UAV Dual-Camera Pose Compensation for Accurate Local Region Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image capturing systems on flying vehicles face challenges in maintaining accurate image focus and zoom due to time delays in signal transmission and environmental factors like wind and mechanical vibrations, leading to shifts in the UAV's pose.
Innovation Solution
An image capturing method and apparatus that calculates relative translation and rotation relationships between two image capturing devices on a flying vehicle, allowing for precise adjustment of the second device to capture a local region image by converting three-dimensional coordinate values, and includes a self-stabilized device to compensate for pose changes caused by environmental factors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If image transmission technology is integrated for remote inspection, then the ability to capture images from suitable viewing angles is improved, but time delay in signal transmission causes the UAV pose to shift due to environmental factors, resulting in image quality degradation
Solution Approach 1:
The system captures a first image at a first time point before receiving the control signal at a second time point. By performing the initial image capture in advance, the system establishes a reference state that can be used to compensate for subsequent pose changes during the transmission delay period.
Solution Approach 2:
The processing unit calculates relative translation and rotation relationships between the first and second image capturing devices based on the time delay and environmental factors. This feedback mechanism allows the system to adjust and compensate for pose shifts, maintaining accurate image focus and zoom despite transmission delays.
2Duration of action of moving object
If the UAV stays in the air for extended periods to capture images, then the ability to maintain stable viewing angles is improved, but mechanical vibrations and wind cause continuous pose changes, leading to image instability
Solution Approach 1:
The system performs preliminary calculation of the relative translation and rotation relationships between image capturing devices at different time points. By establishing this reference framework in advance, the system can continuously compensate for vibrations and wind effects throughout the extended inspection duration.
Solution Approach 2:
The processing unit dynamically adjusts parameters such as translation amount and rotation amount based on environmental conditions and time delays. These parameter changes enable the system to adapt to continuous pose variations caused by wind and vibrations, maintaining image stability throughout the inspection.
3Measurement precision
If coordinate transformation is performed to adjust for pose changes, then the accuracy of local region capture is improved, but the calculation complexity between multiple image capturing devices increases
Solution Approach 1:
The coordinate transformation is divided into separate components: translation relationship calculation and rotation relationship calculation. By segmenting the transformation process, the system handles each aspect independently, reducing overall complexity while maintaining accuracy in local region capture.
Solution Approach 2:
The processing unit acts as an intermediary that performs the coordinate transformation calculations between the first and second image capturing devices. This centralized intermediary approach simplifies the system architecture by consolidating the complex calculation logic in a single component rather than distributing it across multiple devices.
Data Source
AI summary
An image capturing method and an image capturing apparatus are provided. An inspection region image of an inspection region is captured by a first image capturing device at a first time point. A control signal is received at a second time point after the first time point. A relative translation relationship and/or a relative rotation relationship of the first image capturing device at the first time point and a second image capturing device at a second time point are calculated. A first three-dimensional coordinate value of the local region relative to the first image capturing device is calculated and the first three-dimensional coordinate value is transformed to a second three-dimensional coordinate value relative to the second image capturing device according to the relative translation relationship and the relative rotation relationship. The second image capturing device is driven to rotate. A local region image of the local region is captured.