Image Processing Apparatus Polarization Light Source Dependency Area Division
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing techniques fail to accurately separate specular and diffuse reflection components due to insufficient precision, especially in areas with large emittance angles, leading to inaccurate model-based image synthesis and texture misrepresentation in digital archives and augmented reality applications.
Innovation Solution
An image processing method that utilizes both light source condition dependency and polarization information to accurately separate specular and diffuse reflection areas by changing the polarization state of the illumination and using a patterned polarizer to capture polarization images, allowing for precise area division and parameter estimation in model-based image synthesis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional area division based solely on polarization information is used, then the separation process is simple, but the separation precision deteriorates in areas with large emittance angles
Solution Approach 1:
The invention changes the parameter basis for area division from solely polarization information to a combination of polarization information and light source condition dependency. By introducing light source condition dependency as an additional parameter, the system achieves more accurate separation in occluding edge areas without significantly increasing overall system complexity.
Solution Approach 2:
The invention introduces light source condition dependency as an intermediary parameter that mediates between the polarization information and the final area division result. This intermediary helps resolve the ambiguity in occluding edge areas by providing additional discriminative information about the reflection type.
2Ease of manufacture
If polarization information alone is used for area division, then the method is simple to implement, but texture representation accuracy deteriorates in synthesized images
Solution Approach 1:
The invention enhances the parameter set used for area division by incorporating light source condition dependency alongside polarization information. This parameter enhancement leads to more accurate texture representation in synthesized images while maintaining reasonable implementation complexity through systematic processing.
Solution Approach 2:
The invention uses light source condition dependency as a feedback mechanism to improve area division accuracy. By analyzing how polarization characteristics change with light source conditions, the system refines its classification of specular and diffuse reflection areas, leading to better texture representation in synthesized images.
3Productivity
If conventional area division is used, then processing speed is maintained, but image synthesis accuracy deteriorates due to incorrect area classification
Solution Approach 1:
The invention performs preliminary calculation of light source condition dependency before the final area division step. By pre-computing this parameter from the captured images, the system prepares accurate classification data in advance, ensuring both high processing speed and high image synthesis accuracy without requiring complex real-time calculations.
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 approach enables faithful image synthesis by accurately distinguishing between specular and diffuse reflection components, improving the precision of area division and reducing texture discrepancies in synthesized images, thereby enhancing the quality of digital archives and augmented reality applications.
Implementation Method 1
an illumination section for emitting linearly-polarized light onto an object while changing a polarization state of the linearly-polarized light
Implementation Method 2
a polarization information obtaining section for obtaining polarization information of the object
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A polarization image capturing section 103 obtains a first polarization image containing polarization information. A polarization information obtaining section 104 obtains first polarization information from the first polarization image. An illumination control section 102 changes an illumination section 101 and again obtains a second polarization image by means of the polarization image capturing section 103. The polarization information obtaining section 104 obtains second polarization information from the second polarization image. A light source dependency estimation section 105 obtains light source dependency from the first and second polarization information obtained from the first and second polarization images. An area dividing section 106 performs an area division by using the polarization information and light source dependency information.