PDAF Pixel Layout With Multi-Orientation Edge Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image capture devices with phase detection auto-focus (PDAF) pixels face limitations in PDAF performance, particularly in low light conditions and images with few vertical edges, due to the sparse distribution of metal shield pixels and the reliance on vertical edge detection, which degrades autofocus accuracy and increases processing burdens.
Innovation Solution
The implementation of an image capture device with a 4×4 subset of pixels divided into quadrants, each with a Bayer pattern color filter and on-chip lenses (OCLs), allowing for PDAF information to be obtained from multiple orientations, reducing the need for signal correction, and enhancing autofocus performance by detecting both vertical and horizontal edges.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If metal shield pixels are sparsely distributed for PDAF, then device complexity is reduced, but PDAF performance degrades in low light conditions and images with few vertical edges
Solution Approach 1:
The pixel array is segmented into different functional regions: standard color pixels for image capture and PDAF pixels with metal shields for phase detection. This segmentation allows simultaneous optimization of both imaging and autofocus functions without requiring all pixels to serve dual purposes, thereby maintaining PDAF accuracy while managing device complexity.
Solution Approach 2:
Different regions of the pixel array are assigned different qualities and functions. PDAF pixels are strategically positioned at specific locations (e.g., edges of the pixel array) where they provide phase detection capability, while central regions maintain standard color pixel functionality. This local differentiation optimizes PDAF performance in critical areas without unnecessarily increasing complexity across the entire array.
2Device complexity
If PDAF pixels rely on vertical edge detection, then device complexity is minimized, but PDAF performance deteriorates in images with few vertical edges
Solution Approach 1:
The PDAF pixel structure is designed to detect edges in multiple orientations (vertical, horizontal, and diagonal) rather than being limited to vertical edges only. The metal shield configuration and signal processing methodology enable these pixels to function universally across different edge orientations, allowing the system to adapt to various image content types without increasing hardware complexity.
Solution Approach 2:
The system dynamically adjusts detection parameters and signal processing methods based on the detected edge orientation. By changing the analytical parameters rather than the physical structure, the system maintains versatility across different image types while avoiding the complexity of multiple specialized detection mechanisms.
3Measurement precision
If signal correction is extensively applied to improve PDAF performance, then PDAF accuracy is improved, but processing burden increases
Solution Approach 1:
The metal shield pixels are pre-configured with specific geometric arrangements and optical properties that facilitate direct phase detection without requiring extensive post-capture signal correction. The preliminary design of the pixel structure and metal shield positioning enables the system to obtain accurate PDAF measurements directly from the captured data, reducing the need for computationally intensive correction algorithms.
4Measurement precision
If PDAF pixels are densely distributed, then PDAF performance is improved, but image resolution is reduced due to fewer pixels available for image capture
Solution Approach 1:
The pixel array is segmented into distinct functional zones: PDAF pixels are concentrated at specific locations (such as array edges) where they provide phase detection without interfering with the central imaging region. This spatial segmentation allows dense PDAF pixel distribution in critical areas while maintaining sufficient pixel density for high-resolution image capture in other areas, resolving the trade-off between PDAF precision and image resolution.
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 configuration improves PDAF performance by providing more comprehensive edge detection, reducing signal noise, and increasing image resolution, especially in low light conditions, while minimizing the processing burden and image resolution loss.
Implementation Method 1
each pixel including a photodetector
Implementation Method 2
A set of 1×1 on-chip lenses (OCLs) may include a different 1×1 OCL disposed over each pixel
Data Source
AI summary
An image capture device is described. The image capture device includes an array of pixels, each pixel including a photodetector. A Bayer pattern color filter is disposed over a 4×4 subset of pixels in the array of pixels. The Bayer pattern color filter defines a first 2×2 subset of pixels sensitive to red light; a second 2×2 subset of pixels sensitive to green light; a third 2×2 subset of pixels sensitive to green light; and a fourth 2×2 subset of pixels sensitive to blue light. A set of 1×1 on-chip lenses (OCLs) includes a different 1×1 OCL disposed over each pixel in the second 2×2 subset of pixels and the third 2×2 subset of pixels. A set of 2×1 OCLs or 2×2 OCLs includes a 2×1 OCL or a 2×2 OCL disposed over each pixel in the first 2×2 subset of pixels and the fourth 2×2 subset of pixels.