Information processing device, image processing method and medium

Patent metadata

Field	Value
Publication number	US-9779488-B2
Application number	US-201414779147-A
Country	US
Kind code	B2
Filing date	Mar 25, 2014
Priority date	Apr 5, 2013
Publication date	Oct 3, 2017
Grant date	Oct 3, 2017

Abstract

An information processing device according to the present invention includes: a proper identifier output unit which outputs proper identifiers for identifying learning images; a feature vector calculation unit which calculates feature vectors of at least a part of patches included in registered patches that are registered in a dictionary for compositing a restored image; and a search similarity calculation unit which calculates a similarity calculation method that classifies the proper identifiers to be given to the registered patches based on the feature vectors.

First claim

The invention claimed is:

1. An information processing device, comprising:
a central processing unit (CPU); and
a memory connected to the CPU;
wherein the CPU reads a program from the memory and achieves functions of:
outputting proper identifiers for identifying learning images;
calculating feature vectors of at least a part of patches included in registered patches that are registered in a dictionary for compositing a restored image; and
calculating a similarity calculation method that calculates similarities between patches for classifying the proper identifiers to be given to the registered patches based on the feature vectors.

2. The information processing device according to claim 1 , wherein the CPU reads the program from the memory and further achieves functions of:
receiving the learning images;
generating blurred images of the learning images;
generating patches included in the registered patches based on the learning images and the blurred images; and
registering the registered patches in the dictionary with including the similarity calculation method.

3. The information processing device according to claim 2 , wherein
the calculating the feature vectors or the generating the patches gives the proper identifiers to the registered patches.

4. The information processing device according to claim 2 , wherein
the outputting proper identifiers includes calculating the proper identifiers based on the learning images.

5. The information processing device according to claim 2 , wherein
the similarity calculation method is calculated with respect to predetermined regions in the learning images.

6. The information processing device according to claim 2 , wherein
the generating the blurred images includes selecting a blurring method to generate blurred images based on the proper identifiers.

7. The information processing device according to claim 1 , wherein the CPU reads the program from the memory and further achieves functions of:
extracting feature vectors that are used for image recognition from learning images.

8. An image processing method, comprising: outputting proper identifiers for identifying learning images;
calculating feature vectors of at least a part of patches included in registered patches that are registered in a dictionary for compositing a restored image; and
calculating a similarity calculation method that calculates similarities between patches for classifying the proper identifiers to be given to the registered patches based on the feature vectors.

9. The image processing method according to claim 8 , further comprising:
receiving the learning images;
generating blurred images of the learning images;
generating patches included in the registered patches based on the learning images and the blurred images; and
registering the registered patches in the dictionary with including the similarity calculation method.

10. The image processing method according to claim 8 , further comprising:
giving the proper identifiers to the registered patches.

11. The image processing method according to claim 8 , further comprising:
calculating the proper identifiers based on the learning images.

12. The image processing method according to claim 8 , further comprising:
calculating the similarity calculation method with respect to predetermined regions in the learning images.

13. The image processing method according to claim 8 , further comprising:
selecting a blurring method to generate blurred images based on the proper identifiers.

14. The image processing method according to claim 8 , further comprising:
extracting feature vectors that are used for image recognition from learning images.

15. A non-transitory computer-readable recording medium embodying a program, the program causing a computer device to perform a method, the method comprising:
outputting proper identifiers for identifying learning images;
calculating feature vectors of at least a part of patches included in registered patches that are registered in a dictionary for compositing a restored image; and
calculating a similarity calculation method that calculates similarities between patches for classifying the proper identifiers to be given to the registered patches based on the feature vectors.

16. The non-transitory computer-readable recording medium embodying the program causing the computer device to perform the method according to claim 15 , the method further comprising:
receiving the learning images;
generating blurred images of the learning images;
generating patches included in the registered patches based on the learning images and the blurred images; and
registering the registered patches in the dictionary with including the similarity calculation method.

17. The non-transitory computer-readable recording medium embodying the program causing the computer device to perform the method according to claim 15 , the method further comprising:
giving the proper identifiers to the registered patches.

18. The non-transitory computer-readable recording medium embodying the program causing the computer device to perform the method according to claim 15 , the method further comprising:
calculating the proper identifiers based on the learning images.

19. The non-transitory computer-readable recording medium embodying the program causing the computer device to perform the method according to claim 15 , the method further comprising:
calculating the similarity calculation method with respect to predetermined regions in the learning images.

20. The non-transitory computer-readable recording medium embodying the program causing the computer device to perform the method according to claim 15 , the method further comprising:
selecting a blurring method to generate blurred images based on the proper identifiers.

21. The non-transitory computer-readable recording medium embodying the program causing the computer device to perform the method according to claim 15 , the method further comprising:
extracting feature vectors that are used for image recognition from learning images.

22. An information processing device, comprising: proper identifier output means for outputting proper identifiers for identifying learning images;
feature vector calculation means for calculating feature vectors of at least a part of patches included in registered patches that are registered in a dictionary for compositing a restored image; and
search similarity calculation means for calculating similarities between patches for classifying a similarity calculation method that classifies the proper identifiers to be given to the registered patches based on the feature vectors.

Assignees

• Nec Corp

Inventors

• Shibata Takashi
• Iketani Akihiko
• Senda Shuji

Classifications

• G06V30/1914

  Determining representative reference patterns, e.g. averaging or distorting patterns; Generating dictionaries, e.g. user dictionaries · CPC title

• G06F18/28

  Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• G06V10/758

  Involving statistics of pixels or of feature values, e.g. histogram matching · CPC title

• G06T3/4076

  using the original low-resolution images to iteratively correct the high-resolution images · CPC title