Monocular Camera Deceleration Detection Using Lane Marker Counting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems struggle to accurately detect the deceleration of a preceding vehicle using a monocular camera due to factors like sunlight reflection, unlit brake lamps, and brake pedal play, without relying on expensive stereo cameras.
Innovation Solution
A method and device using a monocular camera to detect deceleration by counting equally spaced stationary objects like lane markers, measuring distance, and analyzing time-series transitions in conjunction with vehicle speed to determine deceleration without relying on brake lamp detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If brake lamp detection is used to detect preceding vehicle deceleration, then the detection method is simple, but the detection reliability is low due to sunlight reflection and brake lamp not being lit during deceleration
Solution Approach 1:
The patent introduces equally spaced stationary objects (road markings) as an intermediary element to indirectly measure the distance to the preceding vehicle. Instead of directly detecting brake lamp status, the system counts the number of stationary objects between the own vehicle and the preceding vehicle, which serves as a reliable proxy for distance measurement and deceleration detection.
2Measurement precision
If stereo camera is used to detect distance change of preceding vehicle, then the deceleration detection accuracy is high, but the device cost is high
Solution Approach 1:
The patent replaces the expensive stereo camera system with a single monocular camera combined with virtual markers (equally spaced stationary objects). This substitution uses a low-cost sensor (monocular camera) augmented with virtual reference objects to achieve distance measurement precision comparable to expensive hardware solutions.
Solution Approach 2:
The patent creates virtual copies of physical reference objects by generating equally spaced stationary object markers in the image space. These virtual markers are positioned at known intervals along the road, allowing the monocular camera to infer distance information through counting, effectively copying the distance measurement capability of stereo vision without the hardware complexity.
3Device complexity
If monocular camera with brake lamp detection is used, then the device cost is low, but the detection accuracy is insufficient
Solution Approach 1:
The patent performs preliminary action by pre-defining and detecting equally spaced stationary objects (road markings) in the environment before they are needed for distance measurement. These virtual markers are established in advance along the road, allowing the system to quickly and accurately measure distance by simply counting them without requiring complex real-time calculations.
Data Source
AI summary
The apparatus of the present disclosure detects a preceding vehicle and a white dashed line provided at regular intervals along a road from an image captured by a monocular camera. In addition, the apparatus of the present disclosure acquires speed information of the host vehicle. Then, the apparatus of the present disclosure counts the number of white dashed lines included in the section from the vertical position of the preceding vehicle to the lower fixed reference position of the vertical position in the camera image, and detects the deceleration of the preceding vehicle based on the time-series transition of the number of white dashed lines and the speed information of the host vehicle.