Published: Engineering Applications of Artificial Intelligence (SCIE | Q1)

Resource constraint crop damage classification using depth channel shuffling[Paper]

Md Tanvir Islam ¹, Safkat Shahrier Swapnil ², Md. Masum Billal ³, Asif Karim ^{4, *}, Niusha Shafiabady ⁵, Md. Mehedi Hassan ^{6, *}

| 1. Sungkyunkwan University, South Korea | 2. RUET, Bangladesh | 3. USTC, China | 4. Charles Darwin University, Australia | 5. Australian Catholic University, Australia | 6. Khulna University, Bangladesh || *Corresponding Authors |

---

CDC Dataset

Download Dataset

Dataset Folder Structure

dataset/
└── data/
    ├── train/
    │   ├── damaged/
    │   │   ├── image1.jpg
    │   │   ├── image2.jpg
    │   │   └── ...
    │   └── non_damaged/
    │       ├── image3.jpg
    │       ├── image4.jpg
    │       └── ...
    └── test/
        ├── damaged/
        │   ├── image101.jpg
        │   ├── image102.jpg
        │   └── ...
        └── non_damaged/
            ├── image103.jpg
            ├── image104.jpg
            └── ...

Model Weight

Download Model Weight (will upload soon)

Dependencies

Create conda environment

conda create --name lightCDC python=3.9

conda activate lightCDC

Install CUDA: Tested using CUDA 11.8

conda install pytorch==2.5.1 torchvision==0.20.1 torchaudio==2.5.1  pytorch-cuda=11.8 -c pytorch -c nvidia

Install other libraries

pip install -r requirements.txt

Training

python train.py --config <name_of_your_config> --batch_size <batch_size> --lr <learning_rate>

For other paramaetrs you can edit the configs.py file of the config folder.

Testing

Single Image Inference

python test.py --mode single --model_weight <model_weight_path> --image_path <your_image_path>

Multiple Image Inference (Folder)

python inference.py --mode multiple --model_weight <model_weight_path>

Cite this Paper

If you find our work useful in your research, please consider citing our paper and star ✨✨ this repository. Thank you!

@article{islam2025resource,
  title={Resource constraint crop damage classification using depth channel shuffling},
  author={Islam, Md Tanvir and Swapnil, Safkat Shahrier and Billal, Md Masum and Karim, Asif and Shafiabady, Niusha and Hassan, Md Mehedi},
  journal={Engineering Applications of Artificial Intelligence},
  volume={144},
  pages={110117},
  year={2025},
  publisher={Elsevier}
}