Image processing device, image processing method, and recording medium

Patent metadata

Field	Value
Publication number	US-12062162-B2
Application number	US-202017433885-A
Country	US
Kind code	B2
Filing date	Feb 26, 2020
Priority date	Feb 27, 2019
Publication date	Aug 13, 2024
Grant date	Aug 13, 2024

Abstract

An image processing device performs projection conversion that makes an image captured of an object to be recognized closer to a normal image captured from front of the object to be recognized based on a correlation between: a pre-specified plurality of feature ranges dispersed within a range of the object to be recognized; and a plurality of feature ranges designated based on the dispersion in the image.

First claim

The invention claimed is:

1. An image processing device comprising:
at least one memory configured to store instructions; and
at least one processor configured to execute the instructions to:
correct an image captured of an object to be recognized so that distortion of a specified attachment object is corrected, the specified attachment object appearing in the image and being attached to the object to be recognized; and
perform, after the correction projection conversion that makes the corrected image closer to a normal image captured from front of the object to be recognized based on a correlation between: a pre-specified plurality of feature ranges dispersed within a range of the object to be recognized; and a plurality of feature ranges designated based on the dispersion in the corrected image.

2. The image processing device according to claim 1 , wherein the at least one processor is configured to calculate a projection conversion matrix based on the correlation, and use uses the projection conversion matrix to perform the projection conversion that makes the image closer to the normal image.

3. The image processing device according to claim 1 , wherein the at least one processor is configured to specify the range of the object to be recognized based on a feature included in the object to be recognized.

4. The image processing device according to claim 1 ,
wherein the object to be recognized is an instrument, and
the feature ranges are a range having a large amount of features in the instrument.

5. The image processing device according to claim 1 , wherein the at least one processor is configured to output information indicating a plurality of ranges having a large amount of features within the range of the object to be recognized, and determine any designated ranges among the plurality of ranges as the feature ranges.

6. The image processing device according to claim 1 , wherein the at least one processor is configured to perform a recognition process of a state of the object to be recognized by using a result of the projection conversion of the image captured of the object to be recognized.

7. An image processing method comprising:
correcting an image captured of an object to be recognized so that distortion of a specified attachment object is corrected, the specified attachment object appearing in the image and being attached to the object to be recognized; and
performing, after the correction, projection conversion that makes the corrected image closer to a normal image captured from front of the object to be recognized based on a correlation between: a pre-specified plurality of feature ranges dispersed within a range of the object to be recognized; and a plurality of feature ranges designated based on the dispersion in the corrected image.

8. A non-transitory recording medium that stores a program that causes a computer to execute:
correcting an image captured of an object to be recognized so that distortion of a specified attachment object is corrected, the specified attachment object appearing in the image and being attached to the object to be recognized; and
performing, after the correction, projection conversion that makes the corrected image closer to a normal image captured from front of the object to be recognized based on a correlation between: a pre-specified plurality of feature ranges dispersed within a range of the object to be recognized; and a plurality of feature ranges designated based on the dispersion in the corrected image.

Assignees

• Nec Corp
• Central Res Inst Electric Power Ind

Inventors

• Sakai Masahito
• Itoh Norihiko

Classifications

• G06K7/1456

  determining the orientation of the optical code with respect to the reader and correcting therefore · CPC title

• G06K7/1417

  2D bar codes · CPC title

• G06V2201/034

  of medical instruments · CPC title

• G06V10/20

  Image preprocessing · CPC title

• G06V10/247

  by affine transforms, e.g. correction due to perspective effects; Quadrilaterals, e.g. trapezoids · CPC title