Camouflaged Object Detection via Complementary Information-Selected Network Based on Visual and Semantic Separation

Camouflaged object detection (COD) is a promising yet challenging task that aims to segment objects concealed within intricate surroundings, a capability crucial for modern industrial applications. Current COD methods primarily focus on the direct fusion of high-level and low-level information, with...

Full description

Saved in:
Bibliographic Details
Published inIEEE transactions on industrial informatics pp. 1 - 11
Main Authors Yin, Chao, Yang, Kequan, Li, Jide, Li, Xiaoqiang, Wu, Yifan
Format Journal Article
LanguageEnglish
Published IEEE 25.07.2024
Subjects
Binary segmentation
camouflaged object detection (COD)
Data mining
deep learning
Feature extraction
Image segmentation
Object detection
Semantics
Task analysis
Visualization
Online AccessGet full text

Cover

More Information
Summary:Camouflaged object detection (COD) is a promising yet challenging task that aims to segment objects concealed within intricate surroundings, a capability crucial for modern industrial applications. Current COD methods primarily focus on the direct fusion of high-level and low-level information, without considering their differences and inconsistencies. Consequently, accurately segmenting highly camouflaged objects in challenging scenarios presents a considerable problem. To mitigate this concern, we propose a novel framework called visual and semantic separation network (VSSNet), which separately extracts low-level visual and high-level semantic cues and adaptively combines them for accurate predictions. Specifically, it features the information extractor module for capturing dimension-aware visual or semantic information from various perspectives. The complementary information-selected module leverages the complementary nature of visual and semantic information for adaptive selection and fusion. In addition, the region disparity weighting strategy encourages the model to prioritize the boundaries of highly camouflaged and difficult-to-predict objects. Experimental results on benchmark datasets show the VSSNet significantly outperforms State-of-the-Art COD approaches without data augmentations and multiscale training techniques. Furthermore, our method demonstrates satisfactory cross-domain generalization performance in real-world industrial environments.
ISSN:1551-3203
1941-0050
DOI:10.1109/TII.2024.3426979