Polarized Image Processing for Accurate Normal Line Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing methods fail to accurately separate and account for reflection components in polarized images, leading to inaccuracies in generating normal line information.
Innovation Solution
An image processing apparatus and method that utilizes an image pickup element with identical polarization pixel blocks and color filters to generate polarized images with high extinction ratios, allowing for the separation and extraction of reflection components, which are then used to generate highly accurate normal line information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional image processing methods are used to generate normal line information from polarized images, then the processing is simpler, but the accuracy of normal line information is insufficient due to inability to separate reflection components
Solution Approach 1:
The invention segments the polarized image into multiple reflection components (specular reflection, diffuse reflection, and substrate reflection) based on their different polarization characteristics. By dividing the image processing into component-specific processing steps, the system achieves high measurement precision for normal line information while managing complexity through systematic segmentation of the reflection types.
Solution Approach 2:
The invention introduces polarization degree as an intermediary parameter to differentiate and separate various reflection components. By calculating and utilizing the polarization degree of each pixel, the system can identify and process different reflection types (specular, diffuse, substrate) separately, enabling accurate normal line information extraction without requiring complex direct separation methods.
2Measurement precision
If polarized images are captured without considering reflection components, then the imaging process is faster, but the generated normal line information contains inaccuracies
Solution Approach 1:
The invention performs preliminary classification of pixels into different reflection component categories (specular, diffuse, substrate) based on their polarization characteristics before generating normal line information. This preliminary segmentation allows subsequent processing to be optimized for each component type, improving both accuracy and processing efficiency by avoiding unnecessary computations for each pixel.
Solution Approach 2:
The invention utilizes changes in polarization degree as a key parameter to differentiate reflection components. By monitoring and utilizing the polarization degree parameter across different image channels, the system can rapidly classify and process different reflection types, maintaining high productivity while achieving accurate normal line information extraction.
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
The solution enables the generation of highly accurate normal line information by effectively separating and accounting for reflection components, improving the accuracy of three-dimensional shape acquisition and object recognition.
Implementation Method 1
a polarized image with a plurality of polarization directions is generated using an image pickup element and a polarizer
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An imaging unit 20 has a configuration in which an identical polarization pixel block made up of a plurality of pixels with an identical polarization direction is provided for each of a plurality of polarization directions and pixels of respective predetermined colors are provided in the identical polarization pixel block. A correction processing unit 31 performs correction processing such as white balance correction on a polarized image generated by the imaging unit 20. A polarized image processing unit 32 separates or extracts a reflection component using the polarized image after the correction processing. By using a polarized image of the separated or extracted reflection component, for example, it is possible to generate normal line information with high accuracy.