Automated tonal balancing

• US 10,311,557 B2
• Filed: 02/28/2017
• Issued: 06/04/2019
• Est. Priority Date: 05/08/2014
• Status: Active Grant

• Current Assignee: Maxar Intelligence Incorporated (Galileo Parent, Inc.)
• Original Assignee: DigitalGlobe, Inc. (Galileo Parent, Inc.)
• Inventors: Swanson, Nathan Reid, Young, Patrick, Stirling, Spencer
• Primary Examiner(s): Mistry, Oneal R
• Application Number:
  US15/445,949
• Publication Number:
  US 20170301070A1
• Time in Patent Office: 826 Days
• CPC Class Codes: G06T 2200/32 involving image mosaicing
• CPC Class Codes: G06T 2207/10024 Color image
• CPC Class Codes: G06T 2207/10032 Satellite or aerial image; Remote sensing
• CPC Class Codes: G06T 2207/20008 Globally adaptive
• CPC Class Codes: G06T 2207/20012 Locally adaptive
• CPC Class Codes: G06T 2207/20208 High dynamic range [HDR] image processing
• CPC Class Codes: G06T 2207/30181 Earth observation
• CPC Class Codes: G06T 3/4038 Image mosaicing, e.g. composing plane images from plane sub-images
• CPC Class Codes: G06T 5/70 Denoising; Smoothing
• CPC Class Codes: G06T 5/73 Deblurring; Sharpening
• CPC Class Codes: G06T 5/92 based on global image properties
• CPC Class Codes: G06T 5/94 based on local image properties, e.g. for local contrast enhancement

Abstract

A system for automated tonal balancing, comprising a rectification server that groups and processes images for use in tone-matching and provides them to a tone-matching server, that then performs tone-matching operations on the images and provides them as output for review or storage, and methods for tonal balancing using the system of the invention.

Claims

1. A system for automated generation of large-scale tonally-balanced orthomosaic images, comprising:

a tone-balancing server computer comprising a memory, a processor, and a plurality of programming instructions stored in the memory, wherein the programming instructions, when loaded into the processor, cause the processor to;

receive a group of orthorectified input images associated with a geographic region;

obtain a reference image for the geographic region;

for each orthorectified input image;

subdivide the respective orthorectified input image into a plurality of patches;

compute a pixel-level tonal mapping function for each patch to match a corresponding set of pixels of the reference image;

compute a first global tonal mapping function for the respective orthorectified input image by, for each pair of patches of the respective orthorectified input image, computing a function that smoothly connects the respective pixel-level tonal mapping functions for the pair of patches; and

apply the first global tonal mapping function to each pixel of the respective orthorectified input image to create a respective tonally-balanced orthorectified output image;

compute a second global tonal mapping reference by smoothly matching first global tonal matching functions for each neighboring pair of orthorectified output images; and

apply the second global tonal matching function to each pixel of each tonally-balanced orthorectified output image.

2. A method for automatically generating large-scale tonally-balanced orthomosaic images, the method comprising the steps of:

receiving, at a tone-balancing server computer comprising program code stored in a memory, a group of orthorectified input images associated with a geographic region;

obtaining a reference image for the geographic region;

for each orthorectified input image, performing the steps of;

subdividing the orthorectified input image into a plurality of patches;

computing a pixel-level tonal mapping function for each patch to match a corresponding set of pixels of the reference image;

computing a first global tonal mapping function for the respective orthorectified input image by, for each pair of patches of the respective orthorectified input image, computing a function that smoothly connects the respective pixel-level tonal mapping functions for the pair of patches; and

applying the first global tonal mapping function to each pixel of the respective orthorectified input image to create a plurality of tonally-balanced orthorectified output images;

computing a second global tonal mapping reference by smoothly matching first global tonal matching functions for each neighboring pair of orthorectified output images; and

applying the second global tonal matching function to each pixel of each tonally-balanced orthorectified output image.