Anomalous Road Sign Detection Using Compliance Score Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing internet-of-things (IoT) applications for measuring driving behaviors face challenges in ensuring accuracy and efficiency, particularly in identifying anomalous traffic control devices where non-compliant driving behaviors are frequently observed, leading to unnecessary expenses and potential safety issues.
Innovation Solution
A method and system for identifying anomalous traffic control devices by determining group and location compliance scores, selectively storing image data from locations with unusually high non-compliant driving behaviors, and using machine learning to improve precision and recall of driving behavior monitoring systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all captured data is reviewed to ensure system accuracy, then measurement precision is improved, but loss of time and productivity deteriorate
Solution Approach 1:
The patent extracts and identifies only the most relevant and anomalous data points for review by computing compliance scores and selecting locations with unusually high non-compliant driving behaviors. This selective extraction approach maintains measurement precision while significantly reducing the time and resources required for data review compared to examining all captured data.
Solution Approach 2:
The system performs self-service by automatically computing compliance scores, identifying anomalous locations, and selecting representative image data for review. This automated self-service mechanism reduces reliance on manual review of all data while maintaining system accuracy through intelligent selection of critical cases.
2Reliability
If comprehensive data collection is performed across all locations, then reliability is improved, but loss of substance and storage requirements worsen
Solution Approach 1:
The patent applies local quality by focusing data collection and storage efforts on specific locations with anomalous compliance characteristics rather than uniformly collecting data from all locations. By computing location non-compliance scores and identifying areas with unusually high non-compliant behaviors, the system concentrates storage resources on regions that most contribute to system reliability improvement.
Solution Approach 2:
The system performs partial action by selectively collecting and storing image data only from locations that meet specific compliance criteria. Rather than comprehensively collecting data from all locations, the patent applies partial action by targeting only those locations with anomalous compliance scores, thereby maintaining reliability while reducing overall data storage requirements.
3Measurement precision
If machine learning models are trained on selected anomalous data, then measurement precision is improved, but device complexity worsens
Solution Approach 1:
The patent applies preliminary action by pre-processing and selecting anomalous data points before training machine learning models. Compliance scores are computed and anomalous locations are identified in advance, creating a curated training dataset that improves detection accuracy. This preliminary preparation simplifies the overall system architecture by separating data selection from model training functions.
Solution Approach 2:
The system introduces an intermediary layer that computes compliance scores and selects representative image data between raw data collection and machine learning model training. This intermediary mechanism bridges the gap between comprehensive data collection and targeted model training, improving measurement precision while managing device complexity through modular architecture.
Data Source
Figure 1A~1B
Figure 2
Figure 3~4
AI summary
Systems and methods are provided for detecting and classifying driving behaviors, as well as techniques for improving such systems and methods, which may include identifying geographic locations where anomalous rates of non-compliant driving behaviors are observed, such as in the vicinity of certain traffic signs, traffic lights, and road markings.