Vehicle Visibility Estimation Using Object Contrast and Distance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating meteorological visibility distance in vehicles are limited by the reliance on ground markings, which are prone to degradation and require expensive dual-camera systems, and do not accurately account for varying weather conditions.
Innovation Solution
A system using a camera and obstacle detection system to calculate the atmospheric extinction coefficient by analyzing contrast changes of detected objects, such as other vehicles, to estimate visibility distance, allowing for real-time meteorological visibility distance estimation independent of road markings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If ground markings are used as reference points for contrast extraction, then visibility distance can be estimated, but the accuracy degrades due to paint defects, broken lines, and road color variations
Solution Approach 1:
The patent introduces a localization system as an intermediary component that provides distance information to the visibility estimation system. This mediator supplies accurate distance data to the processing means, enabling the system to calculate atmospheric extinction coefficients and estimate visibility distance without relying on ground markings, thus resolving the reliability and accuracy degradation issues
Solution Approach 2:
The patent changes the reference parameter from ground marking contrast to object contrast at known distances. By using the localization system to provide accurate distance parameters and selecting objects (such as other vehicles) as reference points, the system transforms the estimation approach to avoid defects in ground markings while maintaining measurement capability
2Measurement precision
If a dual-camera system is used to improve visibility estimation accuracy, then measurement precision improves, but device complexity and cost increase
Solution Approach 1:
The patent makes the existing camera multifunctional by having it serve both the obstacle detection system and the meteorological visibility distance estimation system. The same camera captures images for both obstacle detection and contrast extraction for visibility estimation, eliminating the need for a dedicated second camera and reducing overall system complexity
Solution Approach 2:
The patent merges the visibility estimation function with the existing obstacle detection system. By combining these two functions into a single integrated system that shares the camera and processing resources, the patent achieves improved visibility estimation accuracy without the complexity and cost of a separate dual-camera system
3Adaptability or versatility
If road markings are used as reference, then visibility estimation can be performed, but the method fails when road markings are absent or degraded
Solution Approach 1:
The localization system acts as a mediator that provides distance information independent of ground markings. This intermediary enables the system to estimate visibility distance using any detectable object as a reference point, making the method adaptable to environments where road markings are absent or degraded while maintaining reliable estimates
Solution Approach 2:
Instead of using ground markings to determine distance and then estimating visibility, the patent inverts the approach by using the localization system to provide distance information and then using that distance to select appropriate reference objects for contrast extraction. This inversion makes the system adaptable to various road conditions without relying on markings
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
This approach provides accurate and reliable real-time estimation of meteorological visibility distance, validated through tests under different fog conditions, and is independent of the detected object's distance, improving precision and reducing costs compared to previous methods.
Implementation Method 1
The atmospheric extinction coefficient k can be calculated from a first contrast of the localized object at a first distance estimated by the system, said first contrast being extracted from a first image captured by said camera at a first instant, and from a second contrast of the object located at the second distance estimated by the system
Implementation Method 2
Loss of visibility can be caused by various factors, including road topography, vegetation, obstacles temporarily obscuring the road, or adverse weather conditions, particularly fog
Data Source
Figure 1~2
Figure 3~4
AI summary
The invention relates to a method of estimating the meteorological visibility distance on the basis of a processing of images of landscapes captured successively by a camera on board a motor vehicle, said processing of images comprising a step (S40) of extracting contrasts in at least one zone of interest of the images and a step of determining said meteorological visibility distance on the basis of the contrasts extracted. In accordance with the invention, said at least one zone of interest is centred on at least one object located by a system on board said motor vehicle and able to provide an estimation of the distance separating the motor vehicle from the object located.