US2013287316A1PendingUtilityA1

Image synthesis apparatus and image synthesis method

Assignee: FUJITSU LTDPriority: Nov 4, 2008Filed: Jun 25, 2013Published: Oct 31, 2013
Est. expiryNov 4, 2028(~2.3 yrs left)· nominal 20-yr term from priority
G06T 3/4038G06T 11/60
51
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

An image synthesis apparatus includes an image obtainer obtaining a plurality of images taken by a plurality of cameras; a pixel value corrector which, with respect to each of the images, calculates a correction factor of an average pixel value in overlap regions located at the both ends of an image and, in the adjacent direction of the images, calculates the correction factor of a pixel value in a region other than the overlap regions by a predetermined continuous function connecting the correction factors in the overlap regions and corrects the entire image based on the calculated correction factor; a synthetic image generator synthesizing the images corrected by the pixel value corrector to generate one synthetic image; and an image output outputting the synthetic image to an external apparatus.

Claims

exact text as granted — not AI-modified
1 . An image synthesis apparatus comprising:
 an image obtainer which obtains a plurality of images taken by a plurality of cameras;   a feature quantity calculator which, with respect to each of the images, extracts one or a plurality of attention regions from an upper region and a lower region in an overlap region located at the both ends of an image and the overlap regions comprising the image region and an image region of an image taken by another camera among the plurality of cameras, and calculates an average pixel value of each of the extracted attention regions;   a desired value calculator which, with respect to each of the attention regions, calculates a desired value of the average pixel value after correction, based on the average pixel value of the attention region in one image and the average pixel value of the attention region in another image overlapping the attention region in the one image;   a correction factor calculator which, with respect to each of the images, calculates the correction factor of the representative point, which is one pixel in the attention region, based on the average pixel value of the attention region and the desired value, divides the image into a plurality of regions based on each of the representative points, and calculates the correction factor of each pixel in each of the regions based on the correction factor at the representative point constituting the relevant region;   a correction processor which corrects a pixel value of a pixel of each of the images based on the correction factor calculated by the correction factor calculator and corresponding to the pixel;   a synthetic image generator which synthesizes the images corrected by the correction processor to generate one synthetic image; and   an image output unit which outputs the synthetic image generated by the synthetic image generator to an external apparatus.   
     
     
         2 . The image synthesis apparatus as claimed in  claim 1 , wherein the correction factor calculator which divides the image into inside and outside regions of a convex polygon formed by connecting each representative point in the image, and further divides the inside region into a plurality of triangular regions formed by connecting the representative points. 
     
     
         3 . The image synthesis apparatus as claimed in  claim 1 , wherein when there are four representative points in the image, the correction factor calculator divides a rectangular shape, formed by the four representative points, into two triangular regions by a diagonal with a smaller absolute value of the gradient of two diagonals of the rectangular shape.

Join the waitlist — get patent alerts

Track US2013287316A1 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.