Road scene semantic segmentation method based on convolutional neural network

Abstract

The invention discloses a road scene semantic segmentation method based on a convolutional neural network. The method comprises the steps: firstly building the convolutional neural network which comprises an input layer, a hidden layer, and an output layer, and the hidden layer is composed of 13 neural network blocks, 7 upsampling layers, and 8 cascade layers; inputting each original road scene image in the training set into a convolutional neural network for training to obtain 12 semantic segmentation prediction images corresponding to each original road scene image; calculating a loss function value between a set composed of 12 semantic segmentation prediction images corresponding to each original road scene image and a set composed of 12 single hot coding images processed by the corresponding real semantic segmentation image; obtaining an optimal weight vector and an optimal offset item of the convolutional neural network classification training model; inputting the road scene imageto be subjected to semantic segmentation into a convolutional neural network classification training model for prediction to obtain a corresponding predicted semantic segmentation image. The method has the advantage of high semantic segmentation precision.

Description

technical field [0001] The invention relates to a road scene semantic segmentation technology, in particular to a road scene semantic segmentation method based on a convolutional neural network. Background technique [0002] In recent years, advances in machines capable of performing computationally intensive tasks have allowed researchers to dig deeper into neural networks. Convolutional neural networks have achieved recent success in image classification, localization, and scene understanding. At present, due to the proliferation of tasks such as augmented reality and self-driving vehicles, many researchers have turned their attention to scene understanding. One of the main steps is semantic segmentation, which is to classify each pixel in a given image. . Semantic segmentation is of great significance in mobile and robotics related applications. [0003] Of course, object detection methods can help draw bounding boxes for certain entities, but human scene understanding...

AI Technical Summary

Problems solved by technology

But it has many shortcomings. First, the training is more troublesome. It needs to be trained three times to get FCN-8s; second, it is not sensitive enough to the details of the image. This is because in the decoding process, that is, the process of restoring the original image size, the input The label map (label image) of the sampling layer is too sparse, and the upsampling process is a simple deconvolution (deconvolution), so the obtained segmentation results are still not fine

Images