Multimode optical fiber transmission image migration reconstruction algorithm based on numerical speckles

The invention relates to a multimode optical fiber transmission image migration reconstruction algorithm based on numerical speckles. The method comprises the following steps: firstly, obtaining an experimental speckle image data set; then generating a simulation speckle image data set by using a nu...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: SONG BINBIN, MA XIUHAO, CHEN SHENGYONG, CHEN YONGCHENG, JIN CHANG
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 relates to a multimode optical fiber transmission image migration reconstruction algorithm based on numerical speckles. The method comprises the following steps: firstly, obtaining an experimental speckle image data set; then generating a simulation speckle image data set by using a numerical simulation method; analyzing the characteristics of the experimental speckles and the simulated speckles; designing a loss function fusing speckle characteristics, and the transfer learning ability of the U_Net network is enhanced; inputting the simulation speckles into the U_Net network, and training the simulation speckles into a simulation network model; and reconstructing the original image from the experimental speckle image by directly using the simulation network model subsequently. 本发明涉及一种基于数值散斑的多模光纤传输图像迁移重构算法,该方法首先获取到实验散斑图像数据集;然后利用数值仿真方法生成仿真散斑图像数据集;接下来分析出实验散斑与仿真散斑的特性;设计融合了散斑特性的损失函数,增强U_Net网络的迁移学习能力;然后将仿真散斑输入到U_Net网络中,训练成仿真网络模型;便于后续直接利用仿真网络模型从实验散斑图像中重构出原始图像。