Meandering UAV Flight Path for Aerial Image Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Aerial photograph image pickup methods using small Unmanned Aerial Vehicles (UAVs) face challenges in achieving accurate three-dimensional terrain measurement due to image tilting caused by rolling or pitching, which results in distorted combined images when trying to overlap images taken during linear flights.
Innovation Solution
The method involves extracting feature points from images taken at multiple vertices, performing image tracking, and using a meandering flight pattern to correct tilting issues, allowing for accurate image alignment and combination through triplet matching and photogrammetry.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If linear flight path is used for aerial photography, then flight path is simple and easy to control, but image tilting occurs due to rolling or pitching of the UAV resulting in distorted combined images
Solution Approach 1:
The patent applies a meandering flight path pattern where the UAV flies in a curved or zigzag trajectory rather than a straight line. This curvature allows the optical axis to remain substantially perpendicular to the ground surface throughout the flight, preventing image tilting while maintaining operational simplicity. The meandering path ensures that even as the UAV moves forward, it periodically adjusts its heading to keep the camera oriented correctly, thus resolving the contradiction between easy control and precise image alignment.
2Manufacturing precision
If meandering flight pattern is used for aerial photography, then image tilting is corrected and alignment accuracy is improved, but flight path complexity increases
Solution Approach 1:
The patent implements preliminary action by pre-planning and pre-programming the meandering flight path before the aerial photography mission begins. The flight schedule is calculated in advance to ensure that the UAV follows the optimal curved trajectory that maintains perpendicular orientation of the optical axis to the ground. This preliminary preparation eliminates the need for complex real-time adjustments during flight, thus improving image alignment accuracy while keeping the operational complexity manageable through advance planning.
3Productivity
If images are taken during linear flight, then acquisition speed is fast, but tilting errors occur making accurate three-dimensional terrain measurement difficult
Solution Approach 1:
The patent applies dynamics by making the flight path adaptive and responsive to the terrain being photographed. The meandering pattern is dynamically adjusted based on the topography and the required measurement accuracy. The UAV continuously modifies its trajectory to maintain optimal imaging conditions while moving forward, ensuring that images are captured with correct orientation for accurate three-dimensional terrain measurement. This dynamic adjustment maintains both acquisition speed and measurement precision.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An aerial photographing image pickup method comprises a step of making a flying object, which comprises a GPS device 23 and an image pickup device 13 for taking an image downward, fly meanderingly, a step of taking the image at each vertex where a direction is changed in the meandering flight, a step of extracting feature points from a common overlap portion of the images taken from at least three vertexes adjacent to each other, a step of determining two images of two vertexes in the images as one set and acquiring positional information of the two vertexes by the GPS device for each set regarding at least two sets, a step of performing photogrammetry of the measuring points corresponding to the feature points based on the positional information and based on the feature points of the two images and a step of determining the feature points when the surveying results of the measuring points coincide with each other in at least the two sets as tie points for image combination.