Building Detection from Aerial Imagery using Inception Resnet UNET and UNET Models

authored by

S. Aghayari, A. Hadavand, S. Mohamadnezhad Niazi, Mohammad Omidalizarandi

Abstract

Buildings are one of the key components in change detection, urban planning, and monitoring. The automatic extraction of the building from high-resolution aerial imagery is still challenging due to the variations in their shapes, structures, textures, and colours. Recently, the convolutional neural networks (CNN) show a significant improvement in object detection and extraction that surpasses other methods. To extract building, in this paper two segmentation architectures, the UNet and the Inception ResNet UNet are implemented and then tested on the Inria aerial image datasets. The Inception ResNet UNet utilizes the Inception architecture and residual blocks. This makes the model wide and deep, though there are a few differences between numbers of UNet and Inception ResNet UNet parameters. The analyses show that UNet has a high rate of metrics in the training progress. However, on the unseen dataset, Inception ResNet UNet extracts buildings more accurately (97.95% accuracy and 0.96 in the dice metric) in comparison with UNet (94.30% accuracy and 0.55 in the dice metric).

Organisation(s)

Geodetic Institute

External Organisation(s)

Ideh Pardazan Tosseah Consulting Engineering Company

Type

Conference article

Journal

ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences

Volume

10

Pages

9-17

No. of pages

9

ISSN

2194-9042

Publication date

13.01.2023

Publication status

Published

Peer reviewed

Yes

ASJC Scopus subject areas

Environmental Science (miscellaneous), Instrumentation, Earth and Planetary Sciences (miscellaneous)

Electronic version(s)

https://doi.org/10.5194/isprs-annals-X-4-W1-2022-9-2023 (Access: Open)

 

Details in the research portal "Research@Leibniz University"