Diagonal Stereo Camera Vision for All-Direction Distance Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for acquiring distance information around a moving body, such as LiDAR and RADAR sensors, are costly, power-intensive, and mechanically unstable, while stereo camera systems require multiple cameras and face issues with information propagation and discontinuity.
Innovation Solution
A vision apparatus using four cameras arranged diagonally on a moving body's corner surface, with a processor that stereo-matches images to generate distance information for all directions by projecting images onto a virtual three-dimensional figure, correcting distortion and compensating for camera posture errors in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LiDAR and RADAR sensors are used to acquire distance information, then measurement precision is improved, but cost and power consumption increase significantly
Solution Approach 1:
The patent replaces active sensing systems (LiDAR, RADAR) with a passive optical system using standard cameras. Instead of emitting laser beams or radio waves and measuring reflections, the system uses multiple cameras to capture images and computationally derives distance information through stereo matching and projection techniques, thereby reducing hardware complexity and power consumption while maintaining measurement capability
Solution Approach 2:
The patent creates a virtual three-dimensional model by projecting images from multiple cameras onto a virtual 3D figure. This computational copying of spatial information allows the system to reconstruct distance and depth data without requiring expensive specialized sensing hardware, achieving accurate distance measurement through image processing rather than direct physical measurement
2Device complexity
If general stereo cameras are used to estimate distance, then cost is reduced, but the number of cameras required increases to 12 cameras for all-direction coverage
Solution Approach 1:
The patent transitions from two-dimensional image capture to three-dimensional spatial reconstruction by projecting images onto a virtual 3D figure. This dimensional transformation allows four cameras to achieve all-direction distance estimation coverage that would traditionally require twelve cameras, as the 3D projection synthesizes information from multiple viewing angles and combines data from overlapping fields of view
Solution Approach 2:
The patent merges information from multiple cameras by projecting their images onto a unified virtual three-dimensional model. The processor integrates data from all four cameras, combining their respective fields of view and stereo pairs to generate comprehensive all-direction distance information, thereby reducing the total number of cameras needed while maintaining complete spatial coverage
3Quantity of substance
If four cameras are used to reduce camera quantity, then device complexity is reduced, but distortion and posture errors may cause measurement precision to deteriorate
Solution Approach 1:
The patent incorporates posture information of each camera as feedback into the projection process. The processor calculates and updates posture information at predetermined cycles, using this feedback to correct for camera movements and maintain accurate spatial relationships in the virtual 3D model, thereby preserving measurement precision despite using fewer cameras
Solution Approach 2:
The patent dynamically adjusts projection parameters based on real-time posture information. By changing the projection parameters according to calculated posture data, the system compensates for distortion and positional errors introduced by camera movements, maintaining accurate distance measurement precision with a reduced four-camera configuration
Data Source
AI summary
A vision apparatus for a moving body is provided. The vision apparatus for a moving body includes a plurality of cameras that are arranged to be distanced from one another, and are arranged in a diagonal direction to the moving direction of the moving body, and a processor that receives images photographed at each of the plurality of cameras, and stereo-matches the plurality of received images and generates distance information for all directions of the moving body.