mmrotate-rr360-hiou
Bow Direction Detection Based on Angular Coding with Heading Intersection over Union Loss
Science Score: 67.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
✓DOI references
Found 1 DOI reference(s) in README -
✓Academic publication links
Links to: ieee.org -
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (5.4%) to scientific vocabulary
Repository
Bow Direction Detection Based on Angular Coding with Heading Intersection over Union Loss
Basic Info
- Host: GitHub
- Owner: WUTCM-Lab
- License: apache-2.0
- Language: Python
- Default Branch: main
- Size: 0 Bytes
Statistics
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
Bow Direction Detection Based on Angular Coding with Heading Intersection over Union Loss
简介
这是 Bow Direction Detection Based on Angular Coding with Heading Intersection over Union Loss 论文的代码实现
准确的船首方向检测对于船舶轨迹预测和港口监控至关重要。现有的船舶检测网络通常输出 180 度范围内的角度，而扩展到 360 度会引入周期性问题，影响旋转交集比率（Rotation Intersection over Union, RIoU）的准确性。本文提出了一种新的船首方向检测算法，扩展了网络输出至 360 度，并集成了航向交集比率损失（Heading Intersection over Union Loss, HIoU），以提高检测的准确性和鲁棒性。此外，还设计了一个 HIoU 损失函数，旨在改善船首方向识别并减少哈希码中的量化误差。该算法在三个数据集上进行了评估：FGSD、OHD-SJTU-S 和 OHD-SJTU-L。在 FGSD 数据集上，算法实现了 91.14% 的平均精度（mAP）。在 OHD-SJTU-S 数据集上，达到 63.3% 的 mAP50:95 和 90.7% 的船首方向预测准确率。在OHD-SJTU-L数据集上，mAP50:95 为 29.2%，准确率为 80.2%。
环境配置
```shell
假设已经安装mmengine、mmcv 2.x、mmdetection
git clone https://github.com/open-mmlab/mmrotate -b dev-1.x cd mmrotate export MMROTATE_HOME=$(pwd) pip install -v -e . ```
快速开始
训练
shell
cd $MMROTATE_HOME
python projects/HIoU/tools/train.py \
projects/HIoU/configs/hiou/rotated_rtmdet_l_l1_pscrn3_tsiou_csta2b0.2-3x-fgsd.py \
--d
测试自行训练的权重
shell
cd $MMROTATE_HOME
python projects/HIoU/tools/test.py \
projects/HIoU/configs/hiou/rotated_rtmdet_l_l1_pscrn3_tsiou_csta2b0.2-3x-fgsd.py \
path/to/epoch_xx.pth
Citation
bibtex
@ARTICLE{10946139,
author={Chen, Yaxiong and Liu, Jiang and Huang, Qiangqiang and Sun, Hao and Xiong, Shengwu and Lu, Xiaoqiang},
journal={IEEE Transactions on Geoscience and Remote Sensing},
title={Bow Direction Detection Based on Angular Coding with Heading Intersection over Union Loss},
year={2025},
volume={},
number={},
pages={1-1},
keywords={Accuracy;Object detection;Marine vehicles;Prediction algorithms;Encoding;Technological innovation;Optimization;Training;Artificial intelligence;Shape;360-Degree Angle Processing;Angle Encoding;Bow Direction Detection;Computer Vision},
doi={10.1109/TGRS.2025.3556480}}
Owner
- Name: WUTCM-Lab
- Login: WUTCM-Lab
- Kind: user
- Repositories: 1
- Profile: https://github.com/WUTCM-Lab
Citation (CITATION.cff)
cff-version: 1.2.0 message: "If you use this software, please cite it as below." authors: - name: "MMRotate Contributors" title: "OpenMMLab rotated object detection toolbox and benchmark" date-released: 2022-02-18 url: "https://github.com/open-mmlab/mmrotate" license: Apache-2.0
GitHub Events
Total
- Watch event: 3
- Push event: 10
- Create event: 3
Last Year
- Watch event: 3
- Push event: 10
- Create event: 3
Dependencies
- asynctest *
- codecov *
- coverage *
- cython *
- e2cnn *
- flake8 *
- imagecorruptions *
- interrogate *
- isort ==4.3.21
- kwarray *
- matplotlib *
- mmcv <2.1.0,>=2.0.0rc2
- mmdet <3.1.0,>=3.0.0rc2
- mmengine >=0.1.0
- numpy *
- parameterized *
- pycocotools *
- pytest *
- scikit-learn *
- scipy *
- six *
- terminaltables *
- torch *
- ubelt *
- wheel *
- xdoctest >=0.10.0
- yapf *
- cython *
- numpy *
- docutils ==0.16.0
- myst-parser *
- sphinx ==4.0.2
- sphinx-copybutton *
- sphinx_markdown_tables *
- sphinx_rtd_theme ==0.5.2
- mmcv >=2.0.0rc2,<2.1.0
- mmdet >=3.0.0rc2,<3.1.0
- mmengine >=0.1.0
- imagecorruptions *
- scikit-learn *
- scipy *
- e2cnn *
- mmcv >=2.0.0rc2
- mmdet >=3.0.0rc2
- mmengine >=0.1.0
- torch *
- torchvision *
- matplotlib *
- numpy *
- pycocotools *
- six *
- terminaltables *
- torch *
- asynctest * test
- codecov * test
- coverage * test
- cython * test
- flake8 * test
- interrogate * test
- isort ==4.3.21 test
- kwarray * test
- matplotlib * test
- parameterized * test
- pytest * test
- scikit-learn * test
- ubelt * test
- wheel * test
- xdoctest >=0.10.0 test
- yapf * test
- Markdown ==3.5.1
- Pillow ==10.1.0
- PyYAML ==6.0.1
- Pygments ==2.17.2
- addict ==2.4.0
- aliyun-python-sdk-core ==2.14.0
- aliyun-python-sdk-kms ==2.16.2
- certifi ==2023.11.17
- cffi ==1.16.0
- charset-normalizer ==3.3.2
- click ==8.1.7
- colorama ==0.4.6
- contourpy ==1.1.1
- crcmod ==1.7
- cryptography ==41.0.5
- cycler ==0.12.1
- fonttools ==4.45.1
- idna ==3.6
- importlib-metadata ==6.8.0
- importlib-resources ==6.1.1
- jmespath ==0.10.0
- kiwisolver ==1.4.5
- markdown-it-py ==3.0.0
- matplotlib ==3.7.4
- mdurl ==0.1.2
- mmcv ==2.0.1
- mmdet ==3.0.0rc5
- mmengine ==0.10.1
- model-index ==0.1.11
- numpy ==1.24.4
- opencv-python ==4.8.1.78
- opendatalab ==0.0.10
- openmim ==0.3.9
- openxlab ==0.0.29
- ordered-set ==4.1.0
- oss2 ==2.17.0
- packaging ==23.2
- pandas ==2.0.3
- platformdirs ==4.0.0
- pycocotools ==2.0.7
- pycparser ==2.21
- pycryptodome ==3.19.0
- pyparsing ==3.1.1
- python-dateutil ==2.8.2
- pytz ==2023.3.post1
- requests ==2.28.2
- rich ==13.4.2
- scipy ==1.10.1
- shapely ==2.0.2
- six ==1.16.0
- tabulate ==0.9.0
- termcolor ==2.3.0
- terminaltables ==3.1.10
- tomli ==2.0.1
- tqdm ==4.65.2
- typing_extensions ==4.8.0
- tzdata ==2023.3
- urllib3 ==1.26.18
- yapf ==0.40.2
- zipp ==3.17.0
- mkl-fft ==1.3.8
- mkl-random ==1.2.4
- mkl-service ==2.4.0
- torch ==1.11.0
- torchvision ==0.12.0