DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Rule-Based, Noisy Labels for Overhead Imagery Segmentation

Albrecht, Conrad M and Liu, Chenying and Wang, Yi and Klein, Levente J and Zhu, Xiao Xiang (2022) Rule-Based, Noisy Labels for Overhead Imagery Segmentation. Living Planet Symposium, 23.-27. May 2022, Bonn, Germany.

[img] PDF


Within the past decade, modern statistical and machine learning methods significantly advanced the field of computer vision. For a significant portion, success stories trace back to training deep artificial neural networks on massive amounts of labeled data. However, generating human labor-intensive annotations for the ever-growing volume of earth observation data at scale renders Sysiphus-like. In the realm of weakly-supervised learning, methods operating on sparse labels attempt to exploit a small set of annotated data in order to train models for inference on the full domain of input. Our work presents a methodology to utilize high resolution geospatial data for semantic segmentation of aerial imagery. Specifically, we exploit high-quality LiDAR measurements to automatically generate a set of labels for urban areas based on rules defined by domain experts. The top of the figure attached provides a visual sample for such automatized classifications in suburbs: vegetation (dark madder purple), roads (lime green), buildings (dark green), and bare land (yellow). A challenge to the approach of auto-generated labels is introduction of noise due to inaccurate label information. Through benchmarks and improved architecture design of the deep artificial neural networks, we provide insights on success and limitations of our approach. Remarkably, we demonstrate that models trained on inaccurate labels have the ability to surpass annotation quality when referenced to ground truth information (cf. bottom of figure attached). Moreover, we investigate boosting of results when weak labels get auto-corrected by domain expert-based noise reduction algorithms. We propose technology interacting with deep neural network architectures that allows human expertise to re-enter weakly supervised learning at scale for semantic segmentation in earth observation. Beyond the presentation of results, our contribution @LPS22 intends to start a vital scientific discussion on how the approach substantiated for LiDAR-based automatic annotation might get extended to other modalities such as hyper-spectral overhead imagery.

Item URL in elib:https://elib.dlr.de/186653/
Document Type:Conference or Workshop Item (Poster)
Title:Rule-Based, Noisy Labels for Overhead Imagery Segmentation
AuthorsInstitution or Email of AuthorsAuthor's ORCID iD
Albrecht, Conrad MConrad.Albrecht (at) dlr.deUNSPECIFIED
Liu, Chenyingchenying.liu (at) dlr.deUNSPECIFIED
Wang, YiYi.Wang (at) dlr.deUNSPECIFIED
Klein, Levente Jkleinl (at) us.ibm.comUNSPECIFIED
Zhu, Xiao Xiangxiao.zhu (at) dlr.deUNSPECIFIED
Date:May 2022
Refereed publication:No
Open Access:Yes
Gold Open Access:No
In ISI Web of Science:No
Keywords:LiDAR, deep learning, semantic segmentation, weak supervision, noisy labels
Event Title:Living Planet Symposium
Event Location:Bonn, Germany
Event Type:international Conference
Event Dates:23.-27. May 2022
Organizer:European Space Agency
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Earth Observation
DLR - Research area:Raumfahrt
DLR - Program:R EO - Earth Observation
DLR - Research theme (Project):R - Artificial Intelligence
Location: Oberpfaffenhofen
Institutes and Institutions:Remote Sensing Technology Institute > EO Data Science
Deposited By: Albrecht, Conrad M
Deposited On:03 Jun 2022 10:52
Last Modified:07 Jun 2022 18:05

Repository Staff Only: item control page

Help & Contact
electronic library is running on EPrints 3.3.12
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.