LPR Camera Speed Estimation via Pixel Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing License Plate Recognition (LPR) systems face challenges in accurately estimating vehicle speed over large distances due to the need for multiple LPR units and the masking of details in calculated average speed, which can be costly and inefficient.
Innovation Solution
The method involves estimating vehicle speed by tracking pixel locations of a vehicle in successive images and using a look-up table or mapping between camera coordinate space and roadway surface coordinates, with corrections for 3D to 2D image sensor distortions, and employing calibration passes to establish the relationship between camera and physical movements, allowing for precise speed estimation using a single camera unit.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If two or more LPR systems are physically separated by a large distance to perform speed estimation, then speed calculation can be performed, but the calculated average speed masks important details about the actual speed of the vehicle and the system becomes costly
Solution Approach 1:
The patent segments the speed estimation task into multiple discrete steps performed by a single LPR system: (1) capturing multiple images of the vehicle at different positions, (2) identifying tracking points on the vehicle in each image, (3) calculating pixel displacement between images, (4) converting pixel displacement to physical distance using calibration data, and (5) computing speed from distance and time. This segmentation allows one system to perform functions that traditionally required multiple systems.
Solution Approach 2:
The patent introduces several intermediary elements to enable speed estimation with a single LPR system: (1) a calibration tarp with known markings placed on the roadway serves as an intermediary reference object to establish the relationship between pixel displacement and physical distance, (2) tracking points identified on the vehicle act as intermediaries to measure vehicle position changes, and (3) a lookup table or mathematical model serves as an intermediary to convert pixel space measurements to physical space measurements.
2Measurement precision
If two or more LPR systems are used to capture images at different locations, then vehicle speed can be estimated, but the system cost and complexity increase
Solution Approach 1:
The patent makes the single LPR system multi-functional by enabling it to perform both license plate recognition and vehicle speed estimation using the same hardware infrastructure. The system captures images for LPR purposes and simultaneously extracts speed information by tracking vehicle features across multiple images, eliminating the need for separate speed measurement equipment.
Solution Approach 2:
The patent performs preliminary calibration by capturing images of a calibration tarp with known dimensions and markings before actual speed measurement. This preliminary action establishes a mapping between pixel coordinates and physical distances, creating a lookup table or mathematical model that enables accurate speed estimation during subsequent vehicle tracking without requiring additional calibration equipment during operation.
3Measurement precision
If multiple LPR units are deployed to estimate vehicle speed, then speed data can be obtained, but the aggregation of information from multiple systems increases complexity
Solution Approach 1:
The single LPR system performs self-service by automatically capturing multiple images of the same vehicle as it passes by, automatically identifying tracking points on the vehicle in each image, automatically calculating the displacement between images, and automatically computing the speed. The system serves itself by generating all necessary measurement data and performing all processing steps without requiring data aggregation from external sources.
Data Source
AI summary
Methods and systems for estimating the speed of passing vehicles based on License Plate Recognition (LPR) image information are described herein. The distance traversed by a vehicle between image frames is estimated based on a difference between a pixel location of a tracking point on the vehicle in a first image and a second pixel location of the tracking point in second image of the vehicle. In some embodiments, the displacement identified in the images is converted to a projected displacement on the roadway surface based on the mapping between pixel locations within the field of view of the image sensor and locations on the surface of the roadway. The projected displacement is corrected by accounting for the difference in height of the camera module and tracking point over the roadway surface. An estimate of vehicle speed is calculated from the corrected displacement and the elapsed time between image frames.