Autors: Hensel S., Marinov, M. B., Koch M., Arnaudov, D. D.
Title: Evaluation of Deep Learning-Based Neural Network Methods for Cloud Detection and Segmentation
Keywords: machine learning, generation, ground-based sky image, irradi

Abstract: This paper presents a systematic approach for accurate short-time cloud coverage prediction based on a machine learning (ML) approach. Based on a newly built omnidirectional ground-based sky camera system, local training and evaluation data sets were created. These were used to train several state-of-the-art deep neural networks for object detection and segmentation. For this purpose, the camera-generated a full hemispherical image every 30 min over two months in daylight conditions with a fish-eye lens. From this data set, a subset of images was selected for training and evaluation according to various criteria. Deep neural networks, based on the two-stage R-CNN architecture, were trained and compared with a U-net segmentation approach implemented by CloudSegNet. All chosen deep networks were then evaluated and compared according to the local situation.



    Energies, vol. 14, issue 19, pp. 1-14, 2021, Switzerland, MDPI,

    Copyright MDPI

    Цитирания (Citation/s):
    1. K. Kroics and J. Zarembo, "Concept of Inductor with a Virtual Air Gap for Increasing Fault Current Capability in Traction Drive Applications," 2022 13th National Conference with International Participation (ELECTRONICA), Sofia, Bulgaria, 2022, pp. 1-4, doi: 10.1109/ELECTRONICA55578.2022.9874364. - 2022 - в издания, индексирани в Scopus или Web of Science
    2. Nedyalkov, I., Application of GNS3 to Study the Security of Data Exchange between Power Electronic Devices and Control Center, Computers 12(5),101, Open Access, DOI 10.3390/computers12050101 - 2023 - в издания, индексирани в Scopus или Web of Science
    3. Liu, Y., Jing, Z., Power System Relay Protection Based on Faster R-CNN Algorithm, International Journal of Information Technology and Web Engineering 18(1), Open Access, DOI 10.4018/IJITWE.333475 - 2023 - в издания, индексирани в Scopus или Web of Science

    Вид: статия в списание, публикация в издание с импакт фактор, публикация в реферирано издание, индексирана в Scopus