Heavy Rain Detection via Temporal Blur Transitions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting heavy rain on a vehicle windshield using cameras focused on infinity struggle to differentiate between rain and other environmental factors, leading to inaccurate detection and potential interference with driver assistance systems.
Innovation Solution
A method that utilizes a camera focused on a far range outside the vehicle, analyzing image blur caused by windshield wiper movements to detect heavy rain by setting threshold values for blur transitions and calculating the 'BlurExtent' using HAAR transforms and edge classification, allowing for reliable detection of heavy rain without focusing directly on the windshield.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the camera is focused on the windshield to detect raindrops, then rain detection accuracy is improved, but the camera cannot be used for monitoring surroundings focused at infinity
Solution Approach 1:
Instead of focusing the camera on the windshield to detect raindrops directly, the patent inverts the approach by focusing the camera at infinity to monitor surroundings, and detecting rain through the blur effect that raindrops create on the windshield. The windshield is intentionally kept out of focus, and rain detection is achieved by analyzing the blur in the captured images rather than by direct focus on the windshield surface.
2Adaptability or versatility
If the camera is focused at infinity to monitor surroundings, then driver assistance system functionality is improved, but rain detection accuracy deteriorates due to inability to resolve windshield droplets
Solution Approach 1:
The patent uses image blur as an intermediary indicator to detect rain. Instead of directly observing windshield droplets (which requires focusing on the windshield), the system uses the blur effect that droplets create in the out-of-focus regions of the image as a mediator to infer rain presence. The blur serves as an intermediate signal that connects the focused-at-infinity camera with the rain detection objective.
Solution Approach 2:
The patent replaces the mechanical focusing mechanism with an image processing approach. Instead of mechanically adjusting the camera focus to resolve droplets, the system uses computational methods to analyze blur characteristics in the captured images. The focus remains fixed at infinity, and rain detection is achieved through algorithmic analysis of blur patterns rather than through optical focusing on the windshield.
3Adaptability or versatility
If blur analysis is used to detect rain with a camera focused at infinity, then camera versatility is improved, but detection reliability worsens due to difficulty differentiating rain blur from other environmental factors
Solution Approach 1:
The patent applies local quality analysis by examining blur characteristics in specific regions of the image rather than analyzing the entire image uniformly. The system evaluates blur in local areas to detect rain patterns, allowing it to differentiate rain-induced blur from other environmental factors by analyzing the spatial distribution and local characteristics of blur in different image regions.
Solution Approach 2:
The patent uses dynamic analysis by examining changes in blur characteristics over time rather than relying on a single static image. The system analyzes temporal variations in blur patterns to distinguish rain from other environmental factors, as rain-induced blur exhibits specific dynamic patterns that differ from static or differently evolving blur caused by other conditions.
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 determines heavy rain conditions by measuring image blur changes caused by wiper movements, reducing false positives and ensuring accurate detection of low image quality, thus enabling timely intervention in driver assistance systems.
Implementation Method 1
a camera (1) which is not focused on the pane (2) of the vehicle but is configured to view the scene outside of the vehicle
Implementation Method 2
processing unit (3) is configured to detect heavy rain on the pane (2) of the vehicle by means of image processing of a series of images captured by the camera (1)
Data Source
Figure 1
Figure 2a~2d
Figure 3
AI summary
The present invention relates to a method and a device for detecting heavy rain in images (1), the method comprising the steps: a) providing a series of images (1) as input information, the images (1) relating to a scene external to a vehicle viewed through a pane of the vehicle; b) evaluating a measure of blur in an area of interest (AOI) in the images (1); c) analyzing the course of the measured blur values over time in order to detect transitions between images (1) with low blur values and images (1) with high blur values or vice versa; d) detecting heavy rain from the detected transitions; and e) providing an output information in the case where heavy rain is detected.