Lidar Window Impurity Detection via Image Feature Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting impurities on lidar viewing windows, such as those used in automated vehicles, are inadequate, leading to degraded sensor performance and compromised safety due to contamination.
Innovation Solution
A method involving the emission of a laser beam and detection of reflections, combined with background light analysis, uses edge detection algorithms to compare intensity and background images for similarity, concluding contamination based on shared features exceeding a threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing methods for detecting impurities are used, then detection capability is provided, but sensor performance degrades due to inadequate detection accuracy
Solution Approach 1:
The patent segments the detection process into two distinct phases: active illumination detection (intensity image) and passive background light detection (background light image). This segmentation allows comparison of the same scene under different lighting conditions, enabling more accurate contamination detection by identifying inconsistencies between the two images that indicate impurities on the viewing window.
2Reliability
If contamination detection is implemented, then safety is improved, but false detections occur due to insufficient differentiation between clean and contaminated states
Solution Approach 1:
The patent employs feedback by comparing intensity image data with background light image data from the same scene. The system uses edge detection algorithms to identify features in both images and compares them, providing feedback that reveals inconsistencies caused by contamination. This feedback mechanism enables reliable differentiation between clean and contaminated states while minimizing false detections.
3Device complexity
If simple detection methods are used, then device complexity is reduced, but detection reliability is compromised
Solution Approach 1:
The patent introduces an intermediary approach by using background light images as a reference medium to validate the intensity images. Instead of directly analyzing intensity images for contamination, the system uses background light images captured without active illumination as an intermediary reference, comparing edges and features between the two to indirectly detect contamination with higher reliability.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Effectively distinguishes clean and contaminated lidar viewing windows by analyzing image similarities, enhancing sensor reliability and safety in automated vehicles.
Implementation Method 1
A laser beam is emitted into a detection area by means of a transmitter of the lidar, and light present in the detection area is detected by means of a receiver of the lidar
Implementation Method 2
edges are detected in the intensity image and in the background light image using an edge detection algorithm
Data Source
Figure 1~2
Figure 3~4
AI summary
The invention relates to a method and a device for detecting impurities on a viewing window of a LIDAR. According to the method, - a laser beam is emitted into a detection region by means of a transmitter of the LIDAR, and - light present in the detection region is detected by means of a receiver of the LIDAR. The invention is characterized in that - an intensity image in the form of a grayscale image of laser reflection intensities is generated from the light that is reflected and detected as a result of the emission of the laser beam, - a background light image in the form of a grayscale image of a background light is generated from light detected without the emission of a laser beam, - the intensity image and the background light image are analyzed with respect to common features, and - if a specified number of common features is not attained, the presence of an impurity on the viewing window is inferred.