A road image semantic segmentation method based on a hybrid automatic encoder

Abstract

The invention discloses a road image semantic segmentation method based on a hybrid automatic encoder. The road image semantic segmentation method comprises the following steps of (10) sample set collection of dividing sample set images into a training set and a test set; (20) sample image preprocessing of performing size transformation and contrast normalization processing on the sample set images, and converting the sample set images to be processed into a standard form; (30) acquiring a hybrid automatic encoder network model, respectively training a sparse automatic encoder and a de-noisingautomatic encoder by utilizing the preprocessed training sample, extracting an intermediate encoding weight value and an intermediate decoding weight value, and constructing the hybrid automatic encoder network model by establishing a reasonable model arrangement sequence and a stacking form; and (40) performing road semantic segmentation, performing semantic segmentation on a road image shot bythe vehicle-mounted camera by using the hybrid automatic encoder network model. According to the invention, through a stacking form of the hybrid automatic encoder, the optimization description of image semantics is realized, a concise and effective semantic segmentation model is established, and better road semantic segmentation performance is obtained.

Description

technical field [0001] The invention belongs to the technical field of computer vision, is used for environment perception in automatic / assisted driving of vehicles, and plays an important role in autonomous driving of vehicles, in particular a road area detection semantic segmentation method with simple structure and strong feature description ability. Background technique [0002] Road detection is an important part of the environment perception of unmanned vehicles. Using computer vision technology to realize the semantic segmentation of environmental scenes, identify the semantic category of each area in the scene, and determine the road area in the scene, which is very important for unmanned vehicle navigation. It is of great significance to autonomous driving. At present, most road image semantic segmentation methods are transformed from classification models under the deep learning framework, and generate dense pixel-level prediction outputs through small-step convolu...

AI Technical Summary

Problems solved by technology

[0003] To sum up, the problems existing in the existing technology are: the network structure of the road semantic segmentation model is complex, the training is difficult, and the real-time performance is affected

Images