Image processing apparatus, image processing method, and image system

Patent metadata

Field	Value
Publication number	US-9609212-B2
Application number	US-201514925702-A
Country	US
Kind code	B2
Filing date	Oct 28, 2015
Priority date	Aug 28, 2013
Publication date	Mar 28, 2017
Grant date	Mar 28, 2017

Abstract

An image processing apparatus detecting a plurality of joint positions between a plurality of input images includes a target image generation unit configured to generate a plurality of target images to be searched in a second input image among the input images from a first input image among the input images; a characteristic amount calculation unit configured to calculate a characteristic amount for each of the target images generated by the target image generation unit; and a matching calculation unit configured to calculate a matching between the target image of interest among the target images having the characteristic amounts calculated by the characteristic amount calculation unit, and a part of the second input image, using an evaluation method depending on the characteristic amounts calculated by the characteristic amount calculation unit.

First claim

What is claimed is:

1. An image processing apparatus detecting a plurality of joint positions between input images, comprising:
circuitry configured to:
generate target images to be searched in a second input image based on a first input image, the first input image and the second input image being among the input images;
calculate a characteristic amount for each of the target images; and
calculate a matching between a particular target image of interest among the target images, and a part of the second input image, using an evaluation method depending on the calculated characteristic amounts, wherein the characteristic amount is at least one of an edge amount, dispersion and standard deviation of the respective target image.

2. The image processing apparatus as claimed in claim 1 , wherein the circuitry is further configured to:
correct an evaluation value at each of positions calculated by the matching using a correction method depending on the evaluation method used.

3. The image processing apparatus as claimed in claim 2 , wherein
the circuitry is further configured to:
calculate a temporary position for the particular target image of interest based on a relative relationship of the characteristic amounts between the particular target image of interest and one or more of adjacent target images among the target images, and

in the correction, the circuitry is configured to correct the evaluation value based on similarity calculated so that the evaluation value has a greater value centered around the temporary position.

4. The image processing apparatus as claimed in claim 2 , wherein the circuitry is further configured to calculate a temporary position for the particular target image of interest based on a relative relationship of the characteristic amounts between the particular target image of interest and one or more of adjacent target images among the target images.

5. The image processing apparatus as claimed in claim 1 , wherein, as the evaluation method depending on the characteristic amounts, an evaluation method giving a correlation value is used when the characteristic amount is greater than a reference value, or another evaluation method giving an error difference value is used when the characteristic amount is less than the reference value.

6. The image processing apparatus as claimed in claim 1 , wherein the circuitry is further configured to:
determine a joint position with the second input image for the particular target image of interest, depending on joint positions of adjacent target images having the characteristic amounts greater than that of the particular target image of interest, based on evaluation values calculated by the matching.

7. The image processing apparatus as claimed in claim 6 , wherein the circuitry is further configured to:
convert a plurality of captured images onto a coordinate system different from a coordinate system used when capturing the captured images, using conversion data to project an overlapped region where captured ranges of the captured images are overlapped, onto a neighborhood of an equator of a spherical surface based on a projection model, to generate the input images;
correct the conversion data based on a result of the determination; and
apply a rotational coordinate transform to the corrected conversion data, to generate another conversion data for image synthesis specifying conversion to be applied to the captured images for image synthesis.

8. The image processing apparatus as claimed in claim 6 , wherein the circuitry is further configured to convert a plurality of captured images onto a coordinate system different from a coordinate system used when capturing the captured images, using conversion data to project an overlapped region where captured ranges of the captured images are overlapped, onto a neighborhood of an equator of a spherical surface based on a projection model, to generate the input images.

9. The image processing apparatus as claimed in claim 8 , wherein the circuitry is further configured to correct the conversion data based on a result of the determination.

10. The image processing apparatus as claimed in claim 1 , wherein the characteristic amount corresponds to an edge amount of the respective target image.

11. The image processing apparatus as claimed in claim 1 , wherein the characteristic amount corresponds to a dispersion of the respective target image.

12. The image processing apparatus as claimed in claim 1 , wherein the characteristic amount corresponds to a standard deviation of the respective target image.

13. An image processing method detecting a plurality of joint positions between input images, executed by circuitry of a computer, the method comprising:
generating target images to be searched in a second input image based on a first input image, the first input image and the second input image being among the input images;
calculating a characteristic amount for each of the target images;
calculating a matching between a particular target image of interest among the target images, and a part of the second input image, using an evaluation method depending on the calculated characteristic amounts, wherein the characteristic amount is at least one of an edge amount, dispersion and standard deviation of the respective target image.

14. The image processing method of claim 13 , further comprising:
correcting an evaluation value at each of positions calculated by the matching using a correction method depending on the evaluation method used.

15. The image processing method of claim 14 , further comprising:
calculating a temporary position for the particular target image of interest based on a relative relationship of the characteristic amounts between the particular target image of interest and one or more of adjacent target images among the target images.

16. The image processing method of claim 13 , wherein the characteristic amount is an edge amount, dispersion and standard deviation of the respective target image.

17. The image processing method of claim 13 , wherein, as the evaluation method depending on the characteristic amounts, an evaluation method giving a correlation value is used when the characteristic amount is greater than a reference value, or another evaluation method giving an error difference value is used when the characteristic amount is less than the reference value.

18. An image processing system detecting a plurality of joint positions between input images, comprising:
an image capturing device; and
circuitry configured to:
obtain the input images from a plurality of captured images for a plurality of respective azimuths captured by the image capturing device;
generate target images to be searched in a second input image based on a first input image, the first input image and the second input image being among the input images;
calculate a characteristic amount for each of the target images; and
calculate a matching between a particular target image of interest among the target images, and a part of the second input image, using an evaluation method depending on the calculated characteristic amounts, wherein the characteristic amount is at least one of an edge amount, dispersion and standard deviation of the respective target image.

19. The image processing system of claim 18 , wherein, as the evaluation method depending on the characteristic amounts, an evaluation method giving a correlation value is used when the characteristic amount is greater than a reference value, or another evaluation m

Assignees

• Takenaka Hirokazu
• Yoshida Kazuhiro
• Kawaguchi Keiichi
• Ricoh Co Ltd

Inventors

• Takenaka Hirokazu
• Yoshida Kazuhiro
• Kawaguchi Keiichi

Classifications

• H04N1/387 Primary

  Composing, repositioning or otherwise {geometrically} modifying originals · CPC title

• H04N23/698

  for achieving an enlarged field of view, e.g. panoramic image capture · CPC title

• G06F18/22

  Matching criteria, e.g. proximity measures · CPC title

• H04N1/3872

  Repositioning or masking · CPC title

• G06T3/4038

  Image mosaicing, e.g. composing plane images from plane sub-images · CPC title