Vehicle Camera Optical Blockage Detection via Server Reference Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting optical blockages in vehicle cameras are not robust, leading to reduced system availability and critical system behavior due to incorrect recognition of blockages or failure to recognize actual blockages, especially in varying environmental conditions.
Innovation Solution
A method that records an image of a static scene and compares it with a reference image retrieved from a server, based on averaged image information from multiple vehicles under similar conditions, using features like brightness, edge detection, sharpness, and homogeneity to reliably detect optical blockages, and adjusts the degree of automation or hands over control to the driver if deviations exceed predetermined limits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If image processing algorithms are used to detect optical blockage, then detection capability is provided, but reliability is reduced due to incorrect detection under certain environmental conditions
Solution Approach 1:
The patent introduces a server-based reference image system as an intermediary to mediate between the camera image and the blockage detection algorithm. Instead of directly comparing images, the system retrieves a reference image from a server that contains averaged image information from multiple vehicles under similar conditions. This intermediary reference image provides a more reliable basis for detection, reducing false positives and false negatives that occur when using simple image processing algorithms alone.
Solution Approach 2:
The patent creates a virtual copy of the expected scene by averaging images from multiple vehicles to generate a reference image stored on a server. This copied reference representation is then used for comparison with the actual camera image. By using this averaged reference copy instead of relying solely on algorithmic detection of the single captured image, the system achieves more reliable blockage detection across varying environmental conditions.
2Measurement precision
If a reference image from a server is used for comparison, then detection accuracy is improved, but system complexity increases
Solution Approach 1:
The patent implements a server-based system that serves multiple vehicles with reference images, making the solution universally applicable across the fleet. The server performs the complex task of averaging and storing reference images once, and then provides this reference data to multiple vehicles. This multi-functional approach distributes the computational complexity to the server while keeping individual vehicle systems relatively simple, achieving high detection accuracy without proportionally increasing on-vehicle system complexity.
3Reliability
If image averaging from multiple vehicles is performed, then reference image reliability is improved, but data processing time increases
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing the averaged reference images on a server before they are needed for actual blockage detection. The time-consuming process of averaging images from multiple vehicles is performed in advance and offline, and the results are stored for rapid retrieval. When a vehicle needs to detect blockage, it simply retrieves the pre-computed reference image from the server, avoiding the time cost of real-time averaging and achieving both high reliability and fast detection.
Data Source
Figure 1
AI summary
The invention relates to a method for detecting optical blockage of a camera of a vehicle (1). According to the invention, an image (B) of a static scene from an environment of the vehicle (1) is captured by the camera. During the capturing of the image (B), current capturing conditions are determined. The captured image (B) is compared to a reference image which is retrieved from a server (2) external to the vehicle and is based on images captured by a plurality of vehicles of the same static scene under the same or similar capturing conditions. If differences in image features between the captured image (B) and the reference image exceed a predefined tolerance level, they are taken as an indication that the camera is optically blocked. Furthermore, the invention relates to a method for operating a vehicle (1) which is designed to perform automated driving.