Image processing device is detecting a first repeated pattern from image and extracting object from the first repeated pattern to output second repeated pattern, image processing method, and computer program product

Patent metadata

Field	Value
Publication number	US-12417614-B2
Application number	US-202217932646-A
Country	US
Kind code	B2
Filing date	Sep 15, 2022
Priority date	Mar 23, 2020
Publication date	Sep 16, 2025
Grant date	Sep 16, 2025

Abstract

According to an embodiment, an image processing device includes one or more processors. The one or more processors are configured to: acquire an image; detect a first repeated pattern from the image; detect an object included in the first repeated pattern; and output the object as a second repeated pattern.

First claim

What is claimed is:

1. An image processing device comprising:
one or more processors configured to:
acquire an image;
detect a first repeated pattern from the image;
divide the first repeated pattern into a plurality of partial images;
execute image processing on a first partial image included in the plurality of partial images, and output a processed image;
perform matching between feature points of a second partial image included in the plurality of partial images and feature points of the processed image, and detect similar feature points as corresponding points;
determine whether or not the first partial image and the second partial image are repeated patterns based on relative positional relationships between the corresponding points and the corresponding feature points;
detect the first partial image and the second partial image determined as repeated patterns, as an object included in the first repeated pattern; and
output the object as a second repeated pattern.

2. The image processing device according to claim 1 , wherein
the one or more processors are configured to divide the first repeated pattern into a plurality of partial images by changing the number of divisions when there is no second partial image determined as a repeated pattern.

3. The image processing device according to claim 1 , wherein
each of the feature points is represented by a feature including a value indicating a scale, and
the one or more processors are configured to determine a scale ratio of the partial image to be divided, in accordance with a ratio between the scale included in the feature representing the feature point and a scale included in a feature representing the corresponding point.

4. The image processing device according to claim 1 , wherein
the image processing includes at least one of processing of vertically inverting the first partial image, processing of horizontally inverting the first partial image, and processing of rotating the first partial image.

5. The image processing device according to claim 4 , wherein
the image processing includes processing of rotating the first partial image, and
the one or more processors are configured to determine a rotation angle based on the relative positional relationships between the feature points and the corresponding points, and to rotate the first partial image at the determined rotation angle.

6. The image processing device according to claim 1 , wherein
one or more processors are configured to:
correct a positional shift between the first partial image and the second partial image determined as repeated patterns; and
detect the first partial image and the second partial image whose positional shift has been corrected as the object included in the first repeated pattern.

7. An image processing device comprising:
one or more processors configured to:
acquire an image;
detect a first repeated pattern from the image;
detect a plurality of objects from the image;
classify objects similar to each other among the plurality of objects into a group;
extract the objects classified into the group from the first repeated pattern; and
output the extracted objects as a second repeated pattern that is a more detailed repeated pattern than the first repeated pattern.

8. An image processing method comprising:
acquiring an image;
detecting a first repeated pattern from the image;
dividing the first repeated pattern into a plurality of partial images;
executing image processing on a first partial image included in the plurality of partial images, and outputting a processed image;
performing matching between feature points of a second partial image included in the plurality of partial images and feature points of the processed image, and detecting similar feature points as corresponding points;
determining whether or not the first partial image and the second partial image are repeated patterns based on relative positional relationships between the corresponding points and the corresponding feature points;
detecting the first partial image and the second partial image determined as repeated patterns, as an object included in the first repeated pattern; and
outputting the object as a second repeated pattern.

9. A computer program product comprising a non-transitory computer-readable storage medium including programmed instructions, the instructions causing a computer to execute:
acquiring an image;
detecting a first repeated pattern from the image;
dividing the first repeated pattern into a plurality of partial images;
executing image processing on a first partial image included in the plurality of partial images, and outputting a processed image;
performing matching between feature points of a second partial image included in the plurality of partial images and feature points of the processed image, and detecting similar feature points as corresponding points;
determining whether or not the first partial image and the second partial image are repeated patterns based on relative positional relationships between the corresponding points and the corresponding feature points;
detecting the first partial image and the second partial image determined as repeated patterns, as an object included in the first repeated pattern; and
outputting the object as a second repeated pattern.

10. An image processing device comprising:
one or more processors configured to:
acquire an image;
extract a plurality of feature points from the image;
perform matching for each of the plurality of feature points with other feature points to detect similar feature points as corresponding points; and
detect a repeated pattern included in the image based on relative positional relationships between the corresponding points and the corresponding feature points,
wherein
each of the feature points is represented by a feature including a value indicating a direction, and
the one or more processors are configured to perform the matching between feature points among the plurality of feature points having a difference smaller than a threshold, the difference being a difference between the values indicating the directions.

11. An image processing method comprising:
acquiring an image;
extracting a plurality of feature points from the image;
performing matching for each of the plurality of feature points with other feature points to detect similar feature points as corresponding points; and
detecting a repeated pattern included in the image based on relative positional relationships between the corresponding points and the corresponding feature points,
wherein
each of the feature points is represented by a feature including a value indicating a direction, and
the matching includes matching between feature points among the plurality of feature points having a difference smaller than a threshold, the difference being a difference between the values indicating the directions.

12. A computer program product comprising a non-transitory computer-readable storage medium including programmed instructions, the instructions causing a computer to execute:
acquiring an image;
extracting a plurality of feature points from the image;
performing matching for each of the plurality of feature points with other feature points to detect similar feature points as corresponding points; and
detecting a repeated pattern included in the image based on relative positional relationships between the corresponding points and the corresponding feature points,
wherein
each of the feature points is represented by a feature including a value indicating a direction, and
the matching includes matching between feature points amo

Assignees

• Toshiba Kk
• Toshiba Digital Solutions Corp

Inventors

• Kiyama Ryou

Classifications

• G06V10/48

  by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation · CPC title

• G06V10/764

  using classification, e.g. of video objects · CPC title

• G06V10/7515

  Shifting the patterns to accommodate for positional errors · CPC title

• G06V10/759

  Region-based matching · CPC title

• G06V10/761

  Proximity, similarity or dissimilarity measures · CPC title