Driver Fatigue Detection Using Micro-Expressions and Voice Cues
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing fatigue detection systems relying solely on facial images face challenges such as the need for large datasets, misjudgment due to low resolution images, and incorrect identification of yawning as speaking, leading to inaccurate fatigue assessments.
Innovation Solution
A system that combines an image sensor to capture facial images and a voice sensor to collect driver voices, utilizing micro-expression feature extraction, super-resolution processing, and voice recognition algorithms to determine the driver's mental state, integrating a generative adversarial network for improved training data diversity and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a machine learning model is utilized for fatigue detection, then fatigue detection capability is achieved, but a large number of facial images are required for training
Solution Approach 1:
The patent uses Generative Adversarial Networks (GANs) to generate synthetic facial images that replicate real facial expressions and characteristics. These generated images serve as copies of real training data, allowing the system to create additional training samples without collecting more real facial images, thereby reducing the quantity requirement while maintaining model training effectiveness
2Ease of manufacture
If lower resolution facial images are utilized for feature extraction and training, then data collection is easier, but fatigue detection results are not as good as expected
Solution Approach 1:
The patent implements super-resolution processing to transform low-resolution facial images into high-resolution images by enhancing image quality and detail. This parameter change in image resolution allows the system to maintain ease of data collection while achieving accurate fatigue detection results through improved image quality for feature extraction
3Ease of operation
If mouth opening action is used to determine yawning, then detection is simple, but speaking action is misidentified as yawning
Solution Approach 1:
The patent divides the facial region into multiple segments including eyes, eyebrows, nose, and mouth, and analyzes micro-expression features across these different segments. By examining coordinated changes across multiple facial segments rather than relying solely on mouth opening, the system can distinguish between yawning and speaking actions, improving detection accuracy while maintaining operational simplicity
Data Source
AI summary
A system of fatigue detection includes an image sensor, a voice sensor, a memory and a processor. The image sensor is configured to capture at least one facial image of a driver. The voice sensor is configured to collect a voice of the driver. The memory is configured to store the at least one facial image and the voice. The processor is configured to extract at least one micro-expression feature from the at least one facial image, and establish a fatigue detection model based on the at least one micro-expression feature, and utilize the fatigue detection model to obtain a fatigue detection result. The processor is further configured to utilize a voice detection algorithm to recognize the voice to obtain a voice recognition result. The processor is further configured to determine a mental state of the driver based on the fatigue detection result and the voice recognition result.