BRAILS (Building Recognition using AI at Large-Scale) provides a set of Python modules that utilize deep learning (DL), and computer vision (CV) techniques to extract information from satellite and street level images. The BRAILS framework will also provide turn-key applications that allow the users to put the individual modules together to determine multiple attributes in a single pass.
Online documentation is available at https://nheri-simcenter.github.io/BRAILS-Documentation.
This work is based on material supported by the National Science Foundation under grants CMMI 1612843 and CMMI 2131111.
NHERI-SimCenter [email protected]
The easiest way to install the latest version of BRAILS is using pip
:
pip install git+https://github.com/NHERI-SimCenter/BRAILS
This example demonstrates how to use the InventoryGenerator
method embedded in BRAILS to generate regional-level inventories.
The primary input to InventoryGenerator
is location. InventoryGenerator
accepts four different location input: 1) region name, 2) list of region names, 3) bounding box of a region, 4) A GeoJSON file containing building footprints.
Please note that you will need a Google API Key to run InventoryGenerator
.
#import InventoryGenerator:
from brails.InventoryGenerator import InventoryGenerator
# Initialize InventoryGenerator:
invGenerator = InventoryGenerator(location='Berkeley, CA',
nbldgs=100, randomSelection=True,
GoogleAPIKey="")
# Run InventoryGenerator to generate an inventory for the entered location:
# To run InventoryGenerator for all enabled attributes set attributes='all':
invGenerator.generate(attributes=['numstories','roofshape','buildingheight'])
# View generated inventory:
invGenerator.inventory
@misc{brails,
title={NHERI-SimCenter/BRAILS: Release v2.0.0},
journal={Zenodo},
author={Wang, Charles and Hornauer, Sascha and Cetiner, Barbaros and Guo, Yunhui and McKenna, Frank and Yu, Qian and Yu, Stella X. and Taciroglu, Ertugrul and Law, Kincho H.,
year={2021},
month={Mar},
url={https://zenodo.org/record/4570554}
doi="10.5281/zenodo.4570554"
}