Method and device for image rectification

Patent metadata

Field	Value
Publication number	US-11205281-B2
Application number	US-201816188325-A
Country	US
Kind code	B2
Filing date	Nov 13, 2018
Priority date	Nov 13, 2017
Publication date	Dec 21, 2021
Grant date	Dec 21, 2021

Abstract

The present invention provides a method and a device for image rectification, which is applied to the field of image processing. The method includes: receiving two images, wherein the two images are images of a target object captured at different viewpoints, and an image capturing device configured to capture the two images, includes a first image capturing device and a second image capturing device; determining positions of corresponding feature point pairs in the two images; and aligning the two images according to the positions of the feature point pairs and parameters of the first image capturing device and the second image capturing device. The above-mentioned method can rectify the images captured at different viewpoints, thereby ensuring the alignment of pixels, and avoiding visual discomfort of the viewer.

First claim

The invention claimed is:

1. A method for image rectification, comprising:
receiving two images, wherein the two images are images of a target object captured from different viewpoints, and an image capturing device configured to capture the two images comprises a first image capturing device and a second image capturing device;
determining positions of feature point pairs corresponding to each other in the two images;
removing feature point pairs having a wrong correspondence from the candidate feature point pairs using a Random Sample Consensus (RANSAC) algorithm;
aligning the two images according to the positions of the feature point pairs and parameters of the first image capturing device and the second image capturing device, wherein the step of aligning the two images includes:
determining an error function;
obtaining optimized parameters of the first image capturing device and the second image capturing device according to the error function; and
adjusting the two images according to the optimized parameters of the first image capturing device and the second image capturing device;

wherein
the feature point pair comprises a first point and a second point; and
the step of determining the error function comprises projecting the first point to an image coordinate system of an image with the second point by a three-dimensional point cloud according to the parameters of the first image capturing device and the second image capturing device to obtain a projection point;
the error function is a distance between the projection point and the second point;
the error function comprises a regularization term; and
the regularization term is determined by a multiplication of a coefficient and a sum of squares of a pair of predetermined rotation-related parameters from the first image capturing device and the second image capturing device;

cropping parts of the two images or a whole of the two images for splicing a stereogram from the two images after aligning, according to contents in the two images; and
splicing the parts of the two images or the whole of the two images to form the stereogram, wherein an image area for splicing is selected by adjusting a size and position of a dotted frame.

2. The method of claim 1 , wherein the step of determining the positions of the feature point pairs corresponding to each other in the two images comprises:
determining positions of candidate feature point pairs corresponding to each other in the two images.

3. The method of claim 1 , wherein the feature point pairs comprise one of items listed below:
points having variations in at least one direction greater than a first threshold; and
points with unchanged locations independent of whether the sizes of the two images change.

4. The method of claim 1 , wherein
the pair of predetermined rotation-related parameters correspond to an axis in a three-dimensional coordinate system; and
the regularization term is determined to ensure not excessive rotation over the axis when the error function is minimal.

5. The method of claim 4 , wherein the axis is Z-axis.

6. The method of claim 1 , wherein the optimized parameters of the first image capturing device and the second image capturing device comprise at least one item listed below:
rotation parameters, focal lengths, and coordinates of principal points.

7. A non-transitory computer readable storage medium storing computer programs, wherein when the computer programs are executed by a third processor, the steps of the method of claim 1 are implemented.

8. A device for image rectification, comprising a memory, a fourth processor, and computer programs stored in the memory and operable on the fourth processor, wherein when the computer programs are executed by the fourth processor, the steps of the method of claim 1 are implemented.

9. A device for image rectification comprising:
a receiver configured to receive two images, wherein the two images are images of a target object captured from different viewpoints;
an image capturing device configured to capture the two images, wherein the image capturing device comprises a first image capturing device and a second image capturing device; and
a first processor that
determines positions of feature point pairs corresponding to each other in the two images and removes feature point pairs having a wrong correspondence from the candidate feature point pairs using a Random Sample Consensus (RANSAC) algorithm, wherein the first processor further aligns the two images according to the positions of the feature point pairs and parameters of the first image capturing device and the second image capturing device;
determines an error function;
obtains optimized parameters of the first image capturing device and the second image capturing device according to the error function; and
adjusts the two images according to the optimized parameters of the first image capturing device and the second image capturing device;

wherein the feature point pair comprises a first point and a second point; and
wherein
determining the error function comprises projecting the first point to an image coordinate system of an image with the second point by a three-dimensional point cloud according to the parameters of the first image capturing device and the second image capturing device to obtain a projection point;
the error function is a distance between the projection point and the second point;
the error function comprises a regularization term; and
the regularization term is determined by a multiplication of a coefficient and a sum of squares of a pair of predetermined rotation-related parameters from the first image capturing device and the second image capturing device;

wherein the first processor is further configured to crop parts of the two images or a whole of the two images for splicing a stereogram from the two images after aligning, according to contents in the two images, and splice the parts of the two images or the whole of the two images to form the stereogram.

10. The device of claim 9 , wherein the first processor is further configured to
determine positions of candidate feature point pairs corresponding to each other in the two images.

11. The device of claim 9 , wherein the feature point pairs comprise one of items listed below:
points having variations in at least one direction greater than a first threshold; and
points with unchanged locations independent of whether the sizes of the two images change.

12. A system for image rectification comprising a first image capturing device, a second image capturing device, and a second processor;
wherein
the first image capturing device and the second image capturing device are configured to capture images of a target object from different viewpoints;
the second processor acquires the images of the target object captured by the first image capturing device and the second image capturing device at the different viewpoints, determines positions of feature point pairs corresponding to each other in the image, removes feature point pairs having a wrong correspondence from the candidate feature point pairs using a Random Sample Consensus (RANSAC) algorithm, and aligns the two images according to the positions of the feature point pairs and parameters of the first image capturing device and the second image capturing device;
determines an error function relative to a first parameter obtained according to the positions of the feature point pairs and the parameters of the first image capturing device and the second image capturing device,
obtains optimized parameters of the first image capturing device and the second image captur

Assignees

• Arcsoft Corp Ltd

Inventors

• Yu Li
• Miao Wang
• Fan Xin

Classifications

• H04N23/80

  Camera processing pipelines; Components thereof · CPC title

• H04N23/90

  Arrangement of cameras or camera modules, e.g. multiple cameras in TV studios or sports stadiums · CPC title

• H04N23/951

  by using two or more images to influence resolution, frame rate or aspect ratio · CPC title

• G06T7/337 Primary

  involving reference images or patches · CPC title

• G06T5/50

  using two or more images, e.g. averaging or subtraction · CPC title