PDAF Pixel Correction Using Color Variance for Low-Noise Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing devices face challenges in accurately correcting Phase Detection Auto Focus (PDAF) pixel values, which are treated as defect pixels, leading to noise and imbalance in images, particularly in regions with low complexity.
Innovation Solution
An image processing device and method that determines a target pixel set based on average and variance values of pixel colors, converts PDAF pixel values to normal pixel values, and corrects them based on adjacent pixel values to match the arrangement pattern of the kernel, thereby improving accuracy and reducing noise.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If PDAF pixels are used for focus detection, then auto focus capability is improved, but image quality deteriorates due to noise and imbalance from defect pixels
Solution Approach 1:
The image processing device segments PDAF pixels from normal pixels and processes them separately through dedicated correction circuits. PDAF pixel values are individually corrected based on their specific defect characteristics before being combined with normal pixels for final image output, allowing focus detection functionality to be preserved while eliminating their harmful effects on image quality.
Solution Approach 2:
A correction circuit acts as an intermediary between PDAF pixels and the final image output. This intermediary process converts PDAF pixel values to match surrounding normal pixel values using various correction methods (spatial interpolation, frequency domain filtering, or machine learning-based correction), thereby mediating the conflict between maintaining PDAF functionality and ensuring image quality.
2Object-affected harmful factors
If PDAF pixel values are corrected using conventional methods, then some noise reduction is achieved, but accuracy remains insufficient particularly in low complexity regions
Solution Approach 1:
The correction circuit dynamically changes correction parameters based on local image characteristics. By analyzing variance and complexity metrics of pixel regions, the system adapts correction strength and method - applying stronger correction in low complexity regions where accuracy is most needed, while adjusting for high complexity regions to preserve edge details and avoid over-correction artifacts.
Solution Approach 2:
The correction process incorporates feedback mechanisms where correction results are evaluated against surrounding pixel values and image characteristics. The system uses feedback from variance calculations and complexity assessments to iteratively refine correction parameters, ensuring high accuracy in low complexity regions while maintaining overall image fidelity.
Data Source
AI summary
An image processing device may include: a target pixel set determiner for determining a target pixel set, among a plurality of Phase Detection Auto Focus (PDAF) pixel sets, based on average values and variance values of pixel values output from a plurality of pixels corresponding to a kernel set based on each of the plurality of PDAF pixel sets, wherein the average values and the variance values respectively correspond to colors of color filters included in the plurality of pixels; and a pixel value corrector for converting pixel values of the target pixel set into first pixel values corresponding to a predetermined color and correcting the first pixel values to second pixel values corresponding to an arrangement pattern of the plurality of pixels based on pixel values of a plurality of adjacent pixels adjacent to the target pixel set.