Hidden Path Selection Network for Semantic Segmentation of Remote Sensing Images

Targeting at depicting land covers with pixel-wise semantic categories, semantic segmentation in remote sensing images needs to portray diverse distributions over vast geographical locations, which is difficult to be achieved by the homogeneous pixel-wise forward paths in the architectures of existi...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2021-12
Main Authors: Yang, Kunping, Xin-Yi, Tong, Gui-Song, Xia, Shen, Weiming, Zhang, Liangpei
Format: Article
Language:English
Subjects:
Algorithms
Design optimization
Feature maps
Geographical locations
Image analysis
Image segmentation
Land cover
Pixels
Remote sensing
Semantic segmentation
Semantics
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Targeting at depicting land covers with pixel-wise semantic categories, semantic segmentation in remote sensing images needs to portray diverse distributions over vast geographical locations, which is difficult to be achieved by the homogeneous pixel-wise forward paths in the architectures of existing deep models. Although several algorithms have been designed to select pixel-wise adaptive forward paths for natural image analysis, it still lacks theoretical supports on how to obtain optimal selections. In this paper, we provide mathematical analyses in terms of the parameter optimization, which guides us to design a method called Hidden Path Selection Network (HPS-Net). With the help of hidden variables derived from an extra mini-branch, HPS-Net is able to tackle the inherent problem about inaccessible global optimums by adjusting the direct relationships between feature maps and pixel-wise path selections in existing algorithms, which we call hidden path selection. For the better training and evaluation, we further refine and expand the 5-class Gaofen Image Dataset (GID-5) to a new one with 15 land-cover categories, i.e., GID-15. The experimental results on both GID-5 and GID-15 demonstrate that the proposed modules can stably improve the performance of different deep structures, which validates the proposed mathematical analyses.
ISSN:2331-8422
DOI:10.48550/arxiv.2112.05220