Procedural Synthesis of Remote Sensing Images for Robust Change Detection with Neural Networks

Maria Kolos, Anton Marin, Alexey Artemov, Evgeny Burnaev

    Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

    4 Citations (Scopus)

    Abstract

    Data-driven methods such as convolutional neural networks (CNNs) are known to deliver state-of-the-art performance on image recognition tasks when the training data are abundant. However, in some instances, such as change detection in remote sensing images, annotated data cannot be obtained in sufficient quantities. In this work, we propose a simple and efficient method for creating realistic targeted synthetic datasets in the remote sensing domain, leveraging the opportunities offered by game development engines. We provide a description of the pipeline for procedural geometry generation and rendering as well as an evaluation of the efficiency of produced datasets in a change detection scenario. Our evaluations demonstrate that our pipeline helps to improve the performance and convergence of deep learning models when the amount of real-world data is severely limited.

    Original languageEnglish
    Title of host publicationAdvances in Neural Networks – ISNN 2019 - 16th International Symposium on Neural Networks, ISNN 2019, Proceedings
    EditorsHuchuan Lu, Huajin Tang, Zhanshan Wang
    PublisherSpringer Verlag
    Pages371-387
    Number of pages17
    ISBN (Print)9783030228071
    DOIs
    Publication statusPublished - 2019
    Event16th International Symposium on Neural Networks, ISNN 2019 - Moscow, Russian Federation
    Duration: 10 Jul 201912 Jul 2019

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume11555 LNCS
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Conference

    Conference16th International Symposium on Neural Networks, ISNN 2019
    Country/TerritoryRussian Federation
    CityMoscow
    Period10/07/1912/07/19

    Keywords

    • Deep learning
    • Remote sensing
    • Synthetic imagery

    Fingerprint

    Dive into the research topics of 'Procedural Synthesis of Remote Sensing Images for Robust Change Detection with Neural Networks'. Together they form a unique fingerprint.

    Cite this