Focus Detection Pixel Segmentation for Phase Difference Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing focus detection systems using image sensors for phase-difference detection suffer from reduced accuracy due to low similarity between image signals generated from different pixel groups, leading to errors in focus detection, especially when the arrangement interval between focus detection pixels is large, resulting in incomplete capture of high-frequency components and generation of noise.
Innovation Solution
A focus detection apparatus comprising specific pixel configurations and computation units that compute phase differences between image signals from strategically positioned pixels to improve defocus amount calculation, including a first pixel detecting a partial area, a second pixel detecting a different partial area, a third pixel detecting the entire area, and a fourth pixel also detecting the entire area but positioned differently, with computation units summing phase differences to determine the defocus amount.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If focus detection pixels are arranged with large interval to reduce pixel count and manufacturing cost, then device complexity is reduced, but measurement precision degrades due to loss of high-frequency components and generation of folding noise
Solution Approach 1:
The patent divides the photoelectric conversion unit into multiple independent photoelectric conversion regions (first, second, third, and fourth regions) within a single pixel. Each region captures light flux from different exit pupil areas, enabling the pixel to function as multiple focus detection pixels simultaneously. This segmentation allows fine-grained sampling of high-frequency components without requiring physically separated pixels, thus resolving the contradiction between device complexity and measurement precision.
Solution Approach 2:
The patent transitions from spatial arrangement (arranging separate pixels for focus detection) to dimensional arrangement (dividing the photoelectric conversion unit into multiple regions with different exit pupil reception characteristics). By utilizing the internal dimensional structure of a single pixel rather than external pixel placement, the system achieves high-frequency component capture while maintaining simple device architecture.
2Measurement precision
If the photoelectric conversion unit of each pixel is divided into multiple areas to improve focus detection accuracy, then measurement precision is improved, but device complexity increases due to additional readout processing and calculation operations
Solution Approach 1:
The patent merges the functions of multiple focus detection pixels into a single pixel by dividing its photoelectric conversion unit into multiple regions. Instead of requiring separate pixels for first and second image signals, the single pixel's divided regions generate both signals simultaneously. This merging reduces the number of readout operations and calculation processes while maintaining the ability to compute accurate phase differences, thus resolving the contradiction between measurement precision and device complexity.
3Measurement precision
If focus detection pixels are positioned to capture different exit pupil areas to enable phase difference detection, then measurement precision is improved, but reliability degrades when object image patterns reduce similarity between captured signals
Solution Approach 1:
The patent applies local quality by creating different photoelectric conversion regions within a single pixel, each optimized to receive light flux from specific exit pupil areas. The first region captures light from one exit pupil area while the second region captures light from another area, allowing the single pixel to provide diverse sampling information. This local differentiation ensures reliable phase difference detection across various object patterns by maintaining signal similarity through strategic regional positioning.
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 enhances focus detection accuracy by reducing noise and improving the capture of high-frequency components, leading to more precise defocus amount calculations and improved focus detection performance.
Implementation Method 1
an image sensor comprising a plurality of pixels, each pixel having a photoelectric conversion unit that converts optical energy to electrical energy
Data Source
AI summary
A first phase difference between a first image signal output from a first pixel configured to detect a light flux having passed through a first partial area of the exit pupil, and a second image signal output from a third pixel configured to detect a light flux having passed through the entire exit pupil is computed. A second phase difference between a third image signal output from a second pixel configured to detect a light flux having passed through a second partial area of the exit pupil, and a fourth image signal output from a fourth pixel arranged at a position different from the third pixel and configured to detect a light flux having passed through the entire exit pupil is computed. The defocus amount of the imaging optical system is computed by using the sum of the first and second phase differences.