Picture processing method and apparatus
First Claim
Patent Images
1. An image processing method comprising:
- a feature extracting step of extracting feature points of two or more images;
a matching step of comparing feature points of one of the two or more images to those of other images for effecting matching;
a computing step of performing computations for varying the relative position between said one image and the other image based on the results of said matching step;
wherein said matching step detects the largest one of common point sets between a feature point group made up of all feature points on said one image and feature point groups made up of all feature points on said other images; and
wherein said matching step includes a distance computing step of computing all of distances of the feature points making up the feature point group of said one image and the feature points making up the feature point groups of the other images;
a selecting step of selecting a common distance having the smallest number of the repetitive distances from common distances common to the distances of the feature points making up a feature point group for said one image, as computed by said distance computing step, and to the distances of the feature points making up feature point groups for the other images, similarly as measured by said distance computing step; and
a testing step of testing all positions equal in number to twice said smallest number of the repetitive distances;
said matching step detecting said largest common point set.
1 Assignment
0 Petitions
Accused Products
Abstract
A method and apparatus for processing plural pictures to generate a synthesized picture, in which plural images are synthesized by application of a geometrical model. An image processing apparatus 10 extracts feature points from each of two or more images. Feature points of one of the images and those of the other image or images are compared to each other to effect matching. Based on the results of this matching, computations are conducted to change the relative position of the one and the other images to synthesize the two or more images.
70 Citations
12 Claims
-
1. An image processing method comprising:
-
a feature extracting step of extracting feature points of two or more images;
a matching step of comparing feature points of one of the two or more images to those of other images for effecting matching;
a computing step of performing computations for varying the relative position between said one image and the other image based on the results of said matching step;
wherein said matching step detects the largest one of common point sets between a feature point group made up of all feature points on said one image and feature point groups made up of all feature points on said other images; and
wherein said matching step includes a distance computing step of computing all of distances of the feature points making up the feature point group of said one image and the feature points making up the feature point groups of the other images;
a selecting step of selecting a common distance having the smallest number of the repetitive distances from common distances common to the distances of the feature points making up a feature point group for said one image, as computed by said distance computing step, and to the distances of the feature points making up feature point groups for the other images, similarly as measured by said distance computing step; and
a testing step of testing all positions equal in number to twice said smallest number of the repetitive distances;
said matching step detecting said largest common point set. - View Dependent Claims (2, 3, 4, 5, 6)
an overlapping step of overlapping at least portions of said two or more images prior to said feature extracting step;
said feature extracting step preferentially extracting feature points lying in overlapping portions formed on at least partially overlapping said two or more images by said overlapping step and in the vicinity thereof.
-
-
3. The image processing method according to claim 1, wherein said matching step effects Hausdorff matching to synthesize said two or more images.
-
4. The image processing method according to claim 1
wherein said matching step effects bottleneck matching to synthesize said two or more images. -
5. The image processing method according to claim 1
wherein said computing step causes translation and/or rotation of said one and the other images on a space, based on the results of said matching step, to vary the relative position between said one image and the other images. -
6. The image processing method according to claim 5
wherein said computing step executes affine transformation or adjusts the perspective relation to carry out computations such as to change the relative position between said one image and the other images.
-
7. An image processing apparatus comprising:
-
feature extracting means for extracting feature points of two or more images;
matching means for comparing feature points of one of the two or more images to those of other images for effecting matching; and
computing means for performing computations for varying the relative position between said one image and the other image based on the results of matching by said matching means;
wherein said matching means detects the largest one of common point sets between a feature point group made up of all feature points on said one image and feature point groups made up of all feature points on said other images; and
wherein said matching means computes all of distances of the feature points making up a feature point group of said one image and the feature points making up feature point groups of the other images, selects a common distance having the smallest number of the repetitive distances from the computed common to the computed distances of the feature points making up a feature group for said one image and to the computed distances of the feature points making up feature point for the other images;
testing being made for all positions equal in number to twice said smallest number of the repetitive distances to detect said largest common point set.- View Dependent Claims (8, 9, 10, 11, 12)
overlapping means for overlapping at least portions of said two or more images;
said feature extracting means preferentially extracting feature points lying in overlapping portions formed on at least partially overlapping said two or more images by said overlapping means, and in the vicinity thereof.
-
-
9. The image processing apparatus according to claim 7, wherein said matching process effects Hausdorff matching to synthesize said two or more images.
-
10. The image processing apparatus according to claim 7
wherein said matching process effects bottleneck matching to synthesize said two or more images. -
11. The image processing apparatus according to claim 7
wherein said computing means causes translation and/or rotation of said one and the other images on a space, based on the results of said matching means, to vary the relative position between said one image and the other images. -
12. The image processing apparatus according to claim 11
wherein said computing means executes affine transformation or adjusts the perspective relation to carry out computations such as to change the relative position between said one image and the other images.
Specification