Driver State Detection Using Event-Driven Vision Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional drowsy driving prevention devices require significant data processing to determine if a driver is dozing, making it difficult to quickly detect the driver's state and increasing power consumption.
Innovation Solution
A state detection device utilizing a dynamic vision sensor (DVS) with a first solid-state imaging device that detects events based on light amounts, allowing for rapid state detection and reduced data processing, combined with a state detection unit that analyzes these events to determine the driver's state.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If continuous image data is processed at a predetermined frame rate, then the driver's state can be detected with sufficient accuracy, but the data processing amount becomes large and detection speed decreases
Solution Approach 1:
The patent extracts only the essential information needed for driver state detection by using event-driven data from the solid-state imaging device. Instead of processing complete continuous image frames, the system extracts event data that represents changes in the driver's state, thereby reducing data processing amount while maintaining detection accuracy and improving detection speed.
2Measurement precision
If continuous image data is processed at a predetermined frame rate, then the driver's state can be detected with sufficient accuracy, but the power consumption increases
Solution Approach 1:
The system extracts only event-driven data representing changes in driver state rather than processing complete continuous image frames. This extraction approach significantly reduces the computational load and power consumption while maintaining sufficient detection accuracy for monitoring driver alertness.
3Productivity
If event-driven data processing is used, then the detection speed increases and power consumption decreases, but the data processing approach becomes different from conventional methods
Solution Approach 1:
The patent changes the fundamental parameter of data representation from continuous image frames to discrete event-driven data. This parameter change enables faster processing and lower power consumption. The system manages this complexity by using a state detection unit that processes event data according to specific detection criteria, making the complexity manageable while achieving superior performance.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables quicker detection of the driver's state with reduced power consumption by using event-driven data processing instead of continuous image data, improving the efficiency of drowsy driving prevention systems.
Implementation Method 1
a first solid-state imaging device provided with a plurality of pixels arranged in a matrix, the first solid-state imaging device that detects, according to a light amount incident on each of the pixels, occurrence of an event in the pixel
Data Source
AI summary
A state of a driver is detected more quickly. A state detection device according to an embodiment is provided with a first solid-state imaging device provided with a plurality of pixels arranged in a matrix, the first solid-state imaging device that detects, according to a light amount incident on each of the pixels, occurrence of an event in the pixel, and a state detection unit that detects a state of a driver on the basis of the occurrence of the event detected by the first solid-state imaging device.