Roadside Monocular Camera Calibration with Nonlinear Vehicle Speed Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional vehicle speed detection methods using ground induction coils, radar, and laser are costly, complex, and susceptible to environmental factors, while video-based methods lack accuracy and timeliness, and existing camera calibration methods are inefficient for changing road scenes.
Innovation Solution
A method and device using roadside monocular camera calibration that segments traffic video, applies a multi-target detection algorithm, tracks vehicle movement, and calculates speed using a fitted pixel-to-world coordinate conversion curve, reducing manual recalibration and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a simple linear function is used for coordinate conversion, then the device complexity is reduced, but the measurement precision deteriorates due to foreshortening effects
Solution Approach 1:
The patent changes the parameter model from a simple linear function to a nonlinear function that includes foreshortening correction terms. The coordinate conversion uses parameters such as camera height, vertical angle, and nonlinear correction coefficients to accurately model the perspective distortion, thereby improving measurement precision while maintaining reasonable device complexity through automated parameter estimation.
Solution Approach 2:
The patent replaces manual calibration operations with an automated image processing system. By using deep learning-based object detection and tracking algorithms, the system automatically extracts reference objects (such as lane markings or road features) and computes the coordinate conversion parameters without manual intervention, reducing both device complexity and calibration time.
2Measurement precision
If camera external parameters are recalibrated for scene changes, then the measurement precision is maintained, but the loss of time and operational complexity increase
Solution Approach 1:
The system performs self-calibration by automatically detecting reference objects in the current scene and computing the coordinate conversion parameters based on their known real-world dimensions. The deep learning model automatically identifies and tracks reference features (such as lane markings, road signs, or pavement markings) to determine camera parameters without external intervention, enabling the system to adapt to scene changes autonomously.
Solution Approach 2:
The patent pre-establishes a database of reference objects with their standard dimensions and visual characteristics. When a scene change occurs, the system quickly matches current image features against this pre-established database to rapidly compute new calibration parameters, avoiding time-consuming manual recalibration while maintaining measurement precision.
3Device complexity
If conventional video-based methods are used, then the device complexity is reduced, but the measurement precision and timeliness deteriorate
Solution Approach 1:
The patent replaces traditional mechanical or optical detection methods (such as induction coils or radar) with a digital image processing system based on deep learning. The system uses convolutional neural networks to detect and track vehicles in video frames, extracting position and motion information that is then converted to speed using the calibrated coordinate transformation, achieving high precision with moderate device complexity.
Solution Approach 2:
The patent creates a universal detection system that can handle various vehicle types, road conditions, and environmental scenarios through a single deep learning model. The model is trained to recognize diverse reference objects and vehicle patterns, making the system adaptable to different scenes without requiring separate calibration or detection algorithms for each scenario, thereby improving both precision and timeliness.
Data Source
AI summary
The present disclosure provides an expressway vehicle speed measuring method and device based on roadside monocular camera calibration. The expressway vehicle speed measuring method based on roadside monocular camera calibration includes: acquiring a traffic monitoring video in a roadside monocular camera; segmenting the traffic monitoring video into a traffic background image set in frames, and transmitting the traffic background image set to a well-trained multi-target detection algorithm to obtain a pixel position of a target vehicle; tracking a target vehicle detection result in each frame through a multi-target tracking algorithm to obtain a movement locus of the target vehicle; and calculating an actual moving distance of the target vehicle according to the movement locus of the target vehicle and a fitted pixel-to-world coordinate conversion curve, thereby calculating a speed of the target vehicle. The present disclosure can achieve a higher accuracy and a lower operational complexity in vehicle speed detection.