Polarization Imaging Normal Calculation for Mobile Occlusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for calculating the normal of an object surface using polarization information often face indeterminacy due to occlusion and positional changes when imaging devices are mobile, leading to incomplete or inaccurate normal calculations.
Innovation Solution
An information processing apparatus and method that generates normal candidate information for each pixel based on polarization images in multiple directions, selects relevant pixels, and calculates the normal using zenith and azimuth angles, or their combinations, to determine the plane's orientation, thereby addressing indeterminacy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple polarization imaging units are used to capture images from different positions, then indeterminacy of normal calculation is reduced, but device complexity and cost increase
Solution Approach 1:
The patent segments the polarization imaging process into multiple sequential captures with the same imaging unit, taking images at different positions along the optical axis. This divides the function of multiple simultaneous imaging units into temporal segments, achieving the same effect with a single unit.
Solution Approach 2:
The patent performs preliminary positioning and focusing adjustments before capturing polarization images at different positions. By pre-establishing the correct imaging geometry and focal planes, the system ensures accurate normal calculation without requiring complex real-time adjustments or multiple synchronized imaging units.
2Loss of information
If images are captured from different positions to resolve occlusion, then normal information completeness improves, but positional alignment accuracy becomes more difficult to maintain
Solution Approach 1:
The patent creates a virtual equipotential surface by calculating what the image positions would be if all views were taken from the same reference position. This virtual alignment eliminates the effects of actual positional variations, allowing accurate normal calculation even when images are captured from different locations.
Solution Approach 2:
The patent introduces a virtual reference position as an intermediary concept. All image positions are mathematically transformed to reference this virtual point, serving as a common coordinate system that reconciles images captured from different actual positions and eliminates alignment errors.
3Device complexity
If a single imaging unit is used to reduce device complexity, then indeterminacy clearance capability deteriorates due to occlusion
Solution Approach 1:
The patent makes the imaging system dynamic by capturing images at multiple positions along the optical axis sequentially. This temporal dynamics compensates for the spatial limitation of a single imaging unit, allowing the system to overcome occlusion by observing the target from different depths while maintaining simple hardware configuration.
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 allows for accurate and efficient calculation of normals with cleared indeterminacy, even when imaging devices are mobile, without the need for multiple polarization imaging units.
Implementation Method 1
a polarization image in a plurality of polarization directions
Data Source
AI summary
A normal candidate information generation unit of an information processor generates normal candidate information for each pixel indicating, for example, a zenith angle, or an azimuth angle, or a zenith angle and an azimuth angle, on the basis of a polarization image in a plurality of polarization directions obtained by a polarization imaging unit. The in-plane pixel selection unit selects a plurality of pixels indicating the plane to be observed in the polarization image. A normal calculation unit calculates a normal of the plane to be observed on the basis of the normal candidate information of the pixels selected by the in-plane pixel selection unit.