A method and apparatus for road identification

Abstract

A method and apparatus for road identification in video surveillance is disclosed in embodiments of the present application. The method comprises the steps of creating a symmetrical total convolutionneural network, using a training sample with corresponding annotation data, testing samples and calibrating data to optimize and adjust the parameters, and recognizing the road monitoring picture to be recognized by the symmetrical full convolution neural network after the parameter optimization and adjustment. As a result of applying the technical solution proposed in the embodiments of the present application, end-to-end road detection for each pixel is realized, which can solve the problem that the traditional scene adaptive road segmentation method based on DCNN can not effectively solve the road segmentation problem in the scene of road video surveillance, and improve the accuracy of road recognition in the video surveillance picture.

Description

technical field [0001] The present application relates to the field of video surveillance, in particular to a method and device for road identification. Background technique [0002] Video monitoring and processing is an important part of intelligent transportation system. In the practical application of road monitoring, such as congestion detection, road spill detection, etc., the road needs to be detected accurately before further processing can be carried out on this basis. [0003] The traditional method is based on background modeling technology and uses the road as the background to detect. The disadvantage is that it is easily affected by the weather, and the background update error is easy to update the target. In recent years, machine learning-based frameworks have been gradually introduced into road detection, where pixel blocks in an image are input into a classifier for "road" and "non-road" classification. However, due to the complexity and diversity of the sc...

AI Technical Summary

Problems solved by technology

[0011] The embodiment of the present application provides a road recognition method and device to realize the road recognition in the video surveillance screen through the symmetrical full convolutional neural network, and solve the problem that the traditional scene adaptive road segmentation method based on DCNN cannot effectively solve the problem of road video surveillance. The problem of road segmentation in the scene improves the accuracy of road recognition in the video surveillance screen

Images