TJDe-Net

TJDehaze Dataset & TJDe-Net model

Environment Requirements

The model has been tested and is confirmed to work in the following environment:

• Python: 3.7
• PyTorch: 1.10.2
• CUDA: 11.3

---

Pretrained Model

The pretrained model is saved in the following directory: TJDe-Net/checkpoint/outdoor

---

Test Dataset

The test dataset is located in: TJDe-Net/data/TJDehaze/test The dataset contains:

• GT: Ground truth images (original images).
• hazy: Hazy images generated using a monocular depth estimation method combined with the atmospheric imaging principle.

---

Directory Structure

TJDe-Net/
│
├── checkpoint/
│   └── outdoor/        # Pretrained model weights
│       └── TJDeNet.pth # Pretrained model file
│
├── data/
│   └── TJDehaze/
│       ├── test/       # Test dataset
│       │   ├── GT/     # Ground truth images
│       │   │   └── 1_0.85_0.25.jpg   # Example ground truth image
│       │   └── hazy/   # Hazy images
│       │       └── 1_0.85_0.25.jpg   # Example hazy image
│
├── result/             # Directory for saving test results
│
└── test.py             # Testing script

---

Test Results

After running the model, the test results will be saved in the result directory.

---

Usage

To test the model, use the following command:

python test.py --model TJDeNet --dataset TJDehaze --exp outdoor