Pixel Capture-Time Reliability in Multi-Camera Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing stereo camera systems face errors in distance estimation due to asynchronous image capturing by multiple cameras, leading to erroneous parallax and potential safety hazards, particularly in rolling shutter type cameras, and lack extensibility to configurations with three or more cameras.
Innovation Solution
An image processing device that estimates distances using multiple cameras, calculates image-capturing times, and selects the most reliable distance estimation based on the reliability of image-capturing times to minimize errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple cameras capture images simultaneously for distance measurement, then distance measurement capability is improved, but distance measurement precision deteriorates due to asynchronous capturing and parallax errors
Solution Approach 1:
The system performs preliminary actions by calculating image-capturing times for each pixel before distance determination, and selects the most appropriate distance estimation result based on these timing calculations. This preliminary timing analysis allows the system to choose the most reliable distance measurement while avoiding parallax errors from asynchronous capturing.
Solution Approach 2:
The system changes the parameter of image-capturing time by calculating specific capturing times for each pixel in multiple images and using this temporal parameter to select the optimal distance estimation. This parameter-based selection resolves the contradiction by choosing measurements taken at the most appropriate time points.
2Device complexity
If rolling shutter type cameras are used for image capture, then device complexity is reduced, but measurement precision deteriorates due to time delay between lines
Solution Approach 1:
The system performs preliminary calculation of image-capturing times for each pixel in the rolling shutter images, identifying which pixels were captured at the most appropriate time points. This preliminary timing analysis compensates for the line-by-line capture delay inherent in rolling shutters without requiring hardware changes.
Solution Approach 2:
The system creates a temporal map of when each pixel was captured, effectively copying timing information onto the image data. This allows the system to select distance measurements from pixels captured at optimal time points, compensating for the rolling shutter's sequential capture mechanism.
3Measurement precision
If parallax correction is applied to all pixel regions, then measurement precision is improved, but device complexity increases due to additional correction processing
Solution Approach 1:
Instead of applying uniform parallax correction to all regions, the system calculates image-capturing times for each pixel and selectively applies distance determination based on local timing characteristics. This local quality approach processes only the necessary regions with the appropriate method, reducing overall complexity while maintaining precision where needed.
Solution Approach 2:
The system uses image-capturing time as a selective parameter to determine which distance estimation results to use for each pixel region. This parameter-based selection avoids unnecessary correction processing in regions where it is not needed, reducing computational complexity while maintaining accuracy where timing differences are significant.
4Measurement precision
If stereo camera system is used for distance measurement, then measurement capability is improved, but adaptability deteriorates when extending to three or more cameras or other distance measurement devices
Solution Approach 1:
The system achieves universality by using image-capturing time as a common parameter for selecting distance estimation results, regardless of whether the distance is measured by stereo parallax, monocular cues, or other distance measurement devices. This time-based selection mechanism works across different camera configurations and measurement methods, enabling easy extension to three or more cameras or hybrid systems.
Solution Approach 2:
The system dynamically selects which distance estimation result to use based on calculated image-capturing times and reliability assessments. This dynamic selection approach allows the system to adapt to different camera configurations and measurement devices without requiring fixed architectural changes, as the selection logic automatically adjusts based on the available data and their timing characteristics.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An image processing device includes an image acquisition unit that acquires a plurality of images captured by each of a plurality of cameras, a first distance estimation unit that estimates a distance to an object using the plurality of images, a second distance estimation unit that estimates a distance to the object using at least one of the plurality of images, an image-capturing-time computation unit that obtains image-capturing times of at least some pixels included in the plurality of images, a reliability computation unit that obtains reliability of the plurality of images based on a difference between image-capturing times of the pixel in the plurality of images, and a distance determination unit that determines a distance to the object by using the distance estimated by at least one of the first distance estimation unit and the second distance estimation unit and the reliability.