Unreliable Pixel Contrast Based on von Mises-Fisher Distribution for Semi-Supervised SAR Segmentation

The unique visual properties and the huge size of synthetic aperture radar (SAR) images pose challenges in labeling data. The scarcity of labeled data limits the training of SAR image segmentation networks. To address this issue, semi-supervised methods are used to train the network. However, tradit...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE geoscience and remote sensing letters 2024, Vol.21, p.1-5
Hauptverfasser:	Wu, Zitong, Hou, Biao, Li, Zhonghua, Li, Zhehan, Guo, Xianpeng, Ren, Bo
Format:	Artikel
Sprache:	eng
Schlagworte:	Algorithms Buildings Contrastive learning convolutional neural network Distribution Feature extraction Image processing Image segmentation Labels Learning Pixels Radar imaging Radar polarimetry Reliability SAR (radar) Self-supervised learning semi-supervised learning Synthetic aperture radar synthetic aperture radar (SAR) Teachers Training
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Schreiben Sie den ersten Kommentar!