Traffic Event Detection Using Oscillation Signal Trajectory Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current traffic event detection systems face challenges in accurately distinguishing and detecting vehicles or pedestrians on infrastructure, particularly in multiple monitoring regions and confined spaces like bridges and tunnels, due to sensitivity to environmental conditions and the need for additional parameter calibration.
Innovation Solution
A traffic event detection apparatus utilizing a deep neural network to estimate trajectories and extract timestamps from oscillation signals induced by moving objects, enabling precise detection of traffic events through a combination of trajectory estimation, timestamp extraction, and event extraction units.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If conventional traffic detection methods are used in multiple monitoring regions and confined spaces, then detection coverage is achieved, but measurement precision deteriorates due to sensitivity to environmental conditions
Solution Approach 1:
The patent segments the oscillation signal into multiple components using signal decomposition techniques (such as empirical mode decomposition or wavelet transform). This allows the system to separate useful traffic event signals from noise and environmental interference, thereby maintaining high measurement precision across multiple monitoring regions and confined spaces like bridges and tunnels.
Solution Approach 2:
The patent introduces an intermediary processing layer that includes oscillation signal processing, trajectory estimation, and timestamp extraction modules. This intermediary system acts as a mediator between the raw sensor data and the final detection results, filtering out environmental noise and enhancing the precision of traffic event detection across diverse monitoring conditions.
2Adaptability or versatility
If conventional detection systems are deployed in confined spaces, then detection capability is provided, but device complexity increases due to need for additional parameter calibration
Solution Approach 1:
The patent implements self-service through automatic trajectory estimation and timestamp extraction algorithms that adapt to different monitoring environments without requiring manual parameter calibration. The system automatically adjusts to confined spaces like bridges and tunnels by processing oscillation signals through neural network-based trajectory estimation, eliminating the need for complex pre-calibration procedures.
Solution Approach 2:
The patent dynamically changes processing parameters based on the monitoring environment. The trajectory estimation module adjusts its parameters automatically based on the characteristics of the oscillation signals received, allowing the system to adapt to confined spaces and different monitoring regions without increasing device complexity or requiring manual calibration.
3Measurement precision
If detailed trajectory analysis is performed for each moving object, then measurement precision is improved, but loss of time increases due to extensive data processing
Solution Approach 1:
The patent performs preliminary action by extracting timestamps and estimating trajectories in real-time as oscillation signals are received, rather than performing detailed analysis on all data after collection. The trajectory estimation module continuously processes signals to maintain precise object distinction while minimizing data analysis time through efficient neural network inference.
Solution Approach 2:
The patent extracts only the essential information (timestamps and trajectory data) from the oscillation signals using dedicated extraction modules. This selective extraction approach maintains measurement precision for distinguishing moving objects while significantly reducing the time required for data processing by avoiding unnecessary analysis of redundant information.
Data Source
AI summary
An object of the present disclosure is to provide a traffic event detection apparatus, traffic event detection system, a method and a non-transitory computer readable medium capable of detecting traffic events correctly. A traffic event detection apparatus includes at least one memory configured to store instructions and at least one processor configured to execute the instructions to: estimate a trajectory of a moving object based on an oscillation signal by using deep neural network, while the oscillation signal is induced by traffic of the moving object; extract a timestamp of the moving object based on the trajectory of the moving object; and extract a part of the oscillation signal corresponding to the timestamp of the moving object.