DETEKCIJA POLJA SOLARNIH ELEKTRANA SA SATELITSKIH SNIMAKA
DOI:
https://doi.org/10.24867/27BE11JakovljevicKljučne reči:
Semantička segmentacija, U-mreža, solarni paneliApstrakt
Ovaj istraživački rad predstavlja automatizovan sistem napravljen korišćenjem dubokog učenja, tačnije konvolucione neuronske mreže (CNN) sa U-Net arhitekturom, za detekciju i precizno lociranje solarnih elektrana na satelitskim snimcima. Sistem postiže impresivno preklapanje od 89% prema metrici preseka i unije (IoU), olakšavajući praćenje obnovljivih izvora energije i doprinoseći održivosti.
Reference
[1] O. Ronneberger, P. Fischer i T. Brox, U-Net: Convolutional Networks for Biomedical Image Segmentation, Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2015.
[2] U. Ramer, An iterative procedure for the polygonal approximation of plane curves, Computer Graphics and Image Processing, 1972.
[3] D. Douglas i P. Peucker, Algorithms for the Reduction of the Number of Points Required to Represent a Digitized Line or its Caricature, Cartographica: The International Journal for Geographic Information and Geovisualization, 1973.
[2] U. Ramer, An iterative procedure for the polygonal approximation of plane curves, Computer Graphics and Image Processing, 1972.
[3] D. Douglas i P. Peucker, Algorithms for the Reduction of the Number of Points Required to Represent a Digitized Line or its Caricature, Cartographica: The International Journal for Geographic Information and Geovisualization, 1973.
##submission.downloads##
Objavljeno
2024-06-05
Broj časopisa
Rubrika
Elektrotehničko i računarsko inženjerstvo
