Demosaicing Unit for High-Resolution Polarized Image Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image processing technologies face challenges in acquiring high-resolution information from polarized images, particularly when using color mosaic filters and polarizers, as they often result in reduced resolution for color luminance and polarization information, especially when the green component in photographic objects is small.

Innovation Solution

An image processing device and method that utilizes a demosaicing unit to calculate pixel signals for each polarization component by employing two-dimensional filtering and component information, allowing for the generation of high-resolution polarized images and non-polarized images without deteriorating resolution, and adjusts white balance to improve color difference calculations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If four types of polarizer units are arranged in pixels of the same color to acquire polarization information, then polarization information can be obtained, but the resolution of color luminance information and polarization information is reduced

Engineering Contradiction:
Improvepolarization information acquisitionVSAvoidcolor luminance resolution
Core Design Contradiction:
Measurement precisionVSManufacturing precision

Solution Approach 1:

The image sensor is divided into two distinct pixel groups: first pixels that capture only color luminance information without polarizers, and second pixels that capture both color luminance and polarization information with four types of polarizer units. This segmentation allows full-resolution color luminance imaging while dedicating specific pixels to polarization measurement, thereby avoiding the resolution degradation that would occur if all pixels contained polarizer units.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Different regions of the image sensor are assigned different functional qualities: first pixels are optimized for high-resolution color luminance capture, while second pixels are specialized for polarization information acquisition. This local differentiation ensures that color luminance information maintains high resolution in regions where it is captured, while polarization information is obtained where specifically needed, without compromising overall image quality.

Inventive Principle:
Principle #3Local quality

2Measurement precision

If the Bayer filter is used as a color mosaic filter and polarization information is generated using only green pixels, then polarization information can be acquired, but polarization information cannot be acquired when the green component in the photographic object light is small

Engineering Contradiction:
Improvepolarization information acquisitionVSAvoidpolarization information acquisition reliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The second pixels are designed to capture both color luminance information and polarization information simultaneously, rather than being dedicated solely to polarization measurement. This multi-functionality ensures that polarization information can be acquired from all color components (red, green, and blue), not just green, thereby maintaining reliable polarization data acquisition even when the green component of the photographed object is minimal.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system changes the parameter of pixel functionality from specialized (green-only polarization) to generalized (all-color polarization). By enabling second pixels to capture polarization information across all color components, the system adapts to varying color compositions in photographed objects, ensuring reliable polarization measurement regardless of which color component dominates the scene.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3376759B1Image processing device and image processing method
Publication Date: 2021.04.14 SONY GROUP CORP
  • EP3376759B1 patent drawingFigure 1~2
  • EP3376759B1 patent drawingFigure 3~4
  • EP3376759B1 patent drawingFigure 5(a)~5(b)

AI summary

The polarization imaging unit 21 generates a polarized image including pixels for each of a plurality of polarization components. The demosaicing unit 41 calculates a pixel signal for each polarization component by using the pixel signal of the target pixel of the polarized image and the pixel signal of the pixel for each of the identical polarization components located near the target pixel. In one example, a low frequency component is calculated for each polarization component by using the pixel signal of the pixel located near the target pixel for each of the identical polarization components. In addition, component information indicating the relationship between the low frequency component of the polarization component of the polarized image and the pixel signal of the target pixel is acquired. Furthermore, the pixel signal for each polarization component in the target pixel is calculated on the basis of the low frequency component and the component information for each polarization component. The pixel signal for each polarization component is obtained for each pixel of the polarized image, so it is possible to acquire normal information or a non-polarized light image with high resolution.