Change detection is an important topic in remote sensing to study the effects of climate change, natural disasters, urbanization, etc. However, the need for labeled data has posed significant challenges. In this paper, we introduce a self-supervised learning model to overcome this problem. To evaluate our model performance, we propose a novel evaluation metric called recall-based operational reliability. In our study, we used a large-scale multispectral image dataset called DynamicEarthNet for testing.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Large Scale Multispectral Image Dataset Change Detection Based on Self-Supervised Learning with Novel Evaluation Metric


    Contributors:


    Publication date :

    2024-07-15


    Size :

    2220121 byte





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Learning a nonlinear distance metric for supervised region-merging image segmentation

    Sobieranski, A. C. / Comunello, E. / von Wangenheim, A. | British Library Online Contents | 2011


    Shared median-scaling metric for multi-camera self-supervised depth evaluation

    GUIZILINI VITOR / AMBRUS RARES ANDREI / GAIDON ADRIEN DAVID et al. | European Patent Office | 2023

    Free access

    A New Pedestrian Dataset for Supervised Learning

    Overett, G.M. / Petersson, L. / Brewer, N. et al. | British Library Conference Proceedings | 2008


    Scenario-based threat metric evaluation based on the highd dataset

    Schneider, Patrick / Butz, Martin / Heinzemann, Christian et al. | IEEE | 2020


    A new pedestrian dataset for supervised learning

    Overett, Gary / Petersson, Lars / Brewer, Nathan et al. | IEEE | 2008