Deep learning methods for high-resolution microscale light field image reconstruction: a survey DOI Creative Commons

Bingzhi Lin,

Yuan Tian, Y. Zhang

и другие.

Frontiers in Bioengineering and Biotechnology, Год журнала: 2024, Номер 12

Опубликована: Ноя. 18, 2024

Deep learning is progressively emerging as a vital tool for image reconstruction in light field microscopy. The present review provides comprehensive examination of the latest advancements techniques based on deep algorithms. First, briefly introduced concept and techniques. Following that, application was discussed. Subsequently, we classified learning-based microscopy algorithms into three types contribution learning, including fully method, enhanced raw with numerical inversion volumetric reconstruction, resolution, comprehensively analyzed features each approach. Finally, discussed several challenges, neural approaches increasing accuracy to predict temporal information, methods obtaining training data, strategies data enhancement using existing interpretability networks.

Язык: Английский

Churn prediction with GraphSAGE model based on the derived features using RFM and sentiment analysis DOI

M. Anitha,

K. K. Sherly

Journal of the Chinese Institute of Engineers, Год журнала: 2025, Номер unknown, С. 1 - 13

Опубликована: Март 30, 2025

Язык: Английский

Процитировано

0
Deep learning methods for high-resolution microscale light field image reconstruction: a survey DOI Creative Commons

Bingzhi Lin,

Yuan Tian, Y. Zhang

и другие.

Frontiers in Bioengineering and Biotechnology, Год журнала: 2024, Номер 12

Опубликована: Ноя. 18, 2024

Deep learning is progressively emerging as a vital tool for image reconstruction in light field microscopy. The present review provides comprehensive examination of the latest advancements techniques based on deep algorithms. First, briefly introduced concept and techniques. Following that, application was discussed. Subsequently, we classified learning-based microscopy algorithms into three types contribution learning, including fully method, enhanced raw with numerical inversion volumetric reconstruction, resolution, comprehensively analyzed features each approach. Finally, discussed several challenges, neural approaches increasing accuracy to predict temporal information, methods obtaining training data, strategies data enhancement using existing interpretability networks.

Язык: Английский

Процитировано

0