Divisional Pixel Signal Processing for Autofocus Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing devices for image sensors with multiple photoelectric conversion elements per pixel face challenges in efficiently processing and utilizing the phase difference information from divisional pixel signals, which affects autofocus accuracy and image quality.
Innovation Solution
An image processing device with a Bayer layout of color filters, microlenses, and photoelectric conversion elements that generates Bayer-type divisional image data from divisional pixel signals, allowing for advanced image processing and phase difference detection, including low-light shading correction and phase difference autofocus.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple photoelectric conversion elements are provided per pixel to detect phase difference information, then autofocus accuracy is improved, but device complexity increases
Solution Approach 1:
The pixel is divided into multiple photoelectric conversion elements (first, second, third, and fourth photoelectric conversion elements) arranged in a specific pattern. Each element captures light from different directions, enabling phase difference detection through segmentation of the pixel structure.
Solution Approach 2:
Different photoelectric conversion elements are positioned to receive light from specific directions (e.g., first and second elements for one direction, third and fourth for another direction). This local specialization allows efficient phase difference measurement while maintaining manageable complexity.
2Manufacturing precision
If multiple divisional pixel signals are processed to generate Bayer-type divisional image data, then image quality is improved, but processing complexity increases
Solution Approach 1:
The image processing separates divisional pixel signals into different Bayer-type divisional image data groups based on their directional characteristics. This segmentation allows independent processing of each group, simplifying the overall processing complexity while maintaining high image quality.
Solution Approach 2:
The processing circuit performs selective processing on divisional pixel signals, focusing computational resources on generating specific Bayer-type divisional image data that is most critical for image quality, rather than processing all signals equally.
3Illumination intensity
If advanced image processing is performed on divisional image data, then low-light performance is improved, but processing time increases
Solution Approach 1:
The image processing circuit performs preliminary processing on divisional pixel signals to generate Bayer-type divisional image data in advance, organizing the data structure optimally before final image generation. This preliminary organization reduces processing time during low-light conditions when speed is critical.
Solution Approach 2:
The processing adapts parameters based on lighting conditions, adjusting the processing intensity and methods for divisional image data. In low-light conditions, optimized parameter settings enable effective processing with reduced time loss.
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
Enhances autofocus accuracy and image quality by effectively processing divisional pixel signals, improving phase difference detection and low-light performance while maintaining high frame rates.
Implementation Method 1
photoelectric conversion elements of m×n which generate divisional pixel signals of m×n respectively
Data Source
AI summary
An image processing device includes an image data generation circuit configured to generate n pieces of Bayer-type divisional image data configured by divisional pixel signals having the same divisional arrangement from divisional pixel signals of m×n generated by photoelectric conversion elements at n parts into which each of m pixels is divided, and an image processing circuit configured to perform image processing on the n pieces of Bayer-type divisional image data.