Stereo Camera Parallax Accuracy with Dynamic Pattern Light
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The accuracy of recognizing subjects in stereo matching processes is compromised by weak textures in images, especially under conditions like blocked-out shadows, saturated states, and light reflections, which affect the derivation of accurate parallax values.
Innovation Solution
A system that uses a semi-global matching (SGM) method combined with a pattern light irradiation system, where a stereo camera derives parallax values by forming a texture with pattern light and adjusts light intensity dynamically to mitigate image abnormalities, ensuring accurate parallax value calculation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If pattern light is irradiated to form texture on the subject, then the accuracy in deriving parallax values is improved, but image abnormalities such as blocked-out shadows, saturated states, and light reflections occur
Solution Approach 1:
The patent applies dynamics by making the light emission state changeable - the light emitting unit can dynamically switch between emitting pattern light and not emitting pattern light based on imaging conditions. This allows the system to adapt to different scenarios: using pattern light when texture is needed for accurate parallax derivation, and not using it when it would cause harmful effects like saturation or reflections.
Solution Approach 2:
The patent changes the parameter of light emission state (emitting vs. not emitting) to resolve the contradiction. By controlling whether pattern light is emitted, the system can optimize for either parallax accuracy or avoid image abnormalities depending on the specific imaging conditions and subject characteristics.
2Measurement precision
If pattern light is continuously emitted to improve texture formation, then parallax derivation accuracy is maintained, but the occurrence of saturated states and reflection issues increases
Solution Approach 1:
The system dynamically adjusts the light emission state rather than continuously emitting pattern light. The light emitting unit switches between emitting and non-emitting states based on real-time assessment of imaging conditions, subject material properties, and potential for harmful effects, thereby maintaining reliability while preserving measurement accuracy when needed.
Solution Approach 2:
The patent implements feedback control where the imaging system assesses the results of pattern light emission and adjusts subsequent emission decisions accordingly. By monitoring for signs of saturation, reflections, or other abnormalities, the system can feedback to the light emitting unit to adjust or stop pattern light emission, ensuring consistent image quality.
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 enhances the accuracy of parallax value derivation even in images with weak textures, effectively reducing the impact of image abnormalities and improving subject recognition.
Implementation Method 1
a light emitting unit (8), which irradiates pattern light to the part (20) to be a subject
Implementation Method 2
each of the imaging units (3, 4) generates a brightness image by imaging the subject
Data Source
Figure 1
Figure 2A~2C
Figure 3(a)~4
AI summary
An information processing system comprises: a light emitting unit (8, 8a) that irradiates pattern light having a specific pattern to a subject (20) to form a texture on the subject (20); an imaging unit (106) that captures the subject (20) on which the texture is formed; a deriving unit (62) that derives distance information to the subject (20) based on an image captured by the imaging unit (106); an analyzing unit (200) that analyzes whether or not an image abnormality is present in either one of the captured image and an image based on the distance information; and a dimming unit (201) that performs dimming control when the analyzing unit (200) analyzes that the image abnormality is present.