Full-focus imaging method and system based on event camera

The invention discloses a full-focusing imaging method and system based on an event camera, the method comprises the steps of shooting an event focus stack, searching focusing time based on a golden proportion search algorithm, reconstructing a refocusing image and fusing and reconstructing the focu...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: TENG MINGGUI, LOU HANYUE, SHI BAIXIN, YANG YIXIN
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 a full-focusing imaging method and system based on an event camera, the method comprises the steps of shooting an event focus stack, searching focusing time based on a golden proportion search algorithm, reconstructing a refocusing image and fusing and reconstructing the focus stack, and the system comprises a hybrid camera and corresponding modules in a neural network. According to the method, scene information is more clearly and comprehensively recorded through a focus stack shooting process based on an event camera, meanwhile, a deep learning method is used, an out-of-focus picture and an event focus stack are effectively fused, picture refocusing is achieved, a clear full-focus image is generated, and the problems that in the prior art, the quality of a reconstruction result is unstable and unreliable are solved. 本发明公开了一种基于事件相机的全聚焦成像方法及系统,方法步骤包括拍摄事件焦点堆栈、基于黄金比例搜索算法搜索聚焦时间、重建重聚焦图像和融合重建出焦点堆栈,系统包括混合相机及神经网络中的相应模块。本发明通过基于事件相机的焦点堆栈拍摄过程,更清晰、全面地记录场景信息,同时使用深度学习的方法,有效融合失焦图片与事件焦点堆栈,实现画面重聚焦,产生清