Driving scene classification method based on convolution neural network

Abstract

The invention discloses a driving scene classification method based on a convolution neural network, and the method comprises the following steps: collecting a road environment video image; carrying out the classification of a traffic scene, and building a traffic scene recognition database; extracting sample images of different driving scenes from the traffic scene recognition database, carryingout the feature extraction and multiple convolution training of the sample images through a deep convolution neural network, carrying out the rasterization of pixels, connecting the pixels to form a vector, inputting the vector into a conventional neural network, obtaining convolution neural network output, and achieving the deep learning of different driving scenes; carrying out the parameter optimization of a network structure of the built convolution neural network, obtaining a trained convolution neural network classifier, carrying out the adjustment of a traffic scene recognition model, and selecting an optimal mode as the standard of the traffic scene recognition model; carrying out the real-time collection of the image of a to-be-detected traffic scene, and inputting the image intothe traffic scene recognition model for the recognition of a road environment scene.

Description

technical field [0001] The invention relates to the technical field of self-driving cars, in particular to a method for classifying driving scenes of intelligent vehicles using convolutional neural network technology. Background technique [0002] In recent years, automobile intelligent technology has developed rapidly. In the grading standard of automobile intelligent technology, assisted driving technology and partial automatic driving technology have entered the stage of industrialization; conditional automatic driving and highly automated driving technology have entered the stage of testing and verification. Image processing and recognition technology is the key basic technology for intelligent driver assistance systems and unmanned vehicles to perceive the environment, and its applications are becoming more and more extensive. Based on the vehicle forward vision sensor, various road environment information can be accurately obtained. The vehicle can recognize differen...

AI Technical Summary

Problems solved by technology

[0003] However, the traffic scene is more complex, and has the characteristics of large intra-class distance and small inter-class distance in different traffic scenes.

Features must be extracted before traffic scene recognition. Due to the variability and complexity of traffic scene pictures, explicit feature extraction is not easy

Images