regions of images in respective time series of images;

acquire feature amounts based on deviation values of the observed values in the distributions of feature values; and

associate tracked target objects with one another among multiple time series of images acquired from the plurality of imaging devices based on the feature amounts.

2. The device according to claim 1, wherein the processor is configured to acquire, as a feature amount, a vector of the deviation value of each of observed values corresponding to a plurality of kinds of feature values.

3. The device according to claim 1, wherein the storage further stores the distribution of the feature value for each of the regions into which each image of the time series of images is divided according to change in a moving direction of the target object in the time series of images.

4. The device according to claim 1, wherein the processor is configured to update the distributions stored in the storage according to the observed values.

5. The device according to claim 4, wherein

the storage pre-stores initial distribution, and

the processor is configured to shift the initial distribution according to the observed value extracted in an initial state.

6. The device according to claim 1, wherein the feature values include at least one of height, aspect ratio, moving speed, operation cycle, basic operation interval, and color information of the target object.

7. The device according to claim 1, wherein imaging devices of the plurality of imaging devices have imaging ranges different from one another.

8. An image processing method comprising:

acquiring a time series of images from each of a plurality of imaging devices;

tracking position of a target object in each time series of images;

storing, in a storage, distributions of feature values for respective regions into which each image of a respective time series of images is divided based on an imaging range of an imaging device corresponding to the respective time series of images;

extracting observed values corresponding to feature values in the regions of images in respective time series of images;

acquiring feature amounts based on deviation values of the observed values in the distributions of feature values; and

associating tracked target objects with one another among multiple time series of images acquired from the plurality of imaging devices based on the feature amounts.

9. A computer program product comprising a non-transitory computer-readable medium containing a program that, when executed by a computer, causes the computer to execute:

acquiring a time series of images from each of a plurality of imaging devices;

tracking position of a target object in the time series of images;

storing, in a storage, distributions of feature values for respective regions into which each image of a respective time series of images is divided based on an imaging range of an imaging device corresponding to the respective times series of images;

extracting observed values corresponding to feature values in the regions of images in respective time series of images;

acquiring feature amounts based on deviation values of the observed values in the distributions of feature values; and

associating tracked target objects with one another among multiple time series of images acquired from the plurality of imaging devices based on the feature amounts.

10. An image processing system comprising:

a plurality of imaging devices;

storage; and

a processor configured to at least;

acquire a time series of images from each of the plurality of imaging devices;

track position of a target object in the time series of images;

control storing, in the storage, distributions of feature values for respective regions into which each image of a respective time series of images is divided based on an imaging range of an imaging device corresponding to the respective time series of images;

extract observed values corresponding to feature values in the regions of images in respective time series of images;

acquire feature amounts based on deviation values of the observed values in the distributions of feature values;

associate tracked target objects with one another among multiple time series of images acquired from the plurality of imaging devices based on the feature amounts; and

present information indicating the target objects associated with one another among the multiple time series of images.