Automatically extracting buildings from remotely sensed imagery has always been a challenging task, given the spectral homogeneity of buildings with non-building features as well as the complex structural diversity within the image. Traditional machine learning (ML) based methods deeply rely on a huge number of samples and are best suited for medium-resolution images. Unmanned aerial vehicle (UAV) imagery offers the distinct advantage of very high spatial resolution, which is helpful in improving building extraction by characterizing patterns and structures. However, with increased finer details, the number of images also increases many folds in a UAV dataset, which require robust processing algorithms. Deep learning algorithms, specifically Fully Convolutional Networks (FCNs) have greatly improved the results of building extraction from such high resolution remotely sensed imagery, as compared to traditional methods. This study proposes a deep learning-based segmentation approach to extract buildings by transferring the learning of a deep Residual Network (ResNet) to the segmentation-based FCN U-Net. This combined dense architecture of ResNet and U-Net (Res-U-Net) is trained and tested for building extraction on the open source Inria Aerial Image Labelling (IAIL) dataset. This dataset contains 360 orthorectified images with a tile size of 1500 m2 each, at 30 cm spatial resolution with red, green and blue bands; while covering total area of 805 km2 in select US and Austrian cities. Quantitative assessments show that the proposed methodology outperforms the current deep learning-based building extraction methods. When compared with a singular U-Net model for building extraction for the IAIL dataset, the proposed Res-U-Net model improves the overall accuracy from 92.85% to 96.5%, the mean F1-score from 0.83 to 0.88 and the mean IoU metric from 0.71 to 0.80. Results show that such a combination of two deep learning architectures greatly improves the building extraction accuracy as compared to a singular architecture.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Deep Learning-Based Improved Automatic Building Extraction from Open-Source High Resolution Unmanned Aerial Vehicle (UAV) Imagery


    Weitere Titelangaben:

    Lecture Notes in Civil Engineering


    Beteiligte:
    Jain, Kamal (Herausgeber:in) / Mishra, Vishal (Herausgeber:in) / Pradhan, Biswajeet (Herausgeber:in) / Maniyar, Chintan B. (Autor:in) / Kumar, Minakshi (Autor:in)

    Kongress:

    International Conference on Unmanned Aerial System in Geomatics ; 2021 ; Roorkee, India April 02, 2021 - April 04, 2021



    Erscheinungsdatum :

    16.03.2023


    Format / Umfang :

    16 pages





    Medientyp :

    Aufsatz/Kapitel (Buch)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Automatic Road Pavement Distress Recognition Using Deep Learning Networks from Unmanned Aerial Imagery

    Farhad Samadzadegan / Farzaneh Dadrass Javan / Farnaz Ashtari Mahini et al. | DOAJ | 2024

    Freier Zugriff


    Deep Learning based Vehicle Detection in Aerial Imagery

    Sommer, Lars Wilko | TIBKAT | 2022

    Freier Zugriff

    VEHICLE POSITION ESTIMATION WITH AERIAL IMAGERY FROM UNMANNED AERIAL VEHICLES

    Kruber, Friedrich / Morales, Eduardo Sánchez / Chakraborty, Samarjit et al. | British Library Conference Proceedings | 2020


    Deep Learning based Vehicle Detection in Aerial Imagery

    Sommer, Lars Wilko | TIBKAT | 2021

    Freier Zugriff