A locally-constrained YOLO framework for detecting small and densely-distributed building footprints

Yiqun Xie, Jiannan Cai, Rahul Bhojwani, Shashi Shekhar, Joe Knight

Research output: Contribution to journalArticle

Abstract

Building footprints are among the most predominant features in urban areas, and provide valuable information for urban planning, solar energy suitability analysis, etc. We aim to automatically and rapidly identify building footprints by leveraging deep learning techniques and the increased availability of remote sensing datasets at high spatial resolution. The task is computationally challenging due to the use of large training datasets and large number of parameters. In related work, You-Only-Look-Once (YOLO) is a state-of-the-art deep learning framework for object detection. However, YOLO is limited in its capacity to identify small objects that appear in groups, which is the case for building footprints. We propose a LOcally-COnstrained (LOCO) You-Only-Look-Once framework to detect small and densely-distributed building footprints. LOCO is a variant of YOLO. Its layer architecture is determined by the spatial characteristics of building footprints and it uses a constrained regression modeling to improve the robustness of building size predictions. We also present an invariant augmentation based voting scheme to further improve the precision in the prediction phase. Experiments show that LOCO can greatly improve the solution quality of building detection compared to related work.

Original languageEnglish (US)
JournalInternational Journal of Geographical Information Science
DOIs
StatePublished - Jan 1 2019

Fingerprint

footprint
solar energy
Urban planning
urban planning
Solar energy
learning
voting
Remote sensing
urban area
Availability
regression
experiment
prediction
Group
Experiments
spatial resolution
Deep learning
remote sensing
modeling
Object detection

Keywords

  • Building detection
  • deep learning
  • locally constrained
  • remote sensing
  • YOLO

Cite this

@article{8be354ac5af14d00bc59307c8f07f485,
title = "A locally-constrained YOLO framework for detecting small and densely-distributed building footprints",
abstract = "Building footprints are among the most predominant features in urban areas, and provide valuable information for urban planning, solar energy suitability analysis, etc. We aim to automatically and rapidly identify building footprints by leveraging deep learning techniques and the increased availability of remote sensing datasets at high spatial resolution. The task is computationally challenging due to the use of large training datasets and large number of parameters. In related work, You-Only-Look-Once (YOLO) is a state-of-the-art deep learning framework for object detection. However, YOLO is limited in its capacity to identify small objects that appear in groups, which is the case for building footprints. We propose a LOcally-COnstrained (LOCO) You-Only-Look-Once framework to detect small and densely-distributed building footprints. LOCO is a variant of YOLO. Its layer architecture is determined by the spatial characteristics of building footprints and it uses a constrained regression modeling to improve the robustness of building size predictions. We also present an invariant augmentation based voting scheme to further improve the precision in the prediction phase. Experiments show that LOCO can greatly improve the solution quality of building detection compared to related work.",
keywords = "Building detection, deep learning, locally constrained, remote sensing, YOLO",
author = "Yiqun Xie and Jiannan Cai and Rahul Bhojwani and Shashi Shekhar and Joe Knight",
year = "2019",
month = "1",
day = "1",
doi = "10.1080/13658816.2019.1624761",
language = "English (US)",
journal = "International Journal of Geographical Information Science",
issn = "1365-8816",
publisher = "Taylor and Francis Ltd.",

}

TY - JOUR

T1 - A locally-constrained YOLO framework for detecting small and densely-distributed building footprints

AU - Xie, Yiqun

AU - Cai, Jiannan

AU - Bhojwani, Rahul

AU - Shekhar, Shashi

AU - Knight, Joe

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Building footprints are among the most predominant features in urban areas, and provide valuable information for urban planning, solar energy suitability analysis, etc. We aim to automatically and rapidly identify building footprints by leveraging deep learning techniques and the increased availability of remote sensing datasets at high spatial resolution. The task is computationally challenging due to the use of large training datasets and large number of parameters. In related work, You-Only-Look-Once (YOLO) is a state-of-the-art deep learning framework for object detection. However, YOLO is limited in its capacity to identify small objects that appear in groups, which is the case for building footprints. We propose a LOcally-COnstrained (LOCO) You-Only-Look-Once framework to detect small and densely-distributed building footprints. LOCO is a variant of YOLO. Its layer architecture is determined by the spatial characteristics of building footprints and it uses a constrained regression modeling to improve the robustness of building size predictions. We also present an invariant augmentation based voting scheme to further improve the precision in the prediction phase. Experiments show that LOCO can greatly improve the solution quality of building detection compared to related work.

AB - Building footprints are among the most predominant features in urban areas, and provide valuable information for urban planning, solar energy suitability analysis, etc. We aim to automatically and rapidly identify building footprints by leveraging deep learning techniques and the increased availability of remote sensing datasets at high spatial resolution. The task is computationally challenging due to the use of large training datasets and large number of parameters. In related work, You-Only-Look-Once (YOLO) is a state-of-the-art deep learning framework for object detection. However, YOLO is limited in its capacity to identify small objects that appear in groups, which is the case for building footprints. We propose a LOcally-COnstrained (LOCO) You-Only-Look-Once framework to detect small and densely-distributed building footprints. LOCO is a variant of YOLO. Its layer architecture is determined by the spatial characteristics of building footprints and it uses a constrained regression modeling to improve the robustness of building size predictions. We also present an invariant augmentation based voting scheme to further improve the precision in the prediction phase. Experiments show that LOCO can greatly improve the solution quality of building detection compared to related work.

KW - Building detection

KW - deep learning

KW - locally constrained

KW - remote sensing

KW - YOLO

UR - http://www.scopus.com/inward/record.url?scp=85067053912&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067053912&partnerID=8YFLogxK

U2 - 10.1080/13658816.2019.1624761

DO - 10.1080/13658816.2019.1624761

M3 - Article

JO - International Journal of Geographical Information Science

JF - International Journal of Geographical Information Science

SN - 1365-8816

ER -