Dual Conversion Gain Image Sensor Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
CMOS image sensors face challenges in noise reduction, particularly in low-illuminance environments where exposure time is restricted, leading to inadequate image brightness and noise amplification.
Innovation Solution
Implementing a dual conversion gain mechanism in image sensors, where pixels generate both low and high conversion gain signals, and an image signal processor normalizes and blends these signals to produce a noise-reduced output image, improving signal-to-noise ratio.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Illumination intensity
If exposure time is extended to increase image brightness in low-illuminance environments, then image brightness is improved, but motion blur and noise increase
Solution Approach 1:
The patent changes the conversion gain parameter dynamically based on illuminance conditions. In low-illuminance environments, high conversion gain is applied to amplify weak signals without requiring extended exposure time, thereby maintaining image brightness while avoiding motion blur and noise accumulation associated with long exposures.
Solution Approach 2:
The system dynamically switches between low conversion gain and high conversion gain modes based on real-time illuminance detection. This dynamic adaptation allows the image sensor to optimize signal amplification according to lighting conditions, resolving the contradiction between maintaining brightness and controlling noise in varying environments.
2Power
If conversion gain is increased to amplify weak signals in low-illuminance environments, then signal amplification is improved, but noise amplification increases
Solution Approach 1:
The patent applies different conversion gain values to different pixel regions or different time periods based on local signal characteristics. By selectively applying high conversion gain only where and when needed (in low-illuminance conditions), the system amplifies useful signals while minimizing noise amplification in well-lit regions or time periods.
Solution Approach 2:
The system uses feedback from illuminance detection and signal quality assessment to dynamically adjust conversion gain. This feedback mechanism allows the system to monitor noise levels and signal strength, adjusting amplification in real-time to maintain optimal signal-to-noise ratio, thereby amplifying signals without proportionally amplifying noise.
3Reliability
If dual conversion gain processing is implemented to reduce noise, then signal-to-noise ratio is improved, but device complexity increases
Solution Approach 1:
The patent segments the image processing into distinct stages: low conversion gain processing for well-lit regions/time periods and high conversion gain processing for low-illuminance regions/time periods. This segmentation allows independent optimization of each processing path and simplifies the overall control logic by using clear conditional branching based on illuminance thresholds.
Solution Approach 2:
The image sensor and processing system are designed to perform multiple functions using the same hardware infrastructure. The dual conversion gain mechanism uses a single pixel array and signal processing chain that can operate in both low and high gain modes, eliminating the need for separate hardware systems and reducing overall device complexity despite the enhanced processing capabilities.
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
This approach effectively reduces noise in images captured in low-illuminance conditions by enhancing the signal-to-noise ratio and maintaining image brightness, applicable to both still images and video frames.
Implementation Method 1
The CMOS image sensor includes pixels composed of CMOS transistors and converts light energy into an electrical signal by using a photoelectric conversion element (or device) included in each pixel
Data Source
AI summary
Disclosed is a device for noise reduction using dual conversion gain, which includes an image sensor including a pixel array, the pixel array configured to generate a first pixel signal corresponding to a first conversion gain and a second pixel signal corresponding to a second conversion gain from pixels sharing a floating diffusion region and the image sensor configured to generate first image data and second image data based on the first pixel signal and the second pixel signal, and an image signal processor that generates an output image based on the first image data and the second image data. The image signal processor includes a normalization circuit that normalizes the first image data based on a dynamic range of the second image data to generate third image data, and a blending circuit that generates the output image based on the second image data and the third image data.