Zaznavanje stavb z uporabo nevronskih mrež, učenih s prenosom znanja

Building detection with convolutional networks trained with transfer learning

Author(s):

Simon Šanca, Krištof Oštir, Alen Mangafić

Abstract:

Building footprint detection based on orthophotos can be used to update the building cadastre. In recent years deep learning methods using convolutional neural networks have been increasingly used around the world. We present an example of automatic building classification using our datasets made of colour near-infrared orthophotos (NIR-R-G) and colour orthophotos (R-G-B). Building detection using pretrained weights from two large scale datasets Microsoft Common Objects in Context (MS COCO) and ImageNet was performed and tested. We applied the Mask Region Convolutional Neural Network (Mask R-CNN) to detect the building footprints. The purpose of our research is to identify the applicability of pre-trained neural networks on the data of another colour space to build a classification model without re-learning.

Keywords:

deep learning, convolutional neural networks, classification of buildings, Mask R-CNN, object detection, object segmentation, automatic classification

DOI: 10.15292/geodetski-vestnik.2021.04.559-593

Citation:

Simon Šanca, Krištof Oštir, Alen Mangafić (2021). Zaznavanje stavb z uporabo nevronskih mrež, učenih s prenosom znanja. | Building detection with convolutional networks trained with transfer learning. Geodetski vestnik, 65 (4), 559-593. DOI: 10.15292/geodetski-vestnik.2021.04.559-593