• The PCOD_1200 dataset can be found in this download link (Google Drive)

• Four images with different polarization angles can be found in this download link (Google Drive).

The paper was accepted by Engineering Applications of Artificial Intelligence on 12 October, 2023. The paper link is : - Link.

The paper was accepted by Engineering Applications of Artificial Intelligence on 5 February, 2025. The paper link is : - Link.

• Python 3.9
• Pytorch 1.12.0
• Torchvision 0.13.0
• Numpy 1.26.4

• The training and testing experiments are conducted using PyTorch with a single NVIDIA 3090ti GPU of 24 GB Memory.
• Please run

MyTest.py

• unpolar_rgb.m/untitled2.m/AOP_DOP_new.m: Stokes parameter image computation in MATLAB, with input consisting of four images

  at different polarization angles. Specific differences are detailed in the code.

• data_enhance.py: Data Augmentation, Augment the images in the dataset by horizontally and vertically flipping them.

• convert.py: Process the JSON files generated after using LabelMe for ground truth annotation for subsequent processing.

• Get_gt_from_json.py: Convert the output from the above convert.py script into a binary image (black and white).

• sal2edge:Generate object edges using Ground Truth(GT).

@article{WANG2024107303,
title = {IPNet: Polarization-based Camouflaged Object Detection via dual-flow network},
journal = {Engineering Applications of Artificial Intelligence},
volume = {127},
pages = {107303},
year = {2024},
issn = {0952-1976},
doi = {https://doi.org/10.1016/j.engappai.2023.107303},
author = {Xin Wang and Jiajia Ding and Zhao Zhang and Junfeng Xu and Jun Gao},
}

@article{WANG2025110245,
title = {Polarization-based Camouflaged Object Detection with high-resolution adaptive fusion Network},
journal = {Engineering Applications of Artificial Intelligence},
volume = {146},
pages = {110245},
year = {2025},
issn = {0952-1976},
doi = {https://doi.org/10.1016/j.engappai.2025.110245},
author = {Xin Wang and Junfeng Xu and Jiajia Ding},
}