US 11,232,540 B2Granted

Composite video image creation apparatus, composite video image creation method, and composite video image creation program

AssigneeMurakami Corporation

InventorAtsushi Hayami

Life of this patent

Estimating the term…

Loading patent…

Abstract

With respect to two images acquired from two video images including a mutually-overlapping area in which the two video images overlap each other, an image transformation matrix for mapping coordinate systems is sequentially generated. Coordinate transformation of at least one image of the images is performed using the generated image transformation matrix. A composite image is created by overlaying two images with the at least one image subjected to the coordinate transformation. The currently used image transformation matrix and the newly generated image transformation matrix are compared with each other. If the two image transformation matrices are similar, the coordinate transformation is performed using the currently used image transformation matrix continuously. If the two image transformation matrices are dissimilar, the coordinate transformation is performed using an image transformation matrix corrected with the newly generated image transformation matrix.

A composite video image creation apparatus for creating a composite video image by stitching two video images each including a mutually-overlapping area in which the two video images overlap each other, via the mutually-overlapping areas,

the composite video image creation apparatus comprising an image acquisition section, a feature value extraction section, an image transformation matrix generation section, an image transformation section, an image output section and a correction determination section, wherein:
the image acquisition section sequentially acquires two images, one from each of the two video images;
the feature value extraction section detects respective feature points with respect to the acquired two images and extracts respective feature values of the detected feature points;
the image transformation matrix generation section detects corresponding points in the two images based on the extracted feature values of the feature points, and sequentially generates an image transformation matrix for mapping coordinate systems of the two images to each other, based on coordinates of the detected corresponding points;
the image transformation section performs coordinate transformation of at least one image of the two images using the generated image transformation matrix to perform mapping of the coordinate systems of the two images to each other;
the image output section creates a composite image by overlaying the two images with the respective coordinate systems mapped to each other, and outputs the composite image;
the correction determination section compares the currently used image transformation matrix and the newly generated image transformation matrix with each other to determine whether or not the two image transformation matrices are similar, based on a predetermined determination criterion for determining a degree of difference or a degree of similarity between the two image transformation matrices, and if it is determined that the two image transformation matrices are similar, makes the image transformation section perform the coordinate transformation using the currently used image transformation matrix continuously, and if it is determined that the two image transformation matrices are dissimilar, makes the image transformation section perform the coordinate transformation using an image transformation matrix corrected with the newly generated image transformation matrix;
and
the composite video image creation apparatus creates the composite video image by sequentially performing creation of the composite image according to progress of time.

6 more elements

Facts

Application

17/018,414Filed 11 Sep 2020 · continuation

Priority

JP JP2018-04760015 Mar 2018

Published

US 2020/0410641 A131 Dec 2020

Granted

25 Jan 2022

Term ends

Not estimatedThe source records don’t support an estimate.

Classification7 codes

G06T 3/4038
Scaling of whole images or parts thereof, e.g. expanding or contracting: image mosaicing, e.g. composing plane images from plane sub-images
H04N 7/181
Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast: for receiving images from a plurality of remote…
G06T 7/187
Segmentation; Edge detection: involving region growing; involving region merging; involving connected component labelling
G06T 7/37
Determination of transform parameters for the alignment of images, i.e. image registration: using transform domain methods

Abstract

Claims12