Driving suitability detecting device and method

Abstract

The invention discloses a driving suitability detecting device and a detecting method. An integrated sealed-type spectacle structure is adopted, and the outer layer of the spectacle structure is applied with a black light-absorbing material, so that light leaking and pupil change caused by visual attention of human eye pupils can be effectively prevented; by applying a flexible silica gel materialto a part where spectacles fit to a face, the face can fit to the spectacles, and fatigue can be directly detected, so that the precision of fatigue detection can be improved; by stimulating pupils with a rectangular surface light source, the pupils can be stimulated by the light source uniformly; by irradiating the pupils with an infrared surface light source, interference of the infrared pointlight source to pupil image acquisition can be avoided, so that pupil imaging is more uniform and cannot get interfered easily; motion videos of the pupils are analyzed through a three-dimensional long-term deep convolutional learning network, and driver's task suitability is objectively classified and judged, so as to deeply understand the motion states of the pupils; therefore, end-to-end detection is achieved, and detection accuracy and driving suitability detecting efficiency are improved; and the device is simple in structure and is convenient and rapid to use.

Description

technical field [0001] The invention belongs to the field of medical and health technologies, and relates to a driving suitability detection device and method. Background technique [0002] With the rapid development of the transportation industry and the increase in vehicle ownership, how to reduce the number of road traffic accidents and ensure driving safety is one of the research hotspots of relevant researchers. According to statistics, the number of deaths due to road traffic accidents in the world exceeds 600,000 every year, causing direct economic losses of more than 1 billion. my country is one of the countries with a relatively high number of deaths in road traffic accidents. Among the many causes of road traffic accidents, driver factors account for a large proportion, and fatigue driving is the most important reason. Article 22 of the "Road Traffic Safety Law of the People's Republic of China" clearly stipulates: "Motor vehicle drivers...if excessive fatigue aff...

AI Technical Summary

Problems solved by technology

In 2000, Shi Jian from Shanghai Jiaotong University conducted research on fatigue driving through an objective method, and studied the relationship between fatigue and driving safety by using information parameters such as the situation of holding the steering wheel and the pedals stepped on during fatigue driving; Jilin University Prof. Wang Rongben judges the driver's fatigue by detecting the state of the driver's mouth, and uses the size of the mouth opening as the feature value to be extracted to form a feature vector to design the output of shutting up, talking and yawning. different states; on the basis of research on the PERCLOS algorithm, the Jiangsu University Research Institute uses an infrared imager and a differential imager with a wavelength of 850 / 950mm as the image collector for detection, and uses infrared light source imaging to eliminate the interference of ambient light sources. The accuracy of this measurement has been improved; Wu Qun from Zhejiang University used the electrocardiogram and eye state to detect the fatigue of the monitor. This method obtains the electrocardiogram of the human body in different states through experiments, and then classifies the collected samples through the kernel principal component method Then use the support vector data description method (SupportVector Data Description, SVDD) ECG data to classify, and finally combine the PERCLOSE principle to comprehensively judge the fatigue state of the driver for detection, but this kind of method is generally related to fatigue and is susceptible to Interference with eye disease (trachoma, dry eye, etc.); there are many more studies like this

Existing detection devices and detection methods perform fatigue detection through fatigue appearance, which is susceptible to interference and has low detection accuracy

Images