OBJECT RECOGNITION DEVICE, OBJECT RECOGNITION METHOD, AND PROGRAM

• US 20170351934A1
• Filed: 08/24/2017
• Published: 12/07/2017
• Est. Priority Date: 03/05/2015
• Status: Active Grant

• Current Assignee: Omron Corporation
• Original Assignee: Omron Corporation
• Inventors: RUAN, Xiang, YASUDA, Naru, LU, Yanping
• Granted Patent: US 10,599,709 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/50 of still image data
• CPC Class Codes: G06F 16/58 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/583 using metadata automatically derived from the content
• CPC Class Codes: G06F 16/5866 using information manually generated, e.g. tags, keywords, comments, manually generated location and time information
• CPC Class Codes: G06F 18/28 Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries
• CPC Class Codes: G06V 2201/10 Recognition assisted with metadata

Abstract

An object recognition device includes an acquisition unit configured to acquire a recognition target image that serves as an object to be recognized; a retrieval unit configured to search an image database storing a plurality of image data in association with tag information and retrieve a similar image that matches the recognition target image; and a recognition unit configured to recognize the object included in the recognition target image on the basis of tag information associated with a similar image obtained by the retrieval unit. The recognition may select the tag information that appears most frequently among the tag information associated with the similar images as a recognition result. The recognition unit may also compute a tag information reliability score from the similar image in the retrieval result and recognize an object taking into account said reliability score.

Claims

1. An object recognition device comprising:

an acquisition unit configured to acquire a recognition target image that serves as an object to be recognized;

a retrieval unit configured to search an image database storing a plurality of image data in association with tag information and retrieve a similar image that matches the recognition target image; and

a recognition unit configured to recognize the object included in the recognition target image on the basis of tag information associated with a similar image obtained by the retrieval unit.

2. The object recognition device according to claim 1, wherein the recognition unit recognizes the object included in the recognition target image on the basis of the frequency of tag information associated with a similar image.

3. The object recognition device according to claim 2, wherein the recognition unit selects the tag information that appears most frequently among the tag information associated with the similar images as the recognition result for the object included in the recognition target image.

4. The object recognition device according to claim 2, wherein the recognition unit selects the tag information with a high predetermined frequency of appearances among the tag information associated with the similar images as the recognition result for the object included in the recognition target image.

5. The object recognition device according to claim 1, wherein the recognition unit performs natural language processing to unify the variety of meanings among the tag information associated with the similar images, and after natural language processing recognize the object included in the recognition target image on the basis of the processed tag information.

6. The object recognition device according to claim 1, wherein the recognition unit computes a tag information reliability score which indicates the reliability of tag information associated with a similar image and recognizes an object included in the recognition target image taking into account said tag information reliability score.

7. The object recognition device according to claim 6, wherein the recognition unit computes the tag information reliability score on the basis of a correlation between a plurality of tag information when a single similar image includes a plurality of tag information.

8. The object recognition device according to claim 7, wherein a tag information correlation database is configured to store a correlation between a plurality of tags, and the recognition unit queries the tag information correlation database regarding correlations between the plurality of tag information assigned to the single image and increases the tag information reliability score as the correlation obtained increases.

9. The object recognition device according to claim 6, wherein the recognition unit computes the tag information reliability score on the basis of meta-information and the tag information assigned to the similar images.

10. The object recognition device according to claim 6, wherein the recognition unit recognizes the object included in the recognition target image on the basis of a weighted frequency for the tag information with the tag information reliability score as a weight coefficient.

11. The object recognition device according to claim 1, wherein the acquisition unit outputs a local region from the input image as the recognition target image.

12. The object recognition device according to claim 11, further comprising:

a relevance score computer configured to compute a relevance score for the local region; and

the recognition unit configured to output a local region with a relevance score greater than or equal to a threshold as a recognition result.

13. An object recognition method for execution on a computer, the object recognition method comprising:

acquiring a recognition target image that serves as an object to be recognized;

searching an image database storing a plurality of image data in association with tag information and retrieving a similar image that matches the recognition target image; and

recognizing the object included in the recognition target image on the basis of tag information associated with the retrieved similar image.

14. A non-transitory computer-readable recording medium storing a program causing a computer to perform operations comprising a method according to claim 13.