Driver recognition method, device and system based on user behavior

Abstract

The invention discloses a driver identification method, device and system based on user behavior. The method includes the following steps: collecting user data of a vehicle within a preset time period; performing data cleaning on the user data; adopting a word2vec model to encapsulate the user data A feature vector is formed; the feature vector is input into the driver recognition model corresponding to the vehicle, and the corresponding driver recognition result is obtained through recognition. The present invention uses the word2vec model to encapsulate the user data into a feature vector after the user data of the vehicle within a preset time period, so that the obtained feature vector is input into the pre-trained driver recognition model, and the driver can be identified and obtained. For car owners or non-car owners, this method does not need to add additional hardware equipment, and can directly identify the driver according to the user data of the vehicle. The application cost is low, convenient and fast, and can be widely used in the automotive intelligent industry.

Description

technical field [0001] The invention relates to the technical field of vehicle intelligence, in particular to a driver identification method, device and system based on user behavior. Background technique [0002] At present, the methods for driver identification generally include methods such as biometric method and vehicle driving data analysis method. Among them, the biometric method refers to the identification of the driver through the biological characteristics of the camera, fingerprints, voiceprints, and even irises. For example, the face recognition method through the camera needs to pre-record the driver's facial features and train the model. After that, the face needs to be aligned with the camera for a period of time before recognition can be performed. This solution is limited by light, orientation, cost and interaction consumption. time constraints. The vehicle driving data analysis method is a scheme that performs logistic regression analysis and driver iden...

AI Technical Summary

Problems solved by technology

Among them, the biometric method refers to the identification of the driver through the biological characteristics of the camera, fingerprints, voiceprints, and even irises. For example, the face recognition method through the camera needs to pre-record the driver's face features and train the model. After that, the face needs to be aligned with the camera for a period of time before recognition can be performed. This solution is limited by light, orientation, cost and interaction consumption. time constraints

The vehicle driving data analysis method is a scheme that performs logistic regression analysis and driver identification after collecting user data. However, in this type of scheme, the data are discrete, and the sequence relationship between each data cannot be expressed. In the end, It is not possible to accurately identify the driver based on the characteristics of the user's operation sequence

Images