Ibrahim Ali, Isa, Namoun, Abdallah, Ullah, Sami, Alasmary, Hisham, Waqas, Muhammad ORCID: 0000-0003-0814-7544 and Ahmad, Iftekhar (2024) Infrared ship segmentation based on weakly-supervised and semi-supervised learning. IEEE Access, 12. pp. 117908-117920. ISSN 2169-3536 (Online) (doi:https://doi.org/10.1109/ACCESS.2024.3448301)

Preview

PDF (Open Access Article) 48665 WAQAS_Infrared_Ship_Segmentation_Based_On_Weakly-Supervised_And_Semi-Supervised_Learning_(OA)_2024.pdf - Published Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (2MB) | Preview

Abstract

Existing fully-supervised semantic segmentation methods have achieved good performance. However, they all rely on high-quality pixel-level labels. To minimize the annotation costs, weakly supervised methods or semi supervised methods are proposed. When such methods are applied to the infrared ship image segmentation, inaccurate object localization occurs, leading to poor segmentation results. In this paper, we propose an infrared ship segmentation (ISS) method based on weakly-supervised and semi supervised learning, aiming to improve the performance of ISS by combining the advantages of two learning methods. It uses only image-level labels and a minimal number of pixel-level labels to segment different classes of infrared ships. Our proposed method includes three steps. First, we designed a dual branch localization network based on ResNet50 to generate ship localization maps. Second, we trained a saliency network with minimal pixel-level labels and many localization maps to obtain ship saliency maps. Then, we optimized the saliency maps with conditional random fields and combined them with image-level labels to generate pixel-level pseudo-labels. Finally, we trained the segmentation network with these pixel-level pseudo-labels to obtain the final segmentation results. Experimental results on the infrared ship dataset collected on real sites indicate that the proposed method achieves 71.18% mean intersection over union, which is at most 56.72% and 8.75% higher than the state-of-the-art weakly-supervised and semi-supervised methods, respectively.

Item Type:	Article
Uncontrolled Keywords:	marine vehicles, accuracy, location awareness, training, semantic segmentation, object recognition, decoding, infrared imaging, supervised learning, semisupervised learning, infrared ship images, object segmentation, weakly-supervised learning, semi-supervised learning, pixel-level pseudo-labels
Subjects:	Q Science > Q Science (General) Q Science > QA Mathematics Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Faculty / School / Research Centre / Research Group:	Faculty of Engineering & Science Faculty of Engineering & Science > School of Computing & Mathematical Sciences (CMS)
Last Modified:	20 Nov 2024 13:41
URI:	http://gala.gre.ac.uk/id/eprint/48665

Actions (login required)

View Item

Downloads