Traffic sign recognition model training method and system, traffic sign recognition model recognition method and system, equipment and medium

Abstract

The invention provides a traffic sign recognition model training method and system, a traffic sign recognition model recognition method and system, equipment and a medium. The training method comprises the steps of acquiring a sample data set; training a traffic sign recognition model according to the sample data set, wherein the model comprises a down-sampling network, a feature extraction network and a sign prediction network, the down-sampling network comprises a plurality of down-sampling modules, the feature extraction network comprises a spatial attention module, a plurality of up-sampling modules and a weighting module, the spatial attention module is used for processing the feature image output by the predetermined down-sampling module to obtain a target size feature image, the plurality of up-sampling modules are respectively used for amplifying the feature images output by the corresponding down-sampling modules to a target size, the weighting module is used for obtaining anoverall feature image according to the target size feature image, and the sign prediction network is used for obtaining a prediction bounding box of the traffic sign in the corresponding training image according to the overall feature image. According to the invention, the accuracy and efficiency of traffic sign recognition can be improved.

Description

Technical field

[0001] The present invention relates to the field of artificial intelligence technology, particularly to a traffic sign recognition model training methods, identification methods, systems, devices and media. Background technique

[0002] With the rise of the automobile autopilot technology, TSR (traffic sign recognition, Traffic Sign Recognition) as an auxiliary smart driving extremely important part received widespread attention, its road traffic signs appearing in real-time identification in the process of moving vehicle and timely to remind the driver to make the correct driving behavior, in order to ensure traffic safety, prevent traffic accidents.

[0003] Prior to the depth of learning and development, traffic sign recognition is often based on traditional image processing algorithms to achieve, although the accuracy of the algorithm is rising, but there is poor generalization ability, long running time and other issues, from the practical application there ...

Images