Fatigue detection system based on machine vision and fatigue detection method thereof

Abstract

The invention discloses a fatigue detection system based on machine vision and a fatigue detection method thereof. Facial micro-expressions are tracked and captured in real time according to presentation of micro-signals generated by people under the fatigue state on the facial micro-expressions, the captured micro-expressions are matched with a pre-established facial micro-expression model, the fatigue determination parameter F is extracted by analyzing the matching degree and heart rate variability analysis is also performed, and the final fatigue level is obtained by integration of two types of analysis according to a certain weight so that the possible error can be eliminated to the greatest extent, and the current fatigue state of the human body can be accurately judged. The operation is simple and the cost is low, and the micro fatigue signals can still be accurately captured so as to play an important role in enhancing the work efficiency, reducing accidents and protecting human health and avoiding falling sick from overwork.

Description

technical field [0001] The invention belongs to the technical field of image processing, relates to the detection of micro-expressions of human bodies, and in particular to a fatigue detection system and method based on machine vision. Background technique [0002] With the rapid development of economy and technology, the pace of people's life is constantly accelerating, and the long-term continuous study or work mode has become the normal life of people today. This fast-paced lifestyle is promoting social development and human progress. At the same time, it inevitably brings some side effects. In fact, people's physiological endurance is limited, and long-term uninterrupted work will reduce people's physiological functions and cause the body to feel fatigued. Working or studying in a state of fatigue will make it difficult for people to concentrate, resulting in a decrease in learning or work efficiency, and may even cause accidents in some situations. Simultaneously, if ...

AI Technical Summary

Problems solved by technology

The accuracy of fatigue discrimination based on physiological signals (EEG signals, ECG signals, etc.) is high, and relatively subtle signals can be captured, which can meet the requirements for human fatigue detection to a certain extent, but traditional contact physiological signals The collection method has brought a lot of inconvenience to practical applications and has obvious limitations; the fatigue discrimination method based on the characteristics of human physiological responses refers to the use of human eye characteristics, mouth movement characteristics, etc. to infer the fatigue state of the subject. It is considered to be an important feature to reflect fatigue. Blink amplitude, blink frequency, average closing time, and yawning movements can all be directly used to detect fatigue, but usually a more accurate judgment result can only be given when a person is deeply fatigued.

Cannot meet the requirements for detecting mild fatigue state

Images