Image-Based Vehicle Emission Measurement Without Dedicated Infrastructure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating vehicle emissions are inaccurate and require costly infrastructure, especially in large-scale applications.
Innovation Solution
A method and apparatus that utilize image processing and machine learning models to estimate vehicle emissions by analyzing images from vehicles on a roadway, determining vehicle quantity, model, and type, and calculating emission factors based on these characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct measurement infrastructure is established for vehicle emissions, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent uses image copying technology to capture visual representations of vehicles and their exhaust plumes. Instead of requiring complex direct measurement infrastructure, the system creates digital copies (images) of the emission sources and processes these copies through machine learning models to estimate emissions, thereby resolving the contradiction between measurement precision and device complexity
Solution Approach 2:
The patent replaces traditional mechanical measurement infrastructure with an optical-digital system. Image processing and machine learning algorithms substitute for physical measurement devices, allowing emission estimation without requiring complex infrastructure while maintaining reasonable measurement precision
2Productivity
If traffic level data is used for large scale emission estimation, then productivity is improved, but measurement precision deteriorates
Solution Approach 1:
The patent applies local quality by using different estimation approaches for different vehicles based on their visible characteristics. Instead of treating all traffic uniformly, the system identifies specific vehicle types, models, and conditions from images, applying localized emission factors to each vehicle class, thereby improving both precision and productivity simultaneously
Data Source
AI summary
Systems and methods are described for the vehicle emission measurement. An example method may include receiving a plurality of images from a first vehicle traveling on a section of roadway, determining a quantity of surrounding vehicles from the plurality of images, determining a cropped image of at least one of the surrounding vehicles from the plurality of images, identifying a model of the at least one of the surrounding vehicles from the cropped image, and calculating an emission measurement factor for the section of roadway based on at least the quantity of surrounding vehicles for the at least one of the surrounding vehicles.