Deep learning-based ship driver fatigue detection method and system

Abstract

The invention discloses a deep learning-based ship driver fatigue detection method and system. The method comprises the following steps: acquiring a current video frame image; carrying out face detection by an improved Retinaface face detection network, and marking five face key points of the left eye, the right eye, the nose tip, the left mouth corner and the right mouth corner at the same time; cutting adaptively eyes and mouth pictures according to the positions of the five face key points, and identifying the opening and closing states of the eyes and the mouth by an improved ShuffleNet v2 convolutional neural network; calculating PERCLOS parameters of the eyes and the mouth; and comprehensively judging whether a driver is fatigued or not through a random forest model by fusing the PERCLOS parameters of the eyes and the mouth. Face detection and key point positioning can be rapidly realized, manual picture information extraction is not needed, the opening and closing states of the eyes and the mouth can be automatically identified, and eye and mouth feature parameters are fused based on the idea of multi-feature fusion, so that the fatigue state of the ship driver can be rapidly and accurately detected.

Description

technical field [0001] The invention belongs to the technical field of shipping, in particular to a method and system for detecting ship driver fatigue based on deep learning. Background technique [0002] my country is a big ocean country, and the shipping industry is developing very rapidly. In my country, more than 90% of foreign trade materials are shipped by sea. With the rapid development of modern technology and shipbuilding technology, the overall strength of my country's shipping industry has been significantly improved. However, there are still many safety problems in the process of rapid development of the shipping industry, and safety accidents occur from time to time, which pose a great threat to crew members, ships and the marine environment. According to statistics, 80% of marine safety accidents are related to human factors, among which marine accidents caused by fatigue driving account for a considerable proportion. In this context, it is particularly imp...

AI Technical Summary

Problems solved by technology

Advantages: fast speed, better recognition effect under good lighting conditions; Disadvantages: unable to adapt to complex lighting conditions, single application scene

Advantages: The speed is also faster, and the robustness of the model is enhanced; Disadvantages: It is impossible to distinguish the degree of eye opening, and the effect of fatigue recognition for small eyes is very poor

Advantages: accurate face positioning, accurate fatigue classification; Disadvantages: MTCNN face detection speed is relatively slow, the algorithm is not robust, and is easily affected by factors such as lighting changes, wearing glasses, and side faces

Images