Driver fatigue detection based on the long-term and short-term memory network

Abstract

The invention relates to a driver fatigue detection method based on a long-short-term memory network. The method comprises the following steps: 1) acquiring a driver face video image by means of infrared acquisition equipment; 2) utilizing a multi-task cascaded convolutional neural network to carry out face detection and feature point positioning, and obtaining a driver eye image sequence according to a geometrical relationship among the feature points; and 3) designing an end-to-end convolutional recurrent neural network, extracting human eye spatial characteristics, analyzing context relationships between adjacent image frames, and judging whether the driver is in a fatigue state or not by combining time sequence changes of the human eye image characteristics within a period of time. Results show that the method can accurately extract eye features under the conditions of poor light conditions, sunglasses worn by a driver and the like, and compared with a fatigue detection method based on a CNN combined PERCLOS standard, higher fatigue detection accuracy is obtained, and prediction of the driving state video level of the driver is realized.

Description

technical field [0001] The invention relates to a long-short-term memory network-based driver fatigue detection method, which is better than the prior art in terms of sensitivity, robustness and accuracy, has good detection performance, and belongs to the field of image processing and deep learning. Background technique [0002] Studies have shown that fatigue driving is one of the important causes of traffic accidents, and laws and regulations in various countries expressly prohibit fatigue driving. However, due to its obvious gradual change, concealment and subjective inhibition characteristics, it is difficult for regulators to monitor in time. If the driver is reminded in time when the driver is fatigued, the occurrence of traffic accidents will be effectively avoided. Therefore, the fatigue detection of drivers has great research significance and social value. [0003] Fatigue detection methods based on visual features have been widely concerned, but there are some di...

AI Technical Summary

Problems solved by technology

The fatigue detection method based on physiological signals has high detection accuracy in theory, but it is easily affected by environmental factors, detection instruments, instrument measurement accuracy, and the driver's personal driving habits in practical applications. Poor, the detection equipment is expensive, in the actual use process, the measurement of physiological signals needs to wear an instrument, usually a contact detection, which will cause inconvenience and interference to the driver, the operability is not strong, and it is difficult to promote on a large scale; based on The fatigue detection method of driving behavior characteristics will be affected by factors such as differences in driving habits of different drivers, vehicle models, and road conditions. Due to individual differences, it is difficult to determine the threshold of driver fatigue rating indicators, the parameters of fatigue characteristics, and the degree of fatigue. The correlation between the

Images