Driver Monitoring rPPG Using Non-Shadowed Face Zones
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The insufficiency of computing power in driver monitoring systems (DMS) when applying the remote-PhotoPlethysmoGraphy (rPPG) algorithm is exacerbated by shadows on the driver's face, leading to inefficient resource utilization and increased power consumption.
Innovation Solution
A method and system that adaptively monitor and adjust face zones not in shadow, excluding shadowed areas from rPPG calculations to optimize computing power and resource usage, using a camera and processor to dynamically determine and adjust face zones based on signal-to-noise ratio (SNR) thresholds.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the rPPG algorithm processes all face zones including shadowed areas, then the heart beat signal recovery completeness is maintained, but the computing power consumption increases and processing speed decreases
Solution Approach 1:
The patent segments the driver's face into multiple face zones and applies different processing strategies to shadowed and non-shadowed zones. By dividing the face into regions and selectively processing only non-shadowed zones for rPPG calculations, the system reduces computational load while maintaining signal recovery accuracy through the aggregation of multiple zone signals.
Solution Approach 2:
The patent implements local quality assessment by evaluating shadow detection results for each face zone individually. Different quality metrics are applied to different zones based on their lighting conditions, with shadowed zones being excluded from rPPG calculations while non-shadowed zones undergo full processing. This localized approach optimizes resource allocation based on regional signal quality.
2Measurement precision
If the rPPG algorithm processes all face zones including shadowed areas, then the signal data completeness is maintained, but the processing time increases
Solution Approach 1:
The patent extracts and removes shadowed face zones from the rPPG processing pipeline by detecting shadows and excluding those specific zones from calculations. This extraction of problematic data regions prevents wasted computation on low-quality signals while maintaining completeness of useful signal data from non-shadowed zones through selective inclusion.
Solution Approach 2:
The patent applies partial action by processing only the necessary subset of face zones (non-shadowed areas) rather than all zones. By performing rPPG calculations exclusively on zones with adequate lighting conditions, the system achieves sufficient signal completeness for accurate heart beat recovery without the excessive processing time required for full-face analysis.
3Productivity
If shadowed face zones are excluded from rPPG calculations, then the computing efficiency is improved, but the risk of incomplete signal recovery increases
Solution Approach 1:
The patent merges signals from multiple non-shadowed face zones to reconstruct the heart beat signal. By combining data from several qualifying zones rather than relying on a single zone, the system maintains signal recovery reliability while improving computing efficiency through the exclusion of shadowed areas. The aggregation of multiple zone signals compensates for the reduced input data volume.
Solution Approach 2:
The patent implements feedback mechanisms by continuously monitoring the quality and sufficiency of signals from non-shadowed zones. The system assesses whether the available non-shadowed zones provide adequate signal strength and diversity for reliable heart beat recovery, and adjusts processing accordingly to maintain reliability while preserving computing efficiency gains.
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
This approach reduces calculation time and power consumption by focusing on non-shadowed face zones, enhancing the efficiency and effectiveness of heart beat signal recovery in DMS.
Implementation Method 1
The rPPG stands for remote-PhotoPlethysmoGraphy, or simply speaking, a remote heart rate estimation. It is a simple optical technique used to detect volumetric changes in blood in a peripheral circulation. The rPPG algorithm may measure the variance of red, green, and blue light reflection changes from the skin, without contacting.
Implementation Method 2
The rPPG algorithm measures a contrast between a specular reflection and a diffused reflection. The specular reflection is a pure light reflection from the skin
Implementation Method 3
The diffused reflection is a reflection that remains after absorption and scattering by skin tissue, which varies as blood volume changes.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The disclosure describes a method adapted for a driver monitoring system and the driver monitoring system. The method may comprise obtaining sequences of frames, wherein each sequence of frames consists of a plurality of color scalars associated with a plurality of face zones of a driver in a vehicle. For each sequence of frames, the method may determine whether a monitoring function is activated for a first set of face zones that are face zones in shadow; and may perform subsequent steps in response to determining the monitoring function being activated. In some embodiments, the subsequent steps may comprise monitoring the first set of face zones; adaptively determining a second set of face zones that are face zones not in shadow, based on results of the monitoring; and performing calculations for recovering heart beat signals only for the second set of face zones.