Digital Camera Noise Reduction via Multi-Frame Temporal Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional digital cameras face challenges in low-light conditions, where increased shutter time leads to motion blur and high ISO amplification results in noise, with existing anti-shake technologies either being expensive or increasing noise levels.
Innovation Solution
A method involving motion compensated temporal filtering (MCTF) that captures multiple frames at high ISO with short shutter times and combines them to reduce noise and blur, using a digital still camera system with a pre-processing circuit for digital gain correction, an MCTF circuit for temporal frame alignment, and a post-processing circuit for image sharpening and smoothing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If shutter time is increased to capture more light in low-light conditions, then illumination intensity is improved, but motion blur increases
Solution Approach 1:
The patent divides a single long exposure into multiple shorter sub-exposures captured at different time points. Each sub-exposure captures a portion of the total light while freezing motion better, and these segments are later combined through temporal filtering to achieve both brightness and sharpness.
Solution Approach 2:
The patent performs temporal filtering continuously across multiple captured frames to progressively reduce noise and blur while accumulating light information. This continuous processing allows the system to maintain image quality throughout the extended capture period.
2Illumination intensity
If high gain (high ISO) is used to amplify the picture signal, then illumination intensity is improved, but image noise increases
Solution Approach 1:
The patent combines multiple frames captured at high ISO through temporal filtering and averaging. This merging process reduces the random noise inherent in high ISO shots while preserving the amplified signal, effectively achieving brightness enhancement with reduced noise compared to single-frame high ISO capture.
Solution Approach 2:
The temporal filtering process continuously processes multiple high ISO frames to progressively reduce noise while maintaining the beneficial signal amplification. This continuous refinement allows the system to exploit high gain without permanently suffering from its noise penalty.
3Manufacturing precision
If mechanical anti-shake technology is used to counteract camera motion, then image sharpness is improved, but device complexity and cost increase
Solution Approach 1:
The patent replaces mechanical anti-shake mechanisms with a digital signal processing approach using temporal filtering. Instead of physically moving lenses or sensors to counteract motion, the system captures multiple frames and uses computational methods to eliminate blur, thereby reducing mechanical complexity while achieving similar or better results.
Solution Approach 2:
The patent creates multiple digital copies of the scene at different time points and processes these copies through temporal filtering. This digital copying and processing approach achieves motion compensation without requiring physical anti-shake mechanisms, simplifying the overall system.
4Device complexity
If digital anti-shake/anti-blur based on increased ISO is used, then device complexity is reduced, but image noise increases
Solution Approach 1:
The patent combines multiple high ISO frames through temporal filtering to reduce noise while maintaining the simplicity of a digital-only system. By merging information from multiple frames captured at high gain, the system achieves noise reduction without requiring mechanical components or reducing the ISO setting.
Solution Approach 2:
The continuous temporal filtering process progressively reduces noise from multiple high ISO frames while maintaining the benefits of high gain capture. This ongoing processing allows the system to use high ISO settings without permanently suffering from the associated noise increase.
Data Source
AI summary
A method of capturing a still frame is disclosed. The method generally includes the steps of (A) generating a plurality of initial frames with a sensor in response to an optical signal and (B) generating the still frame by combining the initial frames using a noise reduction technique.