Stereo ADAS Camera Alignment for Ground Truth Scene Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Advanced driver assistance systems (ADAS) lack a reliable ground truth system for accurate distance estimation, and existing solutions like LiDAR are costly, have sparse scanning lines, and cannot match the frame rate of video data, leading to inefficiencies in data collection and processing.
Innovation Solution
Implement a binocular stereo vision system using two monocular ADAS cameras mounted on a vehicle chassis, synchronized by a frame synchronization signal, to generate ground truth data for scene reconstruction, eliminating the need for LiDAR and providing denser point clouds and improved distance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LiDAR is used for ground truth system, then distance measurement capability is improved, but system cost increases and scanning lines become sparse
Solution Approach 1:
The patent uses ADAS cameras to capture images that serve as a copy or alternative representation of the scene, replacing the need for LiDAR. The stereo vision system processes these camera images to generate point clouds and ground truth data, achieving distance measurement without expensive LiDAR hardware
Solution Approach 2:
The patent replaces the mechanical LiDAR scanning system with an optical camera-based stereo vision system. Instead of using LiDAR's mechanical scanning lines to measure distance, the system uses camera images processed through stereo vision algorithms to achieve the same measurement function
2Measurement precision
If LiDAR is used for ground truth system, then distance measurement capability is improved, but scanning lines become sparse leading to insufficient point density
Solution Approach 1:
The patent uses high-resolution camera images as a dense copy of the scene, where each pixel can potentially contribute to point cloud generation. This provides much higher point density compared to LiDAR's sparse scanning lines, as camera images capture continuous spatial information across the entire field of view
3Measurement precision
If LiDAR is used for ground truth system, then distance measurement capability is improved, but frame rate becomes low causing mismatch with video frames
Solution Approach 1:
The patent makes the ADAS cameras serve multiple functions: they both capture video frames for ADAS processing and provide ground truth data for scene reconstruction. Since the same cameras are used for both purposes, there is no frame rate mismatch between video and ground truth data, unlike LiDAR which operates at a lower frame rate
4Measurement precision
If LiDAR is used for ground truth system, then distance measurement capability is improved, but field-of-view and installation location differ from ADAS cameras causing FOV loss
Solution Approach 1:
The patent uses the same ADAS cameras for both video capture and ground truth generation, ensuring identical field-of-view and installation location. This eliminates FOV loss and alignment issues that occur when using separate LiDAR and camera systems with different positions and viewing angles
Data Source
AI summary
An apparatus includes a chassis and a processor. The chassis may be configured to be mounted to a vehicle and to hold a first ADAS camera and a second ADAS camera. The chassis generally provides a coarse alignment of the first ADAS camera and the second ADAS camera to obtain stereo images of an area outside of the vehicle. The processor may be configured to (i) generate a frame synchronization signal based on a real-time clock signal, (ii) present the frame synchronization signal and one or more control signals to the first ADAS camera and the second ADAS camera, (iii) receive a first pixel datastream corresponding to the area outside of the vehicle from the first ADAS camera, (vi) receive a second pixel datastream corresponding to the area outside of the vehicle from the second ADAS camera, (v) process the first pixel datastream arranged as first video frames and the second pixel datastream arranged as second video frames, (vi) compute warp parameters for the first ADAS camera and the second ADAS camera to finely align pixel data of the first video frames with pixel data of the second video frames, and (vii) generate ground truth data based on the first video frames from the first ADAS camera and the second video frames from the second ADAS camera.