一种基于深度光学的轻量化双目立体视觉相机的成像方法和装置

By combining optimized optical elements and neural networks, a depth-optical binocular stereo vision camera was developed, which solved the problems of lightweight design and insufficient accuracy in depth estimation of stereo vision cameras, and achieved fast and accurate depth estimation and full-focus image recovery.

CN115767065BActive Publication Date: 2026-07-17YUGUANG LANSHU (HANGZHOU) INTELLIGENT TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
YUGUANG LANSHU (HANGZHOU) INTELLIGENT TECH CO LTD
Filing Date
2022-11-15
Publication Date
2026-07-17

Smart Images

  • Figure CN115767065B_ABST
    Figure CN115767065B_ABST
Patent Text Reader

Abstract

本发明公开了一种基于深度光学的轻量化双目立体视觉相机的成像方法和成像装置，包括：构建物理成像模型，并基于物理成像模型编码得到预模拟左右视图；基于神经网络构建用于预测深度图和全焦图的生成模型，以预模拟左右视图作为生成模型的输入，对生成模型的参数和物理成像模型中左右光学元件的面型高度参数进行联合优化；按照优化的面型高度参数制造左右光学元件，并采用实拍方式获得左右光学元件的实际点扩散函数和编码得到再模拟左右视图；利用再模拟左右视图对生成模型再优化；采集实拍左右视图，并利用再优化后的生成模型预测高质量全焦图和深度图的预测。该成像方法和装置在不同场景下均能实现快速而准确的深度估计，提供可靠的距离信息。
Need to check novelty before this filing date? Find Prior Art