Vehicle Light State Detection Using Neural Network Image Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for detecting the state of vehicle lights, such as blinking or braking, are limited in their ability to differentiate between various lighting states and require separate localization and tracking of individual lights, and are not effective for pulsed LED lights or partially defective lights, especially when the blinking frequency is variable.
Innovation Solution
A method using an artificial neural network to evaluate camera images of vehicle areas, allowing for the detection of vehicle light states without separate localization or tracking, and capable of distinguishing between different lighting states, including blinking and braking, regardless of lighting design or type, using image subsignals with different color information and spectral ranges.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If frequency examination of intensity profile is used to detect blinking, then blinking detection is possible, but the method cannot detect braking states and is limited to specific blinking frequencies
Solution Approach 1:
The patent replaces traditional mechanical signal processing methods (frequency examination of intensity profiles) with an artificial neural network-based image evaluation system. The neural network directly processes camera images to classify vehicle light states into multiple categories including braking, blinking, and their combinations, eliminating the need for separate localization and tracking of individual lights while achieving reliable detection across variable blinking frequencies and different lighting types including pul LED lights.
2Measurement precision
If separate localization and tracking of individual vehicle lights is performed, then precise light state detection is possible, but the device complexity increases
Solution Approach 1:
The patent merges the functions of localization, tracking, and state classification into a single integrated neural network evaluation process. Instead of separately localizing individual lights and tracking their states, the neural network evaluates the entire vehicle light area in the camera image and directly classifies the combined light state, significantly reducing device complexity while maintaining detection precision.
Solution Approach 2:
The neural network evaluation device is designed to universally handle multiple vehicle light states (braking, blinking, combinations) and different lighting types (LED, pulsed LED, defective lights) through a single multi-functional system, eliminating the need for specialized processing pipelines for each light type or state.
3Ease of manufacture
If traditional image evaluation methods are used, then processing is simpler, but detection of pulsed LED lights and partially defective lights is not effective
Solution Approach 1:
The patent changes the evaluation parameters from traditional intensity profile frequency analysis to neural network-based image feature analysis. This parameter change enables effective detection of pulsed LED lights and partially defective lights by learning from training data that includes these challenging cases, achieving both processing efficiency and high detection reliability across diverse lighting conditions.
Data Source
AI summary
A method is described for ascertaining a state of a vehicle light of a vehicle. The method includes a step of reading in and a step of evaluating. In the step of reading in, an image signal is read in that includes a camera image including a vehicle area of a detected vehicle. In the step of evaluating, the image signal is evaluated by using an artificial teachable model, in particular an artificial neural network. Here, at least two state categories, each representing a state of the vehicle light of the vehicle, are differentiated. That state category is ascertained as the evaluation result which represents the instantaneous state of the vehicle light of the vehicle.