单边缘阶段检索用于半透明物体测量使用深度卷积生成对抗网络
Jiayan He1, Yuanchang Huang1, Juhao Wu1
1School of Mechanical and Electrical Engineering, Guangzhou University, Guangzhou 510006, China.
Sensors (Basel, Switzerland)
|April 28, 2025
概括
一种名为GAN-PhaseNet的新方法通过改进边缘投影谱 (FPP) 阶段检索来提高半透明物体的3D重建精度. 这种单方法显著减少了分散效应引起的错误.
科学领域:
- 光学和光子学 在光学和光子学.
- 计算机视觉 计算机视觉
- 计量学 计量学 计量学
背景情况:
- 边缘投影特征测量 (FPP) 对于3D重建至关重要,但由于相位移转和边缘调制减少,它与半透明物体作斗争.
- 在半透明材料中的表面散射效应会降低FPP中的测量精度.
研究的目的:
- 引入GAN-PhaseNet,一种新的单相位检索方法,以提高半透明物体的3D测量精度.
- 为了减轻表面散射对FPP包装相位测量的影响.
主要方法:
- 开发了GAN-PhaseNet,这是一个包含U-net++,Resnet101骨干和多层次注意模块的生成对抗网络.
- 采用单框架方法进行相位检索以简化测量过程.
主要成果:
- GAN-PhaseNet实现了卓越的相位检索精度,与非散射和轻微散射物体的传统方法相匹配.
- 与CDLP,Unet-Phase和DCFPP相比,该方法在严重散射的物体中显示出最小的误差.
- GAN-PhaseNet在不同噪声水平和边缘频率上表现出卓越的稳定性和通用性.
结论:
- GAN-PhaseNet有效地提高了边缘投影特征测量中的半透明物体的3D重建精度.
- 拟议的方法为FPP中的计算成像提供了强大而准确的解决方案,特别是对于具有挑战性的半透明材料.
相关概念视频
01:26Phase Contrast and Differential Interference Contrast Microscopy
15.2K
Phase-Contrast Microscopes
In-phase-contrast microscopes, interference between light directly passing through a cell and light refracted by cellular components is used to create high-contrast, high-resolution images without staining. It is the oldest and simplest type of microscope that creates an image by altering the wavelengths of light rays passing through the specimen. Altered wavelength paths are created using an annular stop in the condenser. The annular stop produces a hollow cone of...
In-phase-contrast microscopes, interference between light directly passing through a cell and light refracted by cellular components is used to create high-contrast, high-resolution images without staining. It is the oldest and simplest type of microscope that creates an image by altering the wavelengths of light rays passing through the specimen. Altered wavelength paths are created using an annular stop in the condenser. The annular stop produces a hollow cone of...
15.2K
01:20Deconvolution
698
Deconvolution, also known as inverse filtering, is the process of extracting the impulse response from known input and output signals. This technique is vital in scenarios where the system's characteristics are unknown, and they must be inferred from the observable signals.
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
Deconvolution involves several mathematical techniques to derive the impulse response. One common approach is polynomial division. In this method, the input and output sequences are treated as coefficients of...
698