A master thesis from Aalborg University
Mask R-CNN for Segmentation of Aerial Data with Edge Aware Loss
Term
4. term
Education
Publication year
2020
Submitted on
2020-06-03
Pages
8 pages
Abstract
In this project we tackle the problem of image segmentation in aerial data, where the goal is to draw accurate segmentation masks unto buildings. We utilize an existing machine learning model, Mask R-CNN, and attempt to optimize the mask loss function, such as to improve the accuracy of the segmentation masks. We therefore propose and implement two different loss functions for the model, to measure their effects on performance, specifically when applied on our dataset. We also produce a dataset, derived from aerial photographical data and LiDAR data. We find that the Edge Aware loss function results in a noteworthy improvement in output masks.
Documents
Colophon: This page is part of the AAU Student Projects portal, which is run by Aalborg University. Here, you can find and download publicly available bachelor's theses and master's projects from across the university dating from 2008 onwards. Student projects from before 2008 are available in printed form at Aalborg University Library.
If you have any questions about AAU Student Projects or the research registration, dissemination and analysis at Aalborg University, please feel free to contact the VBN team. You can also find more information in the AAU Student Projects FAQs.