Park Seonghwan, Kim Youhyun, Moon Inkyu
Department of Robotics Engineering, DGIST, 333 Techno Jungang-daero, Hyeonpung-eup, Dalseong-gun, Daegu, 42988, Republic of Korea.
Biomed Opt Express. 2021 Oct 22;12(11):7064-7081. doi: 10.1364/BOE.440338. eCollection 2021 Nov 1.
Digital holography can provide quantitative phase images related to the morphology and content of biological samples. After the numerical image reconstruction, the phase values are limited between -π and π; thus, discontinuity may occur due to the modulo 2π operation. We propose a new deep learning model that can automatically reconstruct unwrapped focused-phase images by combining digital holography and a Pix2Pix generative adversarial network (GAN) for image-to-image translation. Compared with numerical phase unwrapping methods, the proposed GAN model overcomes the difficulty of accurate phase unwrapping due to abrupt phase changes and can perform phase unwrapping at a twice faster rate. We show that the proposed model can generalize well to different types of cell images and has high performance compared to recent U-net models. The proposed method can be useful in observing the morphology and movement of biological cells in real-time applications.
数字全息术可以提供与生物样本的形态和内容相关的定量相位图像。在进行数值图像重建后，相位值被限制在-π和π之间；因此，由于模2π运算可能会出现不连续性。我们提出了一种新的深度学习模型，该模型通过将数字全息术与用于图像到图像转换的Pix2Pix生成对抗网络（GAN）相结合，能够自动重建展开的聚焦相位图像。与数值相位展开方法相比，所提出的GAN模型克服了由于相位突然变化而导致的精确相位展开的困难，并且可以以快两倍的速度进行相位展开。我们表明，所提出的模型可以很好地推广到不同类型的细胞图像，并且与最近的U-net模型相比具有高性能。所提出的方法在实时应用中观察生物细胞的形态和运动方面可能会很有用。
