Open AccessA new deep distortion convolutional neural network for semantic segmentation of panoramic images
Author(s) -
Xiaozhi Hu,
Yi An,
Cheng Shao,
Pan Ke Qin
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1873/1/012006
Subject(s) - artificial intelligence , computer science , segmentation , distortion (music) , convolutional neural network , computer vision , feature (linguistics) , deep learning , convolution (computer science) , image segmentation , encoder , pattern recognition (psychology) , artificial neural network , amplifier , linguistics , philosophy , bandwidth (computing) , operating system , computer network
Semantic segmentation of panoramic images plays a crucial role in many applications, such as scene understanding, autonomous navigation, and community security. However, the traditional deep learning algorithms achieve lower accuracy for panoramic images due to the serious image distortion. This paper proposes a new deep convolutional neural network to segment panoramic images of outdoor scenes semantically. The proposed network includes two branches: the semantic segmentation sub-branch (SS-branch) and the feature enhancement sub-branch (FE-branch). The SS-branch contains two parts: Encoder and decoder. In the encoder, a new distortion convolution layer (DCL) is defined to reduce the distortion of panoramic images. In the FE-branch, a canny is used to detect the edge and a lightweight network is designed to enhance the features of image boundaries. Finally, a weighted loss function is used. The experimental results show that the proposed semantic segmentation method has better performance for different outdoor scenes.