System for collecting and processing aerial imagery with enhanced 3D and NIR imaging capability
First Claim
Patent Images
1. A system for collection and processing aerial imagery collected from an aerial vehicle, comprising:
- at least one camera;
an avionics box comprising a microprocessor;
a GPS receiver;
a remote computing device in said aerial vehicle and a software application comprising instructions executable by said remote computing device for guidance along a raster flight plan across a predetermined geographic area and automatically triggering said at least one camera to collect overlapping images based on coordinates of said GPS receiver; and
a computing cloud comprising a plurality of cloud computing nodes, at least one data storage device and at least one processing device running at least one software application, wherein said at least one software application further comprises,computer instructions stored on said data storage device and executable by said at least one processing device that locate image tie points as between two or more images via an iterative multi-resolution matching method in which image tie points between said plural images are repetitively cross-correlated in iterations of increasingly finer resolution and stored at said data storage device until all said image tie points at all said resolutions are identified and stored at said data storage device, andcomputer instructions stored on said data storage device and executable by said at least one processing device for state estimation of said at least one camera orientation and GPS receiver error from said stored image tie points, for orthorectification of said two or more images for perspective and relief error by association of at least one pixel in each of said two or more images with a point in a geographic coordinate reference system, and for accurately mapping said corrected two or more images to a map projection.
0 Assignments
0 Petitions
Accused Products
Abstract
A system for guided geospatial image capture, registration and 2D or 3D mosaicking, that employs automated imagery processing and cutting-edge airborne image mapping technologies for generation of geo-referenced Orthomosaics and Digital Elevation Models from aerial images obtained by UAVs and/or manned aircraft.
28 Citations
27 Claims
-
1. A system for collection and processing aerial imagery collected from an aerial vehicle, comprising:
-
at least one camera; an avionics box comprising a microprocessor; a GPS receiver; a remote computing device in said aerial vehicle and a software application comprising instructions executable by said remote computing device for guidance along a raster flight plan across a predetermined geographic area and automatically triggering said at least one camera to collect overlapping images based on coordinates of said GPS receiver; and a computing cloud comprising a plurality of cloud computing nodes, at least one data storage device and at least one processing device running at least one software application, wherein said at least one software application further comprises, computer instructions stored on said data storage device and executable by said at least one processing device that locate image tie points as between two or more images via an iterative multi-resolution matching method in which image tie points between said plural images are repetitively cross-correlated in iterations of increasingly finer resolution and stored at said data storage device until all said image tie points at all said resolutions are identified and stored at said data storage device, and computer instructions stored on said data storage device and executable by said at least one processing device for state estimation of said at least one camera orientation and GPS receiver error from said stored image tie points, for orthorectification of said two or more images for perspective and relief error by association of at least one pixel in each of said two or more images with a point in a geographic coordinate reference system, and for accurately mapping said corrected two or more images to a map projection. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11)
-
-
12. In a system comprising a cloud computing system comprising a plurality of cloud computing nodes, at least one data storage device and at least one processing device, said cloud computing system being connected via a communications network to at least one remote computer workstation, and a portable computing device coupled with at least one imaging device and at least one GPS receiver, a method for collecting and processing aerial images captured along one or more flight paths over a target geographic area comprising the steps of:
-
collecting, by said remote computer workstation, a plurality of GPS data elements pertaining to said flight path from said portable computing device, and a plurality of said images from said at least one imaging device, wherein each one of said plurality of images is associated with at least one of said plurality of GPS data elements; applying, by at least one of said plurality of cloud computing nodes, an iterative multiresolution linking process to said plurality of images whereby image tie points between two or more of said plurality of images are identified in iterations of increasingly finer resolution while refining an estimated position and orientation of said at least one imaging device until all said image tie points at all said resolutions are identified and stored, said multiresolution linking process further comprising, applying a navigation process to said plurality of images whereby estimated position and orientation inclusive of a first roll, pitch, yaw, and absolute three-dimensional location of said imaging device in the geographic coordinate reference system is determined with respect to each one of said plurality of images, and selecting, from said plurality of images, a home frame image and a target frame image, binning said home frame and said target frame; extracting one or more chips from said binned home frame image, each extracted chip comprising a grouping of adjacent pixels; identifying image tie point matches between each of said extracted chips relative to said binned target frame image; and iteratively, repeating said binning, extracting and tie-point identification substeps, binning less on each iteration and saving the image tie point matches at each iteration for subsequent use; applying, by at least one of said plurality of cloud computing nodes, an orthorectification process to each one of said plurality of images, whereby each one of said plurality of images is assigned a geographic referencing metadata tag indicating the association between at least one pixel in each of said plurality of images with a point in the geographic coordinate reference system. - View Dependent Claims (13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25)
-
-
26. A cloud-based system for collecting, processing and distributing aerial imagery, comprising:
-
a first cloud computing node connected in the cloud computing network comprising a processor and a computer readable medium storing a user software application accessible by any number of remote computers, said user software application comprising computer-executable instructions for uploading a plurality of aerial image photos and a corresponding flight log and calibration data from said remote computer(s) to the cloud computing network, and a second cloud computing node connected in the cloud computing network comprising a processor and a computer readable medium storing a cloud-based application comprising computer-executable instructions for carrying out the steps of, successively performing at a plurality of different binned resolutions first-pass of an image tie point registration to determine image tie-point matches between successive pairs of said uploaded aerial image photos, and storing all identified image tie points at all said resolutions at said second cloud computing node, performing between each successive first-pass image tie point registration a state solution estimation by applying a least squares estimation method to estimate camera roll (r), pitch (p), and yaw (y) states for said uploaded aerial image photos, GPS corrections along three axes (x, y, z), and camera calibration states and utilizing said state solution estimation in a successive first-pass image tie point registration, ortho-rectifying each of said uploaded aerial image photos to a map projection in accordance with said estimated state solution, and displaying said orthorectified map projection to one of said remote users. - View Dependent Claims (27)
-
Specification