Training and using methods of driving detection model, training and using devices of driving detection model, equipment and medium

Abstract

The embodiment of the invention discloses training and using methods of a driving detection model, training and using devices of the driving detection model, equipment and a medium. The training method comprises: acquiring a self-labeled training sample, wherein the training sample comprises a driver image sample and a face labeling result, a face key point labeling result and a safety belt labeling result in a driver image; extracting multi-scale image features from the self-labeled training sample by adopting a reference network in the initial driving detection model; and supervising and training the face detection branch network, the face key point detection branch network and the safety belt detection branch network in the initial driving detection model by adopting the multi-scale image features to obtain a target driving detection model. By adopting the scheme of the invention, a plurality of detection tasks are integrated in one model, and compared with a plurality of single-task detection models, the number of models is changed from three to one, which means that less storage space and memory are needed, so that the detection operation efficiency can be effectively improvedand the resource consumption can be reduced.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of computer vision, and in particular to a training and use method, device, equipment and medium of a driving detection model. Background technique [0002] With the rapid development of the economy and transportation industry, automobiles have gradually become popular in every household. The various traffic accidents that follow have become a serious problem to be faced at present, especially the road traffic accidents caused by fatigue driving account for a large proportion of all traffic accidents. Therefore, it is necessary to remind the driver in any unsafe state technology for safe driving. [0003] At present, the detection of drivers in the car is mainly based on traditional image processing, model recognition, and deep learning technology that consumes a lot of time and memory for training and detection. However, based on image processing or pattern recognition, there are...

AI Technical Summary

Problems solved by technology

However, based on image processing or pattern recognition, there are disadvantages such as poor environmental adaptability and low recognition accuracy.

Images