Adversarial convolutional neural network model, OCT image reconstruction algorithm and storage medium

The invention discloses an adversarial convolutional neural network model, an OCT image reconstruction algorithm and a storage medium, the adversarial convolutional neural network model comprises a generator and a discriminator, the generator comprises a frequency domain image feature extraction mod...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: XIA WEI, LIU RUQIAN, LIU RUI, SUN JIAKUI, FAN YONGZHEN, DOU SHUWEI, HAN TINGTING
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention discloses an adversarial convolutional neural network model, an OCT image reconstruction algorithm and a storage medium, the adversarial convolutional neural network model comprises a generator and a discriminator, the generator comprises a frequency domain image feature extraction module and an up-sampling super-resolution reconstruction module, the frequency domain image feature extraction module converts an input spatial domain image into a complex frequency domain image, and the up-sampling super-resolution reconstruction module performs up-sampling super-resolution reconstruction on the complex frequency domain image; the frequency domain image feature extraction module is used for extracting the frequency domain image feature, low-resolution image frequency spectrum feature extraction is achieved through frequency spectrum translation so as to extract important information of an image frequency domain, the up-sampling super-resolution reconstruction module carries out up-sampling on an ima