TSCA-Net: Mars terrain segmentation based on category attention

Guangbin Huang; Li Yang; Haohao Zhang

doi:10.1117/12.3011483

7 December 2023 TSCA-Net: Mars terrain segmentation based on category attention

Guangbin Huang, Li Yang, Haohao Zhang

Proceedings Volume 12941, International Conference on Algorithms, High Performance Computing, and Artificial Intelligence (AHPCAI 2023); 1294148 (2023) https://doi.org/10.1117/12.3011483
Event: Third International Conference on Algorithms, High Performance Computing, and Artificial Intelligence (AHPCAI 203), 2023, Yinchuan, China

Abstract

With the continuous advancement of deep learning technology, its application domain has been expanding extensively. In the field of Mars exploration, terrain segmentation plays a vital role in enabling Mars probes to comprehend the intricacies of the Martian landscape, thereby providing crucial assistance in subsequent missions. Inadequate segmentation performance would have a profound impact on the successful execution of these subsequent missions. Currently, there is a prevalent occurrence of category segmentation errors in existing terrain segmentation methods. In response, we present a novel Mars Terrain Segmentation network named TSCA-Net, which integrates our innovative Category Attention mechanism to enhance the network's capability in prioritizing feature extraction. This approach significantly improves its discernment of diverse terrain categories when compared to the baseline.

(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Guangbin Huang, Li Yang, and Haohao Zhang "TSCA-Net: Mars terrain segmentation based on category attention", Proc. SPIE 12941, International Conference on Algorithms, High Performance Computing, and Artificial Intelligence (AHPCAI 2023), 1294148 (7 December 2023); https://doi.org/10.1117/12.3011483

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