Automatic Vehicle Classification via Unique Identifier Lookup

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing vehicle classification systems using machine vision become obsolete and prone to errors when new vehicles are introduced, as they require manual updates and do not automatically create new classes, leading to inefficiencies and inaccuracies.

Innovation Solution

An automatic learning system that extracts unique vehicle identifiers and visual features from images, looks up class information in a registration database, and creates new classes when necessary, allowing for continuous updating and outlier detection without manual intervention.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If manual updates are used to maintain vehicle classification databases, then classification accuracy can be maintained for known vehicles, but the system becomes obsolete and error-prone when new vehicles are introduced

Engineering Contradiction:

Improveclassification accuracyVSAvoidability to handle new vehicle classes

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The system automatically creates new vehicle classes by extracting visual features from captured images and comparing them against the database. When no matching class is found, the system autonomously generates a new class definition without requiring manual intervention, allowing the database to self-update with new vehicle types.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system continuously learns from new vehicle classifications by comparing visual features of captured vehicles against existing classes. The feedback loop enables the system to refine class definitions and adapt to new vehicle designs, improving classification accuracy over time while maintaining reliability.

Inventive Principle:

Principle #23Feedback

2Productivity

If an existing database of edge data is used for vehicle classification, then classification can be performed for known vehicles, but updating the database becomes cumbersome, time consuming, and susceptible to errors

Engineering Contradiction:

Improveclassification speedVSAvoidtime for database updates

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system automatically updates the vehicle classification database by extracting visual features from captured images and creating new classes when necessary. This self-updating mechanism eliminates the need for manual database maintenance, making the system productive while minimizing time loss.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system performs preliminary classification by comparing visual features against existing classes before creating new classes. This preliminary action filters out vehicles that match known classes, reducing the need for database updates and saving time.

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If manual classification refinement is used, then classification accuracy can be improved, but the process does not automatically create new classes and requires human intervention

Engineering Contradiction:

Improveclassification precisionVSAvoidautomatic class creation

Core Design Contradiction:

Measurement precisionVSExtent of automation

Solution Approach 1:

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system replaces manual classification refinement with an automated computer vision-based classification mechanism. The processor automatically extracts visual features, compares them against existing classes, and creates new classes when necessary, eliminating the need for manual intervention while maintaining precision.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

4Adaptability or versatility

If the system continuously learns and creates new classes, then the database remains current, but the system complexity increases

Engineering Contradiction:

Improveability to adapt to new vehiclesVSAvoidsystem complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system segments the classification process into distinct modules: image capture, visual feature extraction, database querying, class matching, and new class creation. This segmentation manages complexity by organizing functions into manageable, independent components that can operate autonomously.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system uses a universal visual feature extraction process that works for all vehicle types. The same image processing and feature comparison mechanisms handle both known and new vehicles, reducing system complexity while maintaining adaptability.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS10657809B2Automatic learning for vehicle classification

Publication Date: 2020.05.19 3M INNOVATIVE PROPERTIES CO

AI summary

The present disclosure provides an automatically updating vehicle classification system and method. The system comprises a processor which extracts from a vehicle image at least one of: a unique vehicle identifier from the vehicle image and visual features of the vehicle in the vehicle image. If the visual features are below a probability threshold for matching a vehicle class in a local database, the processor looks up the unique vehicle identifier in a registration database. The registration database stores vehicle registration information including unique vehicle identifiers and associated vehicle class information. If the vehicle class information associated with the vehicle identifier is not a class recognized by the processor, the processor creates a new vehicle class associated with the visual features of the vehicle.