- Python 100%
| Filename | Latest commit message | Latest commit date |
|---|---|---|
| dataset | ||
| runs/train/yolo11-efficientViT-cdw | ||
| LICENSE | ||
| README.md | ||
| train.py | ||
| val.py | ||
Overview
Code Implementation of "An Object Detection Algorithm for Construction and Demolition Waste Based on Cascaded Group Attention Mechanism"
DATASET
BTC
The BTC Dataset was collected and made publicly available by Demetriou et.al. It comprises samples extracted from manually sorted Construction and Demolition Waste (CDW) piles, covering three object categories: Bricks, Tiles, and Concrete. Using an RGB camera with a resolution of 1920×1200, the researchers gathered 550 JPG images (with approximately 4 samples per object category), resulting in around 6,600 samples across all object categories.
Download link for the dataset: https://data.mendeley.com/datasets/24d45pf8wm/1
SWP
The SWP Dataset was collected and made publicly available by Fang et.al. The researchers employed a high-definition industrial camera with a 2-megapixel resolution and a frame rate of 60 frames per second to capture the data. Approximately 20,000 images of construction waste were collected, encompassing three categories: Steel, Wood, and Plastic. Please contact the corresponding author for access to the dataset.
Citation
@article{demetriou2023real,
title={Real-time construction demolition waste detection using state-of-the-art deep learning methods; single--stage vs two-stage detectors},
author={Demetriou, Demetris and Mavromatidis, Pavlos and Robert, Ponsian M and Papadopoulos, Harris and Petrou, Michael F and Nicolaides, Demetris},
journal={Waste Management},
volume={167},
pages={194--203},
year={2023},
publisher={Elsevier}
}
@article{fang2024real,
title={Real-time detection of construction and demolition waste impurities using the improved YOLO-V7 network},
author={Fang, Haifeng and Chen, Junji and Wang, Mingqiang and Wu, Qunbiao and Wang, Zhen},
journal={Journal of Material Cycles and Waste Management},
volume={26},
number={4},
pages={2200--2213},
year={2024},
publisher={Springer}
}
Code:
train.py: Training process code, primarily used for hyperparameter configuration and main training logic construction.
val.py:Testing process code, primarily used for evaluating trained models.
dataset:The dataset configuration files are stored here.
runs: Training and testing results are stored here.