Crack Segmentation Dataset
The Crack Segmentation Dataset, available on Roboflow Universe, is an extensive resource designed for individuals involved in transportation and public safety studies. It is also beneficial for developing self-driving car models or exploring various computer vision applications. This dataset is part of the broader collection available on the Ultralytics Datasets Hub.
Watch: Crack segmentation using Ultralytics YOLOv9.
Comprising 4029 static images captured from diverse road and wall scenarios, this dataset is a valuable asset for crack segmentation tasks. Whether you are researching transportation infrastructure or aiming to enhance the accuracy of autonomous driving systems, this dataset provides a rich collection of images for training deep learning models.
Dataset Structure
The Crack Segmentation Dataset is organized into three subsets:
- Training set: 3717 images with corresponding annotations.
- Testing set: 112 images with corresponding annotations.
- Validation set: 200 images with corresponding annotations.
Applications
Crack segmentation finds practical applications in infrastructure maintenance, aiding in the identification and assessment of structural damage in buildings, bridges, and roads. It also plays a crucial role in enhancing road safety by enabling automated systems to detect pavement cracks for timely repairs.
In industrial settings, crack detection using deep learning models like Ultralytics YOLO11 helps ensure building integrity in construction, prevents costly downtimes in manufacturing, and makes road inspections safer and more effective. Automatically identifying and classifying cracks allows maintenance teams to prioritize repairs efficiently, contributing to better model evaluation insights.
Dataset YAML
A YAML (Yet Another Markup Language) file defines the dataset configuration. It includes details about the dataset's paths, classes, and other relevant information. For the Crack Segmentation dataset, the crack-seg.yaml file is maintained at https://github.com/ultralytics/ultralytics/blob/main/ultralytics/cfg/datasets/crack-seg.yaml.
ultralytics/cfg/datasets/crack-seg.yaml
# Ultralytics 🚀 AGPL-3.0 License - https://ultralytics.com/license
# Crack-seg dataset by Ultralytics
# Documentation: https://docs.ultralytics.com/datasets/segment/crack-seg/
# Example usage: yolo train data=crack-seg.yaml
# parent
# ├── ultralytics
# └── datasets
# └── crack-seg ← downloads here (91.2 MB)
# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
path: ../datasets/crack-seg # dataset root dir
train: train/images # train images (relative to 'path') 3717 images
val: valid/images # val images (relative to 'path') 112 images
test: test/images # test images (relative to 'path') 200 images
# Classes
names:
0: crack
# Download script/URL (optional)
download: https://github.com/ultralytics/assets/releases/download/v0.0.0/crack-seg.zip
Usage
To train the Ultralytics YOLO11n model on the Crack Segmentation dataset for 100 epochs with an image size of 640, use the following Python code snippets. Refer to the model Training documentation page for a comprehensive list of available arguments and configurations like hyperparameter tuning.
Train Example
from ultralytics import YOLO
# Load a model
# Using a pretrained model like yolo11n-seg.pt is recommended for faster convergence
model = YOLO("yolo11n-seg.pt")
# Train the model on the Crack Segmentation dataset
# Ensure 'crack-seg.yaml' is accessible or provide the full path
results = model.train(data="crack-seg.yaml", epochs=100, imgsz=640)
# After training, the model can be used for prediction or exported
# results = model.predict(source='path/to/your/images')
# Start training from a pretrained *.pt model using the Command Line Interface
# Ensure the dataset YAML file 'crack-seg.yaml' is correctly configured and accessible
yolo segment train data=crack-seg.yaml model=yolo11n-seg.pt epochs=100 imgsz=640
Sample Data and Annotations
The Crack Segmentation dataset contains a diverse collection of images captured from various perspectives, showcasing different types of cracks on roads and walls. Here are some examples:
-
This image demonstrates instance segmentation, featuring annotated bounding boxes with masks outlining identified cracks. The dataset includes images from different locations and environments, making it a comprehensive resource for developing robust models for this task. Techniques like data augmentation can further enhance dataset diversity. Learn more about instance segmentation and tracking in our guide.
-
The example highlights the diversity within the Crack Segmentation dataset, emphasizing the importance of high-quality data for training effective computer vision models.
Citations and Acknowledgments
If you use the Crack Segmentation dataset in your research or development work, please cite the source appropriately. The dataset was made available via Roboflow:
@misc{ crack-bphdr_dataset,
title = { crack Dataset },
type = { Open Source Dataset },
author = { University },
url = { https://universe.roboflow.com/university-bswxt/crack-bphdr },
journal = { Roboflow Universe },
publisher = { Roboflow },
year = { 2022 },
month = { dec },
note = { visited on 2024-01-23 },
}
We acknowledge the team at Roboflow for making the Crack Segmentation dataset available, providing a valuable resource for the computer vision community, particularly for projects related to road safety and infrastructure assessment.
FAQ
What is the Crack Segmentation Dataset?
The Crack Segmentation Dataset is a collection of 4029 static images designed for transportation and public safety studies. It's suitable for tasks like self-driving car model development and infrastructure maintenance. It includes training, testing, and validation sets for crack detection and segmentation tasks.
How do I train a model using the Crack Segmentation Dataset with Ultralytics YOLO11?
To train an Ultralytics YOLO11 model on this dataset, use the provided Python or CLI examples. Detailed instructions and parameters are available on the model Training page. You can manage your training process using tools like Ultralytics HUB.
Train Example
from ultralytics import YOLO
# Load a pretrained model (recommended)
model = YOLO("yolo11n-seg.pt")
# Train the model
results = model.train(data="crack-seg.yaml", epochs=100, imgsz=640)
# Start training from a pretrained model via CLI
yolo segment train data=crack-seg.yaml model=yolo11n-seg.pt epochs=100 imgsz=640
Why use the Crack Segmentation Dataset for self-driving car projects?
This dataset is valuable for self-driving car projects due to its diverse images of roads and walls, covering various real-world scenarios. This diversity improves the robustness of models trained for crack detection, which is crucial for road safety and infrastructure assessment. The detailed annotations aid in developing models that can accurately identify potential road hazards.
What features does Ultralytics YOLO offer for crack segmentation?
Ultralytics YOLO provides real-time object detection, segmentation, and classification capabilities, making it highly suitable for crack segmentation tasks. It efficiently handles large datasets and complex scenarios. The framework includes comprehensive modes for Training, Prediction, and Exporting models. YOLO's anchor-free detection approach can improve performance on irregular shapes like cracks, and performance can be measured using standard metrics.
How do I cite the Crack Segmentation Dataset?
If using this dataset in your work, please cite it using the provided BibTeX entry above to give appropriate credit to the creators.
