Lightweight object detection algorithm for unmanned surface vehicle in inland river based on improved YOLOv8

Shigang Xiao; Ruixin Dong

doi:10.1117/12.3039331

4 September 2024 Lightweight object detection algorithm for unmanned surface vehicle in inland river based on improved YOLOv8

Shigang Xiao, Ruixin Dong

Author Affiliations +

Proceedings Volume 13259, International Conference on Automation Control, Algorithm, and Intelligent Bionics (ACAIB 2024); 132592J (2024) https://doi.org/10.1117/12.3039331
Event: Fourth International Conference on Automation Control, Algorithm, and Intelligent Bionics (ICAIB 2024), 2024, Yinchuan, China

Abstract

In order to address the challenge of inland water surface object detection, where small objects account for a large proportion and are prone to detection failures, limited GPU resources on embedded platforms, and deployment difficulties in unmanned surface vehicle (USV) object detection tasks, an improved lightweight YOLOv8 algorithm for surface object detection is proposed. Firstly, to mitigate noise interference and omission of small surface objects during downsampling in the backbone network, a Dilation-wise Residual (DWR) module is introduced to enhance the C2f module to efficiently acquire more contextual information and increase the model’s attention to small objects. In addition, given the complexity of the neck of the original network, a new feature fusion network, Multi-scale Screening-feature Pyramid Networks (MSFPN), is designed in this paper to reduce model complexity while maintaining accuracy. The experimental results demonstrate that the proposed algorithm achieves a 1.1% improvement in average accuracy (mAP0.5) for surface objects, with a 36.54% reduction in model parameters and a 14.81% reduction in computational load. It is proved that the proposed DM-YOLOv8 network model has better and more reliable performance in detecting the water surface of inland rivers and can meet the needs of safe autonomous navigation of USV with high real-time and flexibility requirements.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Shigang Xiao and Ruixin Dong "Lightweight object detection algorithm for unmanned surface vehicle in inland river based on improved YOLOv8", Proc. SPIE 13259, International Conference on Automation Control, Algorithm, and Intelligent Bionics (ACAIB 2024), 132592J (4 September 2024); https://doi.org/10.1117/12.3039331

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

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Object detection

Detection and tracking algorithms

Feature extraction

Feature fusion

Convolution

Performance modeling

Autonomous vehicles

Show All Keywords

Keywords/Phrases

Search In:

Publication Years