PyTorch implementation of "Weak Appearance Aware Hybrid Architecture for Leakage Defect Detection".
1. we propose a hybrid semantic segmentation network (named WDA-Net) to achieve pipeline leakage detection.
Model | Dataset | img_size | mIoU | Download |
---|---|---|---|---|
WDA-Net | PLD | 512x512 | 70.04 | model |
WDA-Net | NEU-Seg | 192x192 | 84.55 | model |
The PLD test set is available at [DataSet].
The code was tested on SUSE Linux, with Anaconda Python 3.8 and PyTorch v1.10.0. NVIDIA GPUs are needed for both training and testing. After installing Anaconda:
a. Create a conda virtual environment and activate it.
conda create -n WDA-Net python=3.8 -y
conda activate WDA-Net
b. Install PyTorch and torchvision following the official instructions. Here we use PyTorch 1.10.0 and CUDA 11.1. You may also switch to other version by specifying the version number.
conda install pytorch=1.10.0 torchvision -c pytorch
c. Install MMCV-full 1.5.0 following the official instructions.
pip install mmcv-full -f https://download.openmmlab.com/mmcv/dist/{cu_version}/{torch_version}/index.html
d. Install MMSegmentation.
pip install mmsegmentation
or
pip install git+https://github.com/open-mmlab/mmsegmentation.git
sh tools/dist_train.sh ${CONFIG_FILE} 1 [optional arguments]
sh tools/dist_train.sh ${CONFIG_FILE} ${GPU_NUM} [optional arguments]
sh tools/dist_test.sh ${CONFIG_FILE} 1 [optional arguments]