Image Sensor Pixel Block Averaging for Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image sensing devices face challenges in efficiently interpolating pixel signals without compromising resolution, leading to suboptimal signal-to-noise ratio (SNR) and increased fixed pattern noise (FPN).
Innovation Solution
The proposed image sensing device employs a pixel array with shared pixel structures, where pixel blocks output pixel signals multiple times in specific color orders, and averaging blocks group and average these signals to generate digital signals, reducing noise and improving SNR while maintaining resolution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If pixel signals are interpolated using conventional methods, then the number of readout signals is reduced, but the signal-to-noise ratio deteriorates and fixed pattern noise increases
Solution Approach 1:
The pixel array is divided into multiple pixel blocks (e.g., 2x2 blocks), where each block contains pixels of the same color. This segmentation allows signals from multiple pixels to be read out through shared column lines and then averaged, reducing noise while maintaining readout efficiency. The segmentation creates independent processing units that can be handled separately through the averaging blocks.
Solution Approach 2:
Multiple pixel signals from different pixel blocks are merged and averaged together in the averaging blocks. By combining signals from N pixels (where N≥2) that share the same color information, the system achieves noise reduction through averaging while maintaining the same number of readout channels, thus improving SNR without compromising productivity.
2Reliability
If pixel blocks output signals multiple times in specific color orders, then noise is reduced through averaging, but device complexity increases
Solution Approach 1:
The column lines serve multiple functions: they transfer pixel signals from different pixel blocks and carry signals that will be averaged together. The averaging blocks also serve dual purposes by both averaging signals and preparing them for conversion. This multi-functionality reduces the need for dedicated signal paths, thereby limiting the increase in device complexity.
Solution Approach 2:
Pixel blocks output signals in periodic color orders (e.g., sequentially outputting signals for different colors in a repeating pattern). This periodic arrangement allows the averaging blocks to systematically group and average signals of the same color from multiple blocks, achieving FPN reduction through regular, predictable signal patterns that simplify the processing architecture.
3Measurement precision
If averaging blocks group pixel signals into overlapping groups, then signal-to-noise ratio is improved, but the number of processing blocks increases
Solution Approach 1:
Each averaging block processes a specific local group of pixel signals with particular characteristics (e.g., signals from specific pixel blocks or specific color channels). This local processing approach allows for optimized averaging within each block while maintaining overall system efficiency. The local quality principle ensures that each averaging block is tailored to its specific processing task, improving measurement precision without requiring a uniform increase in the number of blocks throughout the entire system.
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 enhances the signal-to-noise ratio (SNR) and reduces fixed pattern noise (FPN) by interpolating pixel signals effectively, ensuring improved image quality without resolution deterioration.
Implementation Method 1
Image sensing devices capture images using photosensitive properties of semiconductors
Data Source
AI summary
An image sensing device includes: a pixel array including a plurality of pixels arranged at each cross point of rows and columns, wherein the pixel array comprises a plurality of pixel blocks, each including N pixels, N being a natural number equal to or greater than 2, wherein the pixel blocks sequentially output a plurality of pixel signals having pixel information on the same color N times during one or more single row times; a plurality of column lines suitable for sequentially transferring the plurality of pixel signals from the pixel blocks, each column line being shared by two adjacent columns and coupled to at least one of the pixel blocks; a plurality of averaging blocks suitable for grouping the pixel signals to overlap each other, into a plurality of pixel signal groups, and averaging the pixel signal groups to output a plurality of averaged pixel signals, wherein the number of the averaging blocks is smaller than the number of the column lines; and a plurality of conversion blocks suitable for converting the averaged pixel signals into a plurality of digital signals.


