---
task_categories:
- object-detection
tags:
- roboflow
- roboflow2huggingface
---
### Dataset Labels
```
['crack', 'stairstep_crack']
```
### Number of Images
```json
{'test': 11, 'valid': 433, 'train': 947}
```
### How to Use
- Install [datasets](https://pypi.org/project/datasets/):
```bash
pip install datasets
```
- Load the dataset:
```python
from datasets import load_dataset
ds = load_dataset("lipi17/building-cracks-merged", name="full")
example = ds['train'][0]
```
### Roboflow Dataset Page
[https://universe.roboflow.com/lipi-deepaakshi-patnaik-ktyz8/merged-building-cracks/dataset/1](https://universe.roboflow.com/lipi-deepaakshi-patnaik-ktyz8/merged-building-cracks/dataset/1?ref=roboflow2huggingface)
### Citation
```
@misc{ merged-building-cracks_dataset,
title = { Merged-Building-Cracks Dataset },
type = { Open Source Dataset },
author = { Lipi Deepaakshi Patnaik },
howpublished = { \\url{ https://universe.roboflow.com/lipi-deepaakshi-patnaik-ktyz8/merged-building-cracks } },
url = { https://universe.roboflow.com/lipi-deepaakshi-patnaik-ktyz8/merged-building-cracks },
journal = { Roboflow Universe },
publisher = { Roboflow },
year = { 2023 },
month = { oct },
note = { visited on 2023-10-21 },
}
```
### License
MIT
### Dataset Summary
This dataset was exported via roboflow.com on October 21, 2023 at 12:21 PM GMT
Roboflow is an end-to-end computer vision platform that helps you
* collaborate with your team on computer vision projects
* collect & organize images
* understand and search unstructured image data
* annotate, and create datasets
* export, train, and deploy computer vision models
* use active learning to improve your dataset over time
For state of the art Computer Vision training notebooks you can use with this dataset,
visit https://github.com/roboflow/notebooks
To find over 100k other datasets and pre-trained models, visit https://universe.roboflow.com
The dataset includes 1391 images.
Cracks are annotated in COCO format.
The following pre-processing was applied to each image:
* Auto-orientation of pixel data (with EXIF-orientation stripping)
* Resize to 640x640 (Stretch)
No image augmentation techniques were applied.