A loss-modified encoder-decoder network for image semantic segmentation of remotely sensed data

Jingping Pei

doi:10.1117/12.2668160

23 February 2023 A loss-modified encoder-decoder network for image semantic segmentation of remotely sensed data

Jingping Pei

Proceedings Volume 12551, Fourth International Conference on Geoscience and Remote Sensing Mapping (GRSM 2022); 1255113 (2023) https://doi.org/10.1117/12.2668160
Event: Fourth International Conference on Geoscience and Remote Sensing Mapping (GRSM 2022), 2022, Changchun, China

Abstract

Environment monitoring, traffic management, and autonomous driving systems are numerous industries that use semantic segmentation of remote observation sensing data. High-resolution remote sensing satellite images have a large application area, include a lot of object information, and are challenging to extract information features due to the rapid growth of remote sensing technology. Motivated by the recent success of Convolutional Neural Network (CNN) for image semantic segmentation, this paper explores the power of CNN feature learning ability for remote sensing images, and serviceable image segmentation results are provided. A conventional Encoder-Decoder segmentation network is proposed for remotely sensed data. To further boost the performance of semantic segmentation, a variety of segmentation losses are explained and utilized. Evaluation metrics are introduced, and a series of experiments are conducted. The results demonstrate our proposed loss-modified CNN model's competitive performance against other methods.

Citation Download Citation

Jingping Pei "A loss-modified encoder-decoder network for image semantic segmentation of remotely sensed data", Proc. SPIE 12551, Fourth International Conference on Geoscience and Remote Sensing Mapping (GRSM 2022), 1255113 (23 February 2023); https://doi.org/10.1117/12.2668160

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available