Camera Image Data Correctness Validation for Automated Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current surround view camera systems do not meet safety standards required for highly automated driving, as they lack the necessary safety integrity level to provide reliable image data for automated vehicle functions.
Innovation Solution
A method is introduced to validate the correctness of image data captured by multiple cameras by detecting features in the data from one camera and verifying their presence in the data from another, thereby increasing the safety integrity level of the image data output, allowing it to be used for automated driving functions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If image data from multiple cameras are validated by detecting and verifying features across cameras, then the safety integrity level of image data is improved, but the device complexity and computational requirements increase
Solution Approach 1:
The validation process is segmented into distinct functional modules: feature detection module that identifies features in first image data, feature verification module that checks presence of detected features in second image data, and correctness determination module that validates overall image data correctness. This modular segmentation allows the complex validation system to be managed through independent, specialized components.
Solution Approach 2:
The system performs preliminary feature detection and verification before final image data validation. By detecting features in the first camera's image data beforehand and preparing verification criteria based on second camera data, the system reduces the computational burden during the final correctness determination phase.
2Reliability
If features are detected and verified across multiple cameras to ensure data correctness, then the reliability of image data for automated driving is improved, but the processing time and computational resources increase
Solution Approach 1:
The system performs partial validation by focusing on verifying specific detected features rather than analyzing entire image datasets. By identifying key features in the first camera's image data and verifying only their presence in the second camera's data, the system achieves sufficient correctness validation without the excessive processing time required for complete image analysis.
Solution Approach 2:
The system extracts and isolates specific features from the first image data for targeted verification in the second image data. This extraction approach allows the system to focus computational resources on verifying only the most critical features rather than processing entire images, thereby reducing overall processing time while maintaining reliability.
3Measurement precision
If image data from multiple cameras are fused to create environment models, then the accuracy of automated driving functions is improved, but the device complexity and data processing requirements increase
Solution Approach 1:
The data fusion process is segmented into distinct stages: feature detection from first camera data, feature verification against second camera data, correctness determination, and environment model generation. This segmentation allows each stage to be optimized independently, reducing overall system complexity while maintaining measurement precision.
Solution Approach 2:
The system performs preliminary feature detection and correctness verification before generating environment models. By validating image data correctness beforehand through feature verification across multiple cameras, the system ensures high accuracy in the subsequent environment model generation without requiring complex real-time validation during model creation.
Data Source
AI summary
Provided is a method for determining a correctness of image data, the image data being captured by at least two cameras of a camera system installed at a vehicle. The method includes detecting at least one feature in the image data captured by a first one of the at least two cameras, determining if the at least one feature can be detected in the image data captured by a second one of the at least two cameras, and if the at least one feature can be detected in the image data captured by the second one of the at least two cameras, determining the correctness of the image data.