Mask YOLO: a fine-grid method for multi-scale targets detection

Yu Nie

doi:10.1117/12.3006609

26 September 2023 Mask YOLO: a fine-grid method for multi-scale targets detection

Yu Nie

Proceedings Volume 12793, International Conference on Mechatronics and Intelligent Control (ICMIC 2023); 127930Y (2023) https://doi.org/10.1117/12.3006609
Event: International Conference on Mechatronics and Intelligent Control (ICMIC2023), 2023, Wuhan, China

Abstract

Multi-scale and multi-class target detection in complex scenes has always been a very difficult issue. In this paper, we propose a target detection network based on object detection and multi-scale context. The proposed method overcomes the interference caused by the wide dispersion of scale ranges and effectively counteracts the features extracted for targets of different sizes. It combines 3 scale information with Generalized Intersection over Union (GIoU) and FocalLoss to perform the object detection task more accurately. The experimental results show that the method we proposed can effectively improve the accessibility of detection for abnormal size and orientation targets. It successfully generates a mask that highlights the objects on the input image.

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

Citation Download Citation

Yu Nie "Mask YOLO: a fine-grid method for multi-scale targets detection", Proc. SPIE 12793, International Conference on Mechatronics and Intelligent Control (ICMIC 2023), 127930Y (26 September 2023); https://doi.org/10.1117/12.3006609

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