Traffic scene classification method based on multi-scale convolution neural network

Abstract

The invention discloses a traffic scene multi-target classification method, to be specific, discloses a traffic scene classification method based on a multi-scale convolution neural network. The traffic scene classification method is characterized in that recessive characteristics based on the multi-scale convolution neural network are extracted; and an optimal covering segmentation tree is acquired. During the realizing of the traffic scene classification, the multi-scale convolution neural network is adopted, and the excellent recessive characteristics having the invariance property are effectively extracted from an original image in different scales, and by comparing with the single-scale convolution neural network, the acquisition of the abundant and effective characteristic information of the image is realized. The effective information extracted by the convolution neural network is combined with the original segmentation tree of the image to form an optimal purity price tree, and the covering having the optimal purity is carried out, and therefore a clearer target contour is acquired, and the classification accuracy is increased. The RGB-D is used as the convolution neural network input, and by comparing with the conventional RGB convolution neural network input, the training characteristic is additionally provided with the depth information, and the classification of the input image is more accurate.

Description

technical field [0001] The invention belongs to the field of vehicle intelligent transportation, in particular to a method for classifying urban and suburban road traffic scenes. Background technique [0002] Vehicle intelligence is one of the three core technologies in the development of today's automobile industry. The classification of road traffic scenes is an important prerequisite and basis for improving the intelligence of intelligent vehicles and advanced driver assistance systems (ADAS). [0003] Traffic scene classification refers to the use of vehicle-mounted cameras to capture traffic scene images, using different machine learning methods to simulate the human visual perception process, and to classify and mark vehicles, pedestrians, roads, and environmental elements in the captured scene. At present, according to the depth of the hierarchical structure of the machine learning model, the classification methods of traffic scenes can be divided into methods based o...

AI Technical Summary

Problems solved by technology

The deep learning method can directly act on the original image data, can extract the hidden features that reflect the nature of the data, has sufficient model complexity, and can realize multi-object classification in traffic scenes, but the model structure of the deep learning method is diverse, and there are still a lot of problems. The development and optimization space, in the existing deep learning methods, the outline clarity and accuracy of classified images need to be improved

Images