3D Structure Sensing System Reflective Surface Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 3D sensing systems face challenges in accurately detecting reflective surfaces like mirrors and metals, which can cause errors or false readings, impacting safety and reliability in applications such as facial recognition and terrain recognition.
Innovation Solution
A 3D structure sensing system that includes an image sensor to capture reflected light and a depth processing unit to generate depth data, which determines whether the depth data is affected by a reflective surface by comparing image data and depth data, and modifies the depth data accordingly to prevent errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If active light source sensing systems (ToF or SL) are used to capture 3D information, then depth measurement capability is improved, but measurement precision deteriorates when reflective surfaces like mirrors and metals are present
Solution Approach 1:
The patent introduces an intermediary analysis process that compares image data from the image sensor with depth data from the depth sensor. This intermediary comparison mechanism identifies discrepancies caused by reflective surfaces and flags affected depth measurements, preventing them from corrupting the overall 3D reconstruction. The system uses the image data as an intermediary reference to validate depth measurements.
Solution Approach 2:
The system implements a feedback mechanism where the image sensor continuously monitors the visual appearance of surfaces, and this information feeds back to the depth processing unit. When the image data indicates highly reflective characteristics (such as mirror-like surfaces or polished metals), the system adjusts its depth measurement interpretation accordingly, using the feedback loop to correct for reflective interference.
2Measurement precision
If the system processes both image data and depth data to detect reflective surfaces, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent merges the image sensor and depth sensor into a unified sensing system with a shared processing pipeline. Rather than operating as separate systems, the image and depth data streams are integrated and processed together through a single depth processing unit that performs both depth calculation and reflective surface detection. This merging reduces overall system complexity while maintaining high measurement precision.
Solution Approach 2:
The depth processing unit is designed with multi-functionality, serving both to calculate depth from stereo or ToF data and to detect reflective surfaces by analyzing discrepancies between image and depth data. This universal processor handles multiple tasks that could otherwise require separate dedicated components, thereby reducing device complexity while improving measurement precision through integrated analysis.
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 system effectively mitigates the negative effects of reflective surfaces on depth data accuracy, ensuring reliable and accurate 3D information for various applications, including improved facial recognition and terrain mapping.
Implementation Method 1
The image sensor receives a reflected light from an object irradiated by an emitted light, the reflected light being converted into image data representing an image of the object
Data Source
AI summary
A three-dimensional structure sensing system includes an image sensor that receives a reflected light from an object irradiated by an emitted light, the reflected light being converted into image data representing an image of the object; and a depth processing unit that generates depth data according to the image data. It is determined whether the depth data is affected by a reflective surface according to the image data and the depth data.