ISP Fault Detection Using Embedded Image Test Signals
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image signal processors (ISPs) in image processing systems can generate inaccurate images due to faults, leading to issues in advanced driver assistance systems (ADAS) that rely on these images.
Innovation Solution
A method and device for generating an input signal to an ISP using a raw image signal and an input test signal, and a fault detection circuit to determine if a fault occurs by comparing the output test signal to a target value, using cyclic redundancy check (CRC) or secure hash algorithm (SHA) hash.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If an ISP processes a raw image signal to generate an image for ADAS functions, then the image processing capability is improved, but the risk of fault-induced inaccuracies increases
Solution Approach 1:
The patent applies preliminary action by inserting a test signal into the raw image signal before it enters the ISP processing pipeline. This allows the system to pre-establish a reference input signal that can be used later to verify the ISP's processing correctness, thereby detecting faults before they affect actual ADAS operations.
Solution Approach 2:
The patent implements feedback by comparing the ISP's output signal (after removing the test component) with a reference value generated from the known test signal. This closed-loop verification mechanism provides continuous monitoring of ISP processing accuracy, enabling real-time fault detection and maintaining image reliability.
2Measurement precision
If a test signal is inserted into the raw image signal to detect ISP faults, then the fault detection capability is improved, but the complexity of signal processing increases
Solution Approach 1:
The patent applies segmentation by separating the test signal component from the raw image signal in the frequency or time domain. The test signal is inserted into specific intervals (e.g., vertical blanking intervals) of the raw image signal, allowing independent processing and extraction of the test component without affecting the main image data processing pipeline.
Solution Approach 2:
The patent uses an intermediary approach by introducing a known test signal as a mediator between the input and output of the ISP. This test signal serves as a reference that simplifies fault detection, as its known characteristics make it easy to identify and compare against the expected output, reducing the complexity of the verification process.
3Productivity
If the test signal is processed together with the raw image signal through the ISP, then the processing efficiency is maintained, but the difficulty of extracting the test signal from the output increases
Solution Approach 1:
The patent applies preliminary action by strategically positioning the test signal in specific time or frequency intervals (such as vertical blanking intervals) where it does not overlap with the main image data. This pre-planned separation makes it straightforward to extract the test signal from the ISP output by simply isolating the corresponding intervals, without requiring complex signal processing.
Data Source
AI summary
Disclosed is a method of determining an image signal processor (ISP) fault, performed by an image processing device. The method includes receiving a first raw image signal that is at least a portion of an image signal captured by an image sensor of a first camera, based on the first raw image signal and a first input test signal, generating a first input signal, generating, by a first ISP, a first output signal by processing the first input signal, obtaining a first output test signal corresponding to the first input test signal from the first output signal, and based on the first output test signal, determining whether a fault occurs in the first ISP.