Linear-Logarithmic Image Sensor FPN Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image sensors face challenges in reducing fixed pattern noise (FPN) due to characteristic differences among unit pixels, leading to deviations in analog signals generated in response to identical incident light, which affects image quality.
Innovation Solution
A linear-logarithmic image sensor is designed with a pixel array that generates leakage signals and two types of analog signals based on transfer control signals, and a signal generation unit that selects appropriate final analog signals for different illuminance regions, effectively reducing FPN by dividing the signal generation region into linear, extended linear, and logarithmic regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a linear image sensor is used to output electric signals linearly proportional to incident light, then measurement precision is improved for weak light, but fixed pattern noise increases in strong light conditions
Solution Approach 1:
The image sensor dynamically switches between linear and logarithmic conversion modes based on incident light intensity. The control unit adjusts the conversion characteristic in real-time, transitioning from linear conversion for weak light to logarithmic conversion for strong light, thereby adapting the system's response to varying illumination conditions and reducing fixed pattern noise across different light levels
Solution Approach 2:
The patent changes the conversion parameter from linear to logarithmic based on light intensity thresholds. The control unit monitors incident light levels and adjusts the conversion characteristic parameter, using linear conversion when light is weak and logarithmic conversion when light is strong, thus optimizing measurement precision while minimizing fixed pattern noise
2Object-affected harmful factors
If a logarithmic image sensor is used to output electric signals logarithmically proportional to incident light, then fixed pattern noise is reduced in strong light, but measurement precision deteriorates for weak light
Solution Approach 1:
The image sensor dynamically switches between linear and logarithmic conversion modes based on incident light intensity. The control unit adjusts the conversion characteristic in real-time, transitioning from linear conversion for weak light to logarithmic conversion for strong light, thereby adapting the system's response to varying illumination conditions and reducing fixed pattern noise across different light levels
Solution Approach 2:
The patent changes the conversion parameter from linear to logarithmic based on light intensity thresholds. The control unit monitors incident light levels and adjusts the conversion characteristic parameter, using linear conversion when light is weak and logarithmic conversion when light is strong, thus optimizing measurement precision while minimizing fixed pattern noise
3Adaptability or versatility
If multiple transfer control signals are used to sequentially transfer photo-charges, then signal generation flexibility is improved, but device complexity increases
Solution Approach 1:
The transfer of photo-charges is segmented into multiple sequential stages controlled by different transfer control signals. The pixel array first transfers photo-charges to an intermediate node, then selectively transfers them to either a linear signal generation node or a logarithmic signal generation node based on light intensity. This segmentation enables flexible signal generation while managing control complexity through structured, multi-stage transfer operations
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 effectively reduces FPN caused by threshold voltage distribution of transfer transistors, resulting in improved image quality by using different final analog signals for varying light conditions.
Implementation Method 1
photo-charges that are generated in response to incident light by a photoelectric conversion device
Data Source
AI summary
A linear-logarithmic image sensor includes a pixel array, a signal generation unit, and a control unit. The pixel array includes at least one unit pixel that generates a leakage signal corresponding to leakage photo-charges and that sequentially generates a first analog signal corresponding to a portion of accumulated photo-charges and a second analog signal corresponding to a whole of the accumulated photo-charges by resetting a floating diffusion node and transferring the accumulated photo-charges from a storage node to the floating diffusion node in response to first and second transfer control signals that are sequentially activated. The signal generation unit includes at least one signal generation block that generates a final analog signal based on the leakage signal, the first analog signal, and the second analog signal. The control unit controls the pixel array and the signal generation unit.