Automated seamline construction for high-quality high-resolution orthomosaics

Patent metadata

Field	Value
Publication number	US-2016005149-A1
Application number	US-201514789877-A
Country	US
Kind code	A1
Filing date	Jul 1, 2015
Priority date	Jul 1, 2014
Publication date	Jan 7, 2016
Grant date	—

Abstract

A system for semi-automated feature extraction comprising an image analysis server that receives and initializes a plurality of raster images, a feature extraction server that identifies and extracts image features, a mosaic server that assembles mosaics from multiple images, and a rendering engine that provides visual representations of images for review by a human user, and a method for generating a cost raster utilizing the system of the invention.

First claim

What is claimed is:

1 . A system for semi-automated feature extraction, comprising:
an image analysis server computer comprising program code stored in a memory and adapted to process raster images to determine image and pixel information and provide the images and derived information to other components of the system; a feature extraction server computer comprising program code stored in a memory and adapted to identify image features based at least in part on derived information for a raster image, and to provide the image and identified features to other components of the system; a mosaic imaging server computer comprising program code stored in a memory and adapted to receive a plurality of raster images and assemble the images to form mosaics and provide the mosaics to other components of the system; and a rendering engine computer comprising program code stored in a memory and adapted to receive a plurality of data from other components of the system and create visual representations of the data for review by a human user.

2 . The system of claim 1 , further comprising a database computer comprising program code stored in a memory and adapted to store and provide information for other components of the system.

3 . The system of claim 1 , further comprising a viewer device adapted to receive information from other components of the system and present the information for review by a human user.

4 . The system of claim 1 , further comprising a plurality of input devices adapted to receive input from a human user and provide the results of the input to other components of the system.

5 . A method for creating a cost raster for semi-automated feature extraction, comprising the steps of:
receiving, at an image analysis server, a plurality of input images; initializing at least a portion of the plurality of input images for processing; deriving initial image characteristics based at least in part on at least a portion of the plurality of input images; identifying and extracting, using a feature extraction server, a plurality of image features based at least in part on at least a portion of the input images; and computing a cost raster based at least in part on at least a portion of the plurality of image features extracted.

6 . The method of claim 5 , further comprising the step of providing the cost raster as output for storage or further review.

7 . The method of claim 5 , wherein the image features are based at least in part on at least a portion of the identified image characteristics derived in a previous step.

8 . The method of claim 5 , wherein the image characteristics comprise at least a plurality of pixel qualities.

9 . The method of claim 8 , wherein the plurality of pixel qualities comprise at least color qualities.

Assignees

• Digitalglobe Inc

Inventors

• Malitz Seth

Classifications

• G06F16/56

  having vectorial format · CPC title

• G06F16/532

  Query formulation, e.g. graphical querying · CPC title

• G06F16/5838

  using colour · CPC title

• G06T3/4038 Primary

  Image mosaicing, e.g. composing plane images from plane sub-images · CPC title

• G06T2207/30181

  Earth observation · CPC title