Method for establishing traffic sign recognition model

The invention discloses a method for establishing a traffic sign recognition model, and the method comprises the following steps: S1, obtaining a traffic video image, intercepting a traffic sign image, and carrying out the summarization to form a training set; s2, simulating a traffic sign recognition condition appearing in an actual scene, and performing artificial step length on the training setthrough image processing operation; s3, performing preprocessing operation on the training set; s4, establishing an SSD model for traffic sign recognition; and S5, inputting the training set into theSSD model to complete the training of the model, and performing performance test by using the test set to finally obtain a traffic sign recognition model. The constructed traffic sign recognition model has high detection precision and accuracy, is high in robustness and high in recognition real-time degree, and can play an important role in multiple aspects such as advanced auxiliary driving andunmanned driving. 本发明揭示了一种交