Raise the Roof: Towards Generating Lod2 Models Without Aerial Surveys Using Machine Learning

Biljecki, Filip; Dehbi, Youness

doi:10.5194/isprs-annals-iv-4-w8-27-2019

Cited by 17 publications

(21 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Apart from this, most of the buildings in the city of Liege have pitched roofs. The potential of roof greening is relatively lower than other cities (Biljecki, Dehbi, 2019;Silva et al, 2017), mainly due to less number of flat roofs. Building stock in Wallonia region in general and in Liege is very old (Singh et al, 2013), resulting in large number of pitched roofs.…”

Section: Discussionmentioning

confidence: 91%

Potential for Urban Greening With Green Roofs: A Way Towards Smart Cities

Joshi

Selmi

Binard

et al. 2020

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. Cities are experiencing increased pressure on social, economic, and environmental sectors due to the rapid urbanisation and increasing risk owing to climate change affecting the urban environment. Solutions such as green roofs are often discussed in the context of smart and sustainable cities as they present a multi-functional and solution-oriented approach to address these challenges. Green roofs become extremely relevant in the context of highly urbanised and compact cities where impervious surfaces are abundant. Therefore, in this paper, we analyse the potential of green roofs at a city scale with the help of parameters such as area and slope of the roof and structure of the building. We also identify the priority zones based on environmental and socio-economic parameters. The study is carried out in the city of Liege, Belgium. The results suggest that around 20% (350 hectares) of the total buildings in the city have the potential for developing green roofs. Moreover, the potential of green roofs is quite significant in terms of roof area in the priority zone. Due to significant socio-economic deprivation in high priority zones, implementation of green roofs might not be affordable. Buildings with larger roof sizes are mostly owned by companies or commercial establishments, thus, making larger roofs more relevant for retrofitting green roof. Thus, our approach can act as a preliminary decision-making tool for urban planners to analyse the potential of green roofs and prioritize them in deprived areas.

show abstract

Section: Discussionmentioning

confidence: 91%

Potential for Urban Greening With Green Roofs: A Way Towards Smart Cities

Joshi

Selmi

Binard

et al. 2020

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

show abstract

“…As a result, they have reached a mean absolute error of 0.8 m in the inferred heights. Biljecki and Dehbi (2019) demonstrated that it is possible to predict the roof types from lower LoD (i.e., LoD0 and LoD1) datasets and to generate LoD2 models without roof measurements. They achieved an accuracy of 85% of the roof type from sparse data using a multiclass classification and 92% accuracy in predicting whether a roof is flat or not.…”

Section: Machine Learning Methodsmentioning

confidence: 99%

Deep Learning for 3d Building Reconstruction: A Review

Buyukdemircioglu

Kocaman

Kada³

2022

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. 3D building reconstruction using Earth Observation (EO) data (aerial and satellite imagery, point clouds, etc.) is an important and active research topic in different fields, such as photogrammetry, remote sensing, computer vision and Geographic Information Systems (GIS). Nowadays 3D city models have become an essential part of 3D GIS environments and they can be used in many applications and analyses in urban areas. The conventional 3D building reconstruction methods depend heavily on the data quality and source; and manual efforts are still needed for generating the object models. Several tasks in photogrammetry and remote sensing have been revolutionized by using deep learning (DL) methods, such as image segmentation, classification, and 3D reconstruction. In this study, we provide a review on the state-of-the-art machine learning and in particular the DL methods for 3D building reconstruction for the purpose of city modelling using EO data. This is the first review with a focus on object model generation based on the DL methods and EO data. A brief overview of the recent building reconstruction studies with DL is also given. We have investigated the different DL architectures, such as convolutional neural networks (CNNs), generative adversarial networks (GANs), and the combinations of conventional approaches with DL in this paper and reported their advantages and disadvantages. An outlook on the future developments of 3D building modelling based on DL is also presented.

show abstract

“…Untuk objek bangunan, CityGML mendefinisikan tingkat perincian dari LOD0 sampai dengan LOD4. Secara umum, LOD tinggi lebih diinginkan dibandingkan dengan LOD rendah (Biljecki & Dehbi, 2019). Namun, penggunaan LOD yang lebih detail tidak menjamin peningkatan kualitas hasil analisis yang dilakukan dengan menggunakan model tersebut, dikarenakan terdapat faktor kesalahan lain seperti tingkat akurasi akuisisi data geometri (Biljecki et al, 2018).…”

Section: Pendahuluanunclassified

Program Konversi Peta Berstandar Citygml Dari Produk Pemetaan Tiga Dimensi (3d)

Veriandi¹,

Suwardhi²,

Trisyanti³

et al. 2021

SNG

View full text Add to dashboard Cite

CityGML merupakan model data yang bersifat terbuka untuk memodelkan wilayah perkotaan maupun lanskap secara 3D. Dari lima tingkat perincian/Levels of Detail (LOD) pada CityGML, LOD2 menjadi tingkat perincian yang penting karena ketersediaan informasi atap bangunan. Informasi ini merupakan syarat dasar data untuk pelaksanaan aplikasi seperti perencanaan detail maupun penataan lingkungan dan bangunan. Peta kota 3D yang dominan dengan bangunan dapat diperoleh dari metode pemetaan udara lidar maupun fotogrametri. Produk yang dihasilkan biasanya mempunyai struktur data dalam format CAD atau 3D-shapefile dan bukan dalam format CityGML. Makalah ini menjelaskan pembangunan program konversi dari 3D-shapefile menjadi peta 3D CityGML LOD2 berbasis bahasa pemrograman Python. Pembuatan program ini didasari argumen bahwa model bangunan CityGML LOD2 membutuhkan waktu yang lama untuk dihasilkan secara manual. Dengan demikian, program konversi ini dibuat untuk mengurangi waktu yang dibutuhkan dalam menulis file CityGML. Selain program untuk menulis data bangunan, dibentuk program lain yang dapat menggabungkan data dari program 3dfier dan 3Dtree sehingga didapatkan model CityGML wilayah lengkap dengan objek bangunan, lanskap, dan pohon individual. Sebagai percobaan program, pemodelan dilakukan pada wilayah ITB Jatinangor. Program yang dihasilkan berhasil membangun model bangunan LOD2 dengan batasan yakni kesalahan geometri yang dimiliki oleh data masukan juga dimiliki pada model hasil dari program.

show abstract

Raise the Roof: Towards Generating Lod2 Models Without Aerial Surveys Using Machine Learning

Cited by 17 publications

References 35 publications

Potential for Urban Greening With Green Roofs: A Way Towards Smart Cities

Potential for Urban Greening With Green Roofs: A Way Towards Smart Cities

Deep Learning for 3d Building Reconstruction: A Review

Program Konversi Peta Berstandar Citygml Dari Produk Pemetaan Tiga Dimensi (3d)

Contact Info

Product

Resources

About