UAV Georeferencing Stabilization via Lattice Map Averaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing georeferencing methods for UAV-captured videos are prone to high frame-to-frame jittering due to small errors in sensor models, which are magnified at ultrahigh zoom levels, leading to unstable geo projections and hindering further processing.
Innovation Solution
A computer-implemented method and system for stabilizing georeferencing in UAV images in real-time, which constructs a geographic lattice for each frame, aligns frames, averages raw geo coordinates, and builds a lattice map with stabilized geo coordinates, all without relying on ground-truth maps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If ultrahigh resolution mapping is used to capture detailed geolocation information, then measurement precision is improved, but frame-to-frame jittering increases due to magnified sensor errors
Solution Approach 1:
The patent combines multiple consecutive video frames to create a composite georeferenced output. By merging information from multiple frames, the system averages out sensor errors and achieves both high precision and frame-to-frame stability. The composite image integrates georeferenced data from multiple source frames, reducing the impact of individual frame jittering.
Solution Approach 2:
The system processes video frames continuously rather than independently. By maintaining continuous processing of consecutive frames, the georeferencing operation benefits from temporal coherence and cumulative error reduction. The continuous action allows the system to maintain stable geo-projection across frames while preserving ultrahigh resolution details.
2Measurement precision
If ground-truth maps are used for georeferencing, then measurement precision is improved, but computational resources and processing time increase
Solution Approach 1:
Instead of relying on external ground-truth maps, the system creates internal reference frames by georeferencing the first and last frames of the video sequence. These copied georeferenced frames serve as reference points for intermediate frames, eliminating the need for resource-intensive ground-truth map processing while maintaining georeferencing accuracy.
Solution Approach 2:
The system performs georeferencing on the first and last frames in advance before processing intermediate frames. This preliminary action establishes reference georeferenced frames that guide the processing of subsequent frames, reducing computational overhead during real-time processing while maintaining precision.
3Productivity
If real-time processing is implemented, then productivity is improved, but measurement precision may deteriorate due to reduced processing time
Solution Approach 1:
The video processing is segmented into distinct phases: georeferencing the first and last frames, then using these as references for intermediate frames. This segmentation allows computationally intensive operations to be performed only on boundary frames, enabling real-time processing of intermediate frames while maintaining overall georeferencing precision through the reference framework.
Data Source
AI summary
A computing system and methods are provided for georeferencing stabilization. An exemplary method includes: obtaining a video stream capturing an area from a camera of a drone, where the video stream includes a plurality of frames, each including a field of view of the image capturing device and metadata of the image capturing device when the frame is captured; constructing a geographic (geo) lattice for the field of view in each of the plurality of frames, the geo lattice comprises a plurality of points, each being associated with raw coordinates determined based on the corresponding metadata; and building a lattice map with stabilized geo coordinates by (1) aligning the frames, (2) averaging the raw geo coordinates for given intersection points, and (3) building the lattice map based on the averaged geo coordinates of the intersection points.