Image processing method and apparatus for generating panoramic image

• US 7,388,982 B2
• Filed: 11/26/2003
• Issued: 06/17/2008
• Est. Priority Date: 11/29/2002
• Status: Expired due to Fees

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Suzuki, Masahiro, Katayama, Akihiro, Sakagawa, Yukio, Kotake, Daisuke, Endo, Takaaki
• Primary Examiner(s): Bella; Matthew C
• Assistant Examiner(s): LIEW, ALEX KOK SOON
• Application Number:
  US10/721,875
• Publication Number:
  US 20040228544A1
• Time in Patent Office: 1,665 Days
• Field of Search: None
• US Class Current: 382/154
• CPC Class Codes: G06T 15/205 Image-based rendering
• CPC Class Codes: G06T 2207/20212 Image combination
• CPC Class Codes: G06T 3/4038 Image mosaicing, e.g. composing plane images from plane sub-images
• CPC Class Codes: G06T 5/90 Dynamic range modification of images or parts thereof
• CPC Class Codes: G06T 7/30 Determination of transform parameters for the alignment of images, i.e. image registration
• CPC Class Codes: H04N 23/698 for achieving an enlarged field of view, e.g. panoramic image capture

Abstract

There are disclosed an image processing method and apparatus for generating a panoramic image in which the hue or brightness discontinuities generated at the seams of source images are reduced. According to the image processing method of this invention, upon compositing source images (221, 222) having an overlap portion, the hue or brightness of each source image is converted (H211, H221) using a reference image which has overlap portions with both the source images to be composited, and the converted images (241, 242) are composited, thereby reducing the hue or brightness discontinuities generated at the seam of images.

Claims

1. An image processing method for generating a full-view panoramic image by compositing a plurality of images each of which has overlapping portions with neighboring images, comprising the steps of:

acquiring, when an arbitrary one of the plurality of images is defined as a first image, and images to be composited as right and left neighboring images of the first image are defined as second and third images, a first reference image having overlapping portions with both the first and second images, and a second reference image having overlapping portions with both the first and third images;

calculating first and second conversion conditions required to adjust a hue or brightness level of the first image to hue or brightness levels of the first and second reference images;

generating a converted image by converting a left predetermined region of the first image on the basis of the first conversion condition and converting a right predetermined region of the first image on the bases of the second conversion condition, and

generating a full-view panoramic image by compositing the converted images generated for all of the plurality of images,

wherein the conversions based on the first and second conversion conditions are weighted depending on distances from right and left ends of the first image.

2. The method according to claim 1, wherein the first and second reference images are generated from one image.

3. A computer-readable storage medium storing a computer program that makes a computer execute an image processing method for generating a full-view panoramic image by compo siting a plurality of images each of which has overlapping portions with neighboring images, said computer program comprising:

code of a step of acquiring, when an arbitrary one of the plurality of images is defined as a first image, and images to be composited as right and left neighboring images of the first image are defined as second and third images, a first reference image having overlapping portions with both the first and second images, and a second reference image having overlapping portions with both the first and third images;

code of a step of calculating first and second conversion conditions required to adjust a hue or brightness level of the first image to hue or brightness levels of the first and second reference images;

code of a step of generating a converted image by converting a left predetermined region of the first image on the basis of the first conversion condition and converting a right predetermined region of the first image on the basis of the second conversion condition; and

code of a step of generating a full-view panoramic image by compositing the converted images generated for all of the plurality of images,

wherein the conversions on the basis of the first and second conversion conditions are weighted depending on distances from right and left ends of the first image.

4. An image processing apparatus for generating a full-view panoramic image by compositing a plurality of images each of which has overlapping portions with neighboring images, comprising:

an acquiring unit adapted to, when an arbitrary one of the plurality of images is defined as a first image, and images to be composited as right and left neighboring images of the first image are defined as second and third images, acquire a first reference image having overlapping portions with both the first and second images, and a second reference image having overlapping portions with both the first and third images;

a calculating unit adapted to calculate first and second conversion conditions required to adjust a hue or brightness level of the first image to hue or brightness levels of the first and second reference images;

a first generating unit adapted to generate a converted image by converting a left predetermined region of the first image on the basis of the first conversion condition and converting a right predetermined region of the first image on the basis of the second conversion condition; and

a second generating unit adapted to generate a full-view panoramic image by compositing the converted images generated for all of the plurality of images,

wherein, in said first generating unit, the conversions on the basis of the first and second conversion conditions are weighted depending on distances from the right and left ends of the first image.