Aerial Survey Mosaic Creation via 3D Ray Tracing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing aerial surveying methods struggle to rapidly merge heavily overlapped imagery data into a comprehensive mosaic using geo-referencing information without extensive data collection or computation, often resulting in incomplete or distorted images.

Innovation Solution

A two-pass process combining GPS, INS, and camera angle data to create a mosaic image using 3-D ray tracing and graphics methodologies, which eliminates the need for stereo imagery or ortho-rectification, and employs the Best Pixel Rule to select and combine pixels for accurate image registration.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional photogrammetry approaches are used to merge overlapping imagery data into a mosaic, then the mosaic can be created with reasonable accuracy, but the process is labor-intensive and requires extensive data collection and computation

Engineering Contradiction:
Improvemosaic accuracyVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces manual photogrammetry operations with automated computer-based image processing. The system automatically detects features, establishes correspondences between images, and computes the mosaic using algorithms, eliminating the need for manual control points and extensive computational processing required by traditional photogrammetry methods.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system performs self-calibration and automatic feature detection without requiring external reference data or manual intervention. The algorithm automatically identifies common features across overlapping images, establishes their correspondences, and computes the transformation parameters needed for mosaic creation, making the process self-sufficient and highly automated.

Inventive Principle:
Principle #25Self-service

2Measurement precision

If orthorectification is performed before mosaicing, then the image registration accuracy is improved, but the complexity of the process increases and requires stereo images or very high-resolution camera data

Engineering Contradiction:
Improveimage registration accuracyVSAvoiddata collection requirements
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the complex task of image registration into separate manageable steps: feature detection in individual images, feature correspondence establishment between images, and transformation parameter computation. This segmentation allows the system to achieve accurate registration without requiring complex orthorectification procedures or specialized stereo image data.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system introduces an intermediary feature correspondence layer that mediates between the input images and the final mosaic. By first identifying and matching common features across images, then computing transformation parameters based on these correspondences, the system achieves accurate registration without requiring direct complex geometric processing or stereo image pairs.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Device complexity

If the latest image is glued on top of previous images without proper integration, then the mosaic creation is simplified, but the adjacent image edges are not blended resulting in a crude paste up appearance

Engineering Contradiction:
Improvemosaic creation processVSAvoidimage blending quality
Core Design Contradiction:
Device complexityVSManufacturing precision

Solution Approach 1:

The patent merges multiple images into a unified mosaic by establishing correspondences between common features across all images. The system computes transformation parameters that align all images to a common coordinate system, then blends them together seamlessly, combining the information from all images rather than simply stacking them in sequence.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The system performs preliminary feature detection and correspondence establishment for all images before computing the final transformation parameters and creating the mosaic. This preliminary processing ensures that all images are properly aligned and integrated before the final blending operation, preventing the crude paste-up appearance by ensuring smooth transitions at image boundaries.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2212858B1Method and apparatus of taking aerial surveys
Publication Date: 2017.05.03 INTERGRAPH CORP
  • EP2212858B1 patent drawingFigure 1
  • EP2212858B1 patent drawingFigure 2~4
  • EP2212858B1 patent drawingFigure 3a

AI summary

A method of taking an aerial survey maps boundaries of a first image and a second image from a first plane to a second plane to determine boundaries of an output image in the second plane. For a plurality of pixels in the output image, the method determines a corresponding pixel of either the first image or second image in the first plane.