Per-Line CRC Checks for Parallel Image Data Integrity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Detecting data corruption in safety-critical systems, such as vehicle camera modules, is computationally expensive and inefficient when using full-frame CRC checks, especially on parallelized processors like GPUs.
Innovation Solution
Implementing per-line data integrity checks, where a data integrity value is calculated and appended to each line of pixel data, allowing for parallel processing and efficient corruption detection by comparing these values across processors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If full-frame CRC checks are used to detect data corruption, then data integrity validation is achieved, but computational expense and processing time increase significantly
Solution Approach 1:
The patent divides the image data into multiple scanlines, with each scanline having its own separate CRC check value. This segmentation allows parallel processing of multiple scanlines simultaneously on the GPU, rather than processing the entire frame sequentially. The segmentation principle directly resolves the contradiction by enabling both comprehensive data integrity validation (checking every scanline) and high processing efficiency (parallel execution).
Solution Approach 2:
The patent implements CRC checks at a granular scanline level rather than performing a single full-frame check. This partial action approach allows the system to validate data integrity for individual scanlines independently, enabling early termination and partial frame usage when corruption is detected in only certain regions, thus improving overall processing efficiency while maintaining reliability.
2Reliability
If full-frame CRC checks are performed on parallelized processors like GPUs, then data corruption detection is achieved, but the serial nature of CRC calculation reduces processing speed
Solution Approach 1:
By segmenting the image into scanlines with individual CRC values, the patent enables parallelized processors to compute multiple CRC checks simultaneously. Each processing core can handle one or more scanlines in parallel, transforming the inherently serial CRC operation into a parallelizable task that充分利用 the GPU's computational architecture.
Solution Approach 2:
The patent introduces a new dimension of parallelism by organizing CRC checks along the scanline dimension rather than processing the entire frame as a single serial operation. This dimensional transformation allows the system to exploit the parallel processing capabilities of GPUs, where multiple thread blocks can compute CRC values for different scanlines concurrently, dramatically improving processing speed.
3Productivity
If per-line data integrity checks are implemented, then parallel processing efficiency is improved, but data structure complexity increases
Solution Approach 1:
The patent segments both the image data and its associated integrity check data into scanline-level units. Each scanline is paired with its corresponding CRC value, creating a structured organization that naturally supports parallel processing while maintaining manageable data structure complexity through regular, predictable patterns.
4Reliability
If full-frame CRC validation is used, then complete data integrity is ensured, but downstream consumers cannot utilize partially corrupted frames
Solution Approach 1:
By segmenting the frame into independent scanline units with individual CRC validation, the patent enables downstream consumers to process only the uncorrupted portions of the frame. When corruption is detected in specific scanlines, those regions can be excluded while the rest of the frame continues processing, maintaining temporal consistency and reducing time loss.
Solution Approach 2:
The patent allows partial frame usage by validating and processing only the portions of the frame that pass CRC checks. This partial action principle enables the system to maximize utilization of valid data while excluding only the corrupted regions, thereby maintaining temporal consistency and minimizing the impact of data corruption on overall system performance.
Data Source
AI summary
Techniques for image data protection using cyclic redundancy checks are disclosed herein. Some of the techniques may include, at a processor, receiving image data that includes multiple lines of pixel data. The processor may also determine at least a first hash value representing a first line of pixel data of the multiple lines of pixel data and a second hash value representing a second line of pixel data of the multiple lines of pixel data. The processor may also send the image data to a computing device that is configured to determine, based at least in part on the first hash value and the second hash value, whether the image data is corrupt.