Initial Azimuth Calculation via Bundle Adjustment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional methods for determining initial azimuth in photogrammetry-based surveying, such as using compass sensors, are unreliable due to magnetic anomalies, leading to inaccurate results and prolonged processing times in bundle adjustment algorithms.
Innovation Solution
A method that generates initial azimuth values without relying on compass data by using algorithms like the four-station-one-tie-point (4S1P) and two-station-two-tie-point (2S2P) algorithms, which calculate potential tie-point positions and select the most likely one, enabling accurate determination of camera poses and facilitating automatic tie-point finding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If compass sensors are used to determine initial azimuth, then the surveying process can proceed with automated tie-point finding, but the results become unreliable due to magnetic anomalies
Solution Approach 1:
The patent extracts the azimuth determination process from the unreliable compass sensor and relocates it to the bundle adjustment algorithm. Instead of relying on the compass to provide initial azimuth values, the system removes this dependency and calculates azimuth purely from geometric relationships between tie-points and camera positions during bundle adjustment, eliminating the harmful influence of magnetic anomalies
Solution Approach 2:
The patent introduces tie-points as intermediary elements that mediate between camera positions and azimuth determination. By establishing geometric relationships through multiple tie-points observed from different camera stations, the system creates a reliable indirect path to determine azimuth that bypasses the faulty compass sensor entirely
2Reliability
If manual tie-point selection is used to ensure accuracy, then azimuth can be determined reliably, but processing time increases significantly
Solution Approach 1:
The patent performs preliminary automated identification of potential tie-points using image correlation and feature matching algorithms before the bundle adjustment process. This preliminary action prepares a set of candidate tie-points that can be automatically validated during bundle adjustment, eliminating the need for manual tie-point selection while maintaining reliability
Solution Approach 2:
The system enables the bundle adjustment algorithm to automatically validate and select reliable tie-points from the preliminary candidate set through self-consistency checks. The algorithm independently verifies tie-point accuracy by checking geometric constraints and residuals, allowing the system to self-correct and identify reliable tie-points without manual intervention
3Productivity
If bundle adjustment is performed without accurate initial azimuth, then processing can proceed quickly, but the results are inaccurate and require reprocessing
Solution Approach 1:
The patent implements an iterative feedback mechanism within the bundle adjustment algorithm where initial azimuth estimates are continuously refined based on residual errors from tie-point observations. The algorithm uses feedback from each iteration to improve azimuth accuracy, allowing the process to start quickly with rough estimates and converge to precise results through multiple passes
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for solving the initial azimuth for a survey instrument, and other devices, with position and tilt information. The method may be part of bundle adjustment performed during the processing of image and position data collected from each camera of a survey instrument. The use of the initial azimuth generation method (and initial azimuth generator) makes it possible to get accurate azimuth orientation of a camera, such as each camera of a multi-camera survey instrument, without using data from a compass-like sensor. The initial azimuth generated by this method can then be used in later steps/processes of the bundle adjustment to find tie-points with an automatic tie-point finder. Prior to this method, the automatic tie-point finding algorithm relied on the compass and its accuracy, and inaccurate compass values would cause a complete failure, slow runtimes, or less accurate results for the bundle adjustment.