Rolling Shutter Distortion Estimation in Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing devices struggle to accurately calculate the rolling shutter distortion component, especially when multiple subjects with different movements are present in an imaging scene, leading to errors in global motion vector modeling and distortion component calculation.
Innovation Solution
An image processing device that estimates rolling shutter distortion by acquiring a motion vector using successive frame images, calculating a distortion coefficient based on camera motion and pixel sensor settings, and applying correction parameters to correct for distortion, using a combination of hardware components like a gyro sensor for accurate movement detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a global motion vector model is used to calculate rolling shutter distortion, then the distortion component can be estimated, but errors occur when multiple subjects with different movements are present
Solution Approach 1:
The patent segments the global motion vector into multiple independent components: camera motion component, subject motion component, and rolling shutter distortion component. By separating these components, the model can accurately attribute movements to their respective sources even when multiple subjects are present, preventing errors in distortion calculation that occur when using a unified global motion vector.
2Adaptability or versatility
If multiple components of the global motion vector are used to calculate distortion components, then enlargement and reduction effects can be captured, but errors in distortion calculation increase
Solution Approach 1:
The patent extracts and isolates the rolling shutter distortion component from the global motion vector by identifying and removing contributions from camera motion and subject motion. This extraction process allows the model to capture distortion effects without the errors that arise from including multiple motion components in the calculation.
3Adaptability or versatility
If a complex model with multiple motion components is used, then comprehensive motion analysis is achieved, but calculation stability decreases
Solution Approach 1:
The patent applies local quality by treating different motion components with different analytical approaches. The camera motion component is analyzed using gyro sensor data, subject motion is handled through object detection algorithms, and rolling shutter distortion is calculated based on the remaining differential motion. This localized treatment of each component maintains calculation stability while achieving comprehensive motion analysis.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution enables stable and accurate estimation and correction of rolling shutter distortion, even in complex scenes with multiple moving subjects, by simplifying the calculation process and considering specific camera and environmental factors, thereby improving image quality and reducing errors.
Implementation Method 1
a gyro sensor 24 for detecting movement of the camera 20
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An image processing device (1) includes an image input portion (10), a motion vector acquisition portion (11) and a distortion component estimation portion (12). The image input portion (10) inputs a plurality of frame images. The motion vector acquisition portion (11) acquires an amount of parallel movement between two frame images. The distortion component estimation portion (12) performs approximation such that a motion matrix indicating a motion between the frame images consists only of a parallel movement component and a rolling shutter distortion component, uses the amount of parallel movement to calculate, for each of the frame images, a tentative distortion coefficient included in the rolling shutter distortion component and uses the plurality of tentative distortion coefficients calculated to estimate a distortion coefficient.