iLearn-Lab/TGRS24-HFCNet

🚀 HFCNet: Heterogeneous Feature Collaboration Network for Salient Object Detection in Optical Remote Sensing Images

Yutong Liu¹  Mingzhu Xu¹  Tianxiang Xiao¹  Haoyu Tang¹  Yupeng Hu^1✉  Liqiang Nie¹

¹Affiliation (Please update if needed)

Official implementation of HFCNet, a Heterogeneous Feature Collaboration Network for Salient Object Detection (SOD) in Optical Remote Sensing Images.

🔗 Journal: IEEE Transactions on Geoscience and Remote Sensing (TGRS), 2024
🔗 Task: Salient Object Detection (SOD)
🔗 Framework: PyTorch

---

📌 Model Information

1. Model Name

HFCNet (Heterogeneous Feature Collaboration Network)

---

2. Task Type & Applicable Tasks

• Task Type: Salient Object Detection / Remote Sensing
• Core Task: Salient object detection in optical remote sensing imagery
• Applicable Scenarios:
  • Remote sensing scene understanding
  • Aerial object detection
  • Environmental monitoring

---

3. Project Introduction

Salient Object Detection (SOD) in remote sensing images is challenging due to complex backgrounds, scale variations, and heterogeneous feature distributions.

HFCNet proposes a Heterogeneous Feature Collaboration framework, which:

• Integrates multi-level heterogeneous features
• Enhances feature interaction and collaboration
• Improves representation of salient objects across scales
• Strengthens robustness against background interference

---

4. Training Data Source

Supported datasets:

• ORSSD
• EORSSD
• ORSI

---

🚀 Pre-trained Weights

Initialization Weights

Download backbone weights:

• Swin Transformer
• VGG16

Place .pth files into:./pretrained

---

Trained Weights

Download trained model weights:

• Baidu Cloud: https://pan.baidu.com/s/1bVC4uxf3xKhLRcC08EQKMQ?pwd=hfcn

---

🚀 Training

1. Download datasets and pre-trained weights
2. Prepare dataset path lists (.txt files)
3. Update dataset paths in config files

Run training:

nohup python -u main.py --flag train --model_id HFCNet --config config/dataset_o.yaml --device cuda:0 > train_ORSSD.log &

nohup python -u main.py --flag train --model_id HFCNet --config config/dataset_e.yaml --device cuda:0 > train_EORSSD.log &

nohup python -u main.py --flag train --model_id HFCNet --config config/dataset_orsi.yaml --device cuda:0 > train_ORSI.log &

## 🚀 Testing
mkdir ./modelPTH-ORSSD
python main.py --flag test --model_id HFCNet --config config/dataset_o.yaml

mkdir ./modelPTH-EORSSD
python main.py --flag test --model_id HFCNet --config config/dataset_e.yaml 

mkdir ./modelPTH-ORSI
python main.py --flag test --model_id HFCNet --config config/dataset_orsi.yaml

## ⚠️ Notes
Designed for academic research
Performance depends on dataset characteristics
Requires GPU for efficient training

## 📝Citation
@ARTICLE{HFCNet,
  author={Liu, Yutong and Xu, Mingzhu and Xiao, Tianxiang and Tang, Haoyu and Hu, Yupeng and Nie, Liqiang},
  journal={IEEE Transactions on Geoscience and Remote Sensing}, 
  title={Heterogeneous Feature Collaboration Network for Salient Object Detection in Optical Remote Sensing Images}, 
  year={2024},
  volume={62},
  pages={1-14}
}