Multi-Dimensional Topographical Point Registration Using k-D Tree Filtering

Solution Overview

Problem

Current methods for registering multidimensional topographical data points from different perspectives are computationally intensive and time-consuming, particularly when dealing with sparse data due to obstructions like tree canopies, which hinders the generation of accurate and timely three-dimensional models.

Innovation Solution

A method that involves receiving digital elevation model points, finding corresponding points across frames, determining rotation and translation transformations, and iterating to optimize registration, while using a ringing removal filter and k-D tree algorithms to improve signal-to-noise ratio and reduce processing time by focusing on ground surface points and employing polynomial fits for accurate ground surface representation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If known methods of registering frames with large numbers of data points are used, then registration accuracy is improved, but processing time increases significantly

Engineering Contradiction:

Improveregistration accuracyVSAvoidprocessing time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the registration process into multiple stages: first extracting ground surface points using polynomial fits, then performing registration only on these filtered points rather than all data points. This segmentation reduces the computational workload while maintaining registration accuracy on the relevant ground surface data.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent extracts and removes obstructions (tree canopies, buildings, etc.) from the dataset by filtering out points that do not represent ground surface. Using polynomial fits to model ground surface, the system extracts only the relevant ground points for registration, eliminating unnecessary computational processing of obstructing objects.

Inventive Principle:

Principle #2Taking out (Extraction)

2Quantity of substance

If all data points including obstructions are processed, then completeness of data is maintained, but signal-to-noise ratio decreases

Engineering Contradiction:

Improvedata completenessVSAvoidsignal-to-noise ratio

Core Design Contradiction:

Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent extracts and removes obstructions (tree canopies, buildings, etc.) from the dataset by filtering out points that do not represent ground surface. Using polynomial fits to model ground surface, the system extracts only the relevant ground points for registration, eliminating unnecessary computational processing of obstructing objects.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The patent applies different processing quality to different parts of the data. Ground surface points receive full processing attention for registration, while obstruction points are filtered out. This local quality differentiation ensures high signal-to-noise ratio for the critical ground surface registration while maintaining overall data completeness.

Inventive Principle:

Principle #3Local quality

3Measurement precision

If iterative registration process is performed multiple times, then registration optimization is improved, but computational complexity increases

Engineering Contradiction:

Improveregistration optimizationVSAvoidcomputational complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the registration process into multiple stages: first extracting ground surface points using polynomial fits, then performing registration only on these filtered points rather than all data points. This segmentation reduces the computational workload while maintaining registration accuracy on the relevant ground surface data.

Inventive Principle:

Principle #1Segmentation

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach significantly reduces processing time and improves the accuracy of three-dimensional models by filtering out obstructions and using efficient point matching algorithms, enabling faster generation of clear terrain images even under heavy obscuration.

Implementation Method 1

The LIDAR instrument 102 emits a plurality of laser light pulses which are directed toward the ground

Methodology Applied

Scientific EffectLaser: Laser

Implementation Method 2

The instrument 102 comprises a sensor 103 that detects the reflections/scattering of the pulses

Methodology Applied

Scientific EffectReflection: Reflection

Implementation Method 3

finding for each of a plurality of points in a first frame a corresponding closest point in a plurality of subsequent frames

Methodology Applied

Scientific EffectSpatial indexing:

Implementation Method 4

employing polynomial fits for accurate ground surface representation

Methodology Applied

Scientific EffectPolynomial fitting:

Data Source

PatentUS7567731B2Method and system for simultaneously registering multi-dimensional topographical points

Publication Date: 2009.07.28 HARRIS CORP

AI summary

A method for registering multi-dimensional topographical data points representing a multi-dimensional object, comprises: a) receiving a plurality of points representing a plurality of overlapping frames of a surface of the multi-dimensional object; b) finding for each point in a first frame a corresponding closest point in each of a plurality of subsequent frames; c) determining a rotation and translation transformation for each of the plurality of frames so that corresponding closest points are aligned; d) determining a cost for performing each rotation and translation transformation; and e) repeating steps b) through d) for additional frames to provide an optimum transformation.