Motion Invariant Camera Lens Parabolic Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional cameras fail to capture all spatial frequencies in moving scenes, leading to motion blur and loss of texture, especially when objects are in motion or the camera itself moves during exposure, necessitating improved methods for motion de-blurring that do not rely on image-based blur estimation.
Innovation Solution
A motion invariant camera system that stabilizes the lens with parabolic motion during exposure, using existing optical components and sensors to encode all objects with the same Point Spread Function (PSF), allowing for de-convolution without blur estimation by translating the lens to match object velocities, thereby preventing motion blur and reducing ghosting artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional cameras capture images during exposure, then the imaging process is simple, but motion blur occurs and spatial frequencies are lost when objects or camera move
Solution Approach 1:
The patent implements dynamic lens motion during exposure, where the lens is moved according to a predetermined trajectory (e.g., circular or linear paths) to counteract the relative motion between the camera and moving objects. This dynamic adjustment of the lens position allows the system to capture sharp images of moving objects by matching the lens motion to the object motion, thereby resolving the contradiction between image sharpness and exposure time without requiring complex post-processing deblurring algorithms
Solution Approach 2:
The patent employs preliminary calibration to establish the relationship between camera motion and optimal lens compensation trajectories. By pre-determining the lens motion paths based on expected object velocities and directions, the system prepares the compensation mechanism before actual image capture. This preliminary setup enables the lens to automatically follow the correct trajectory during exposure, achieving motion deblurring without real-time complex calculations or post-processing
2Illumination intensity
If exposure time is increased to capture moving objects, then more light is captured, but motion blur increases
Solution Approach 1:
The system dynamically adjusts lens position during exposure based on the predetermined trajectory, allowing longer exposure times to capture more light while maintaining image sharpness. The lens motion continuously compensates for object movement, ensuring that light from moving objects is consistently focused on the corresponding sensor regions throughout the extended exposure period, thus resolving the trade-off between light capture and image sharpness
3Manufacturing precision
If blur estimation is performed to de-blur images, then motion deblurring is achieved, but ghosting artifacts occur and texture is lost
Solution Approach 1:
The patent replaces the computational deblurring process (software-based blur estimation and deconvolution) with a mechanical/optical solution: physically moving the lens during exposure to prevent blur formation in the first place. By substituting the computational approach with direct optical compensation, the system avoids the inherent limitations of deconvolution methods, including ghosting artifacts and texture loss, while achieving superior motion deblurring performance
Solution Approach 2:
The patent converts the harmful effect of object motion during exposure into a beneficial outcome by using the same motion to drive the lens compensation trajectory. The lens follows the object's motion path, transforming what would normally cause blur into a mechanism for maintaining focus, thereby eliminating the need for post-processing deblurring and avoiding associated artifacts
4Loss of information
If coded exposure is used to capture all spatial frequencies, then texture is preserved, but the captured image requires numerically stable de-convolution which needs blur estimation
Solution Approach 1:
The patent replaces the coded exposure approach with direct optical compensation through lens motion. Instead of encoding scene information in a format requiring deconvolution, the system physically prevents blur formation by moving the lens to track moving objects during exposure. This mechanical substitution eliminates the need for numerically stable de-convolution and blur estimation, preserving texture while simplifying processing
Data Source
AI summary
Motion de-blurring systems and methods are described herein. One motion de-blurring system includes an image sensing element, one or more motion sensors in an imaging device, a lens element that undergoes motion during a capture of an image by the sensing element, and a de-blurring element to de-blur the image captured by the sensing element via de-convolving a Point Spread Function (PSF).