Session Index

Graphene based materials have a variety of excellent properties and applications due to its remarkable electrical conductivity, surface and thermal properties. In this paper, the imaging approach to estimate the physical properties of graphene-coated glass with respect to the change collected on the three-dimensional (3D) surface is proposed. The holographic system is employed for the optical thickness measurement of relevance in the analysis of 3D surface distribution. Numerical and experimental results are performed to determine the change in conductivity of coated graphene which can be directly measured by the 3D imaging of digital holographic microscopy.

  Preview abstract

A multifocal confocal fluorescence microscopic system is designed and developed by incorporating photopolymer (PQ-PMMA) based volume holographic optical element in the illumination path for generating multifocal super Gaussian (SG) beam spots to reduce the scanning time.

  Preview abstract

Anomaly detection, an important task for industrial manufacturing, is aimed to distinguish the rare defects from mass production. We proposed a semi-supervised learning model with generative adversarial network (GAN), where the network learned the latent representations through normal data only. For the purpose of anomaly determination, the threshold was determined by measuring the distances between normal data and abnormal data in the latent space. With preliminary experiment, 99% true positive rate (TPR) under 5% false positive rate (FPR) was achieved on a commercial cosmetic patch dataset.

  Preview abstract

We demonstrate the geometric phase writing in a nematic liquid crystal cell by using photo-alignment technique with photopolymer of PAZO. The assembled LC cell can convert an incident circularly light beam into the contrary circularly polarized beam with modulated phase related to the direction of alignment caused from PAZO. Thus, with appropriate designed pattern, novel diffractive waveplate, which provides many unique photonic applications can be written in a LC cell directly. In this paper, we show the comprehensive simulation model to characterize geometric phase property of the LC cell and fabrication of a diffractive geometric phase grating using polarization holography.

  Preview abstract

Volume holograms are known to provide high transmission efficiency, single diffracted order, and compact size. These properties have been utilized for a variety of applications. Volume holography with beam shaping can be utilized for high resolution imaging applications for biological samples. Here, we present an Airy light sheet fluorescence microscope (Airy-LSFM) that uses volume hologram as an Airy beam shaper. Our method is easy to implement and it can provide a larger field of view, deep penetration depth, and better optical sectional capability, which helps measure large biological samples. The imaging performance can be further improved by using deconvolution method.

  Preview abstract

A fringe projection profilometer encounters the challenge of noise disturbance due to projection system. In a short depth of field scanning profilometer, a phase shifting method is proposed to increase two dimensional data and to improve the resolution of the scanning fringe projection profilometer.

  Preview abstract

The diffraction efficiency of traditional holographic storage will decrease as the number of multiplexed pages increases. This research team proposes a set of holographic storage technology to improve the diffraction efficiency of single-point multiplexing when the number of multiplexing pages is limited. Thus, the storage density of holographic storage is improved.

  Preview abstract

We combined the rainbow hologram and the depth-added computer-generated hologram to produce a computer-generated rainbow hologram (CGRH), which can display a three-dimensional color image.

  Preview abstract

A phase unwrapping method using the error-controlled fringe projection scheme for phase-shifting projected fringe profilometry is presented. Most unwrapping methods can not cope with the problems caused by low signal-to-noise ratios. The presenting method can detect such kind of errors and recover the mistake.

  Preview abstract

In this study, photomask positioning system by combing the diffraction-based interferometer and the heterodyne phase measurement was proposed. Since this method used heterodyne phase measurement, the positioning errors caused by ununiform film thickness on the wafer can also be avoided.

  Preview abstract

In this study, we propose a deep learning-based segmentation approach for three-dimensional (3D) composite surface with specular and diffuse regions. The proposed U-net based segmentation has been examined for specular and diffuse images from composite objects. The experimental results show that the proposed method can realize absolute 3D surface measurement and automatic segmentation for 3D composite regions based on integrated coherent system.

  Preview abstract

This study proposes a real-time imaging system, which uses an image sensor to capture retrieving light spot images of an absolute optical rotary encoder, converts the images into coding signals, and outputs corresponding coding angles. The research contribution could be applied to real-time verification of signals in the developing stage of absolute optical rotary encoders, so as to skip the complex steps for specific photodetector design and fabrication.

  Preview abstract

The display quality of a binary computer-generated hologram (BCGH) is usually low because of the high loss of information in the binarization. Here we purpose a new method, simulated-annealing binary search (SABS) algorithm to produce high-quality BCGHs.

  Preview abstract

In this research, two different interferometry results were used to calculate the thickness of the optical plate, the first was the spectrally resolved white light interferometry technique, and the second was the laser interferometry technique. The measured data was compared with the theoretical value using the MATLAB program to further obtain the thickness of the optical plate.

  Preview abstract

We proposed and demonstrated a novel method of an incoherent digital holographic camera. The compact system designed for the incoherent digital holographic camera was contributed by a modified common path architecture in the camera imaging system. The three-dimensional human face object would be directly obtained in a single shot without a stable optical table environment.

  Preview abstract

In this study, the optimal brewing water temperature and brewing time for the content
of theaflavin were found by analyzing the absorption spectrum of the black tea soup, the
brewing time of the tea leaves, and the temperature of the brewing water. The results show that
the optimal brewing temperature is 85 °C and the brewing time is 2 minutes, which has the
highest theaflavin content.

  Preview abstract

In this paper, we will present our studies on the fabrication of holographic optical elements using photoresist S1813. The holographic elements are recorded by using two-beam interfering lithographic method on S1813 spin coated glass substrate so that it is possible to achieve high diffraction efficiency in micro-meter thick thin film due to the huge refractive index difference between polymer and air. We will present how to make high quality thin film for performing holographic recording on S1813 spin-coated glass and we will also demonstrate the fabrication of a holographic waveguide for AR applications

  Preview abstract

In recent years, head-mounted displays have received growing attention and lots of applications are presented. To miniaturize the devices, diffractive waveguide devices are the best solutions. We investigated relief grating through the finite element method rather than the traditional rigorous coupled-wave analysis. The depth and the tilt angle of the slits are considered in different incident angles and show the ability to inhibit the −1 order.

  Preview abstract

This paper proposes an optical system for measuring liquid concentration using phase-enhanced total-internal-reflection (TIR) interferometry. In the proposed system, liquid concentration can be determined by measuring the phase difference between the s- and p- polarized states of a beam totally reflected from a phase-enhanced TIR apparatus with a tested solution. The method produced the measurement resolution and sensitivity better than 0.4 mg/dl and 0.08 deg (dl/ mg),

  Preview abstract

Acoustic camera visualize sound using microphone array in the space to acquire sound signal at different positions. With Huygens-Fresnel principle and Kirchhoff integral theorem, the acoustic field distribution in space and time can be reconstructed as well as the source location.

  Preview abstract

Recently, almost all communication systems on the market use radio waves to transmit information, although there are also papers using visible light as the transmission medium. In light transmission, most published papers use laser light as the output, few papers use Micro LEDs as the output, and they only use a small number of light sources to transmit information by adjusting the frequency and amplitude. Therefore, we hope to design a new communication system with the characteristic of Micro LEDs being able to switch between on and off at a fast speed to provide 5G or even 6G communication.

  Preview abstract

In this study, a surface profile measurement based on interferometry as well as polarization interferometry with polarization cameras was introduced. By analyzing the relationship between the density of interference fringes and the angle between the measurement surface and the reference surface, the surface profile measurement under the minimum number of interference fringes can be calculated. The measuring range of the system is 80 mm in diameter.

  Preview abstract

In this study, a solar photovoltaic module recycling and sorting system based on thermal imaging machine vision was designed. The solar photovoltaic modules after thermal cracking were sorted through the combination of thermal imaging and mechanism control. The sorting accuracy for glass and silicon substrates are 90%.

  Preview abstract

Materials exhibiting photo-induced birefringence are interested in many possible applications of optical information processing and storage. The PQ/PMMA, a kind of doped photopolymer consists of polarization-selectively absorbing PQ molecules dispersed in an amorphous PMMA polymer, and thus can undergo anisotropic photodissociation by linearly polarized light, resulting in photoinduced birefringence. In this article, we will discuss the effect of light-controlled photo-birefringence in PQ/PMMA photopolymers and discuss the correlation coefficients of the induced birefringence properties.

  Preview abstract

FaceNet is a face recognition framework proposed by Google in 2015. Rapid advances in deep learning hardware, software, and technology enable the development of more face recognition techniques. The You Only Look Once (YOLO) is a fast deep learning object detection framework. Implementing the YOLO neural network in a face recognition algorithm should significantly improve the recognition speed. For this purpose, the feasibility of replacing multi-task cascaded convolutional networks (MTCNN) with YOLOv5 to detect human faces was explored. The results showed that YOLOv5 was very effective in detecting faces and reducing the recognition time, thus demonstrating a promising prospect.

  Preview abstract

In this paper, we present the feasibility of using convolutional neural networks (CNNs) for information decryption in digital holography. We established a deterministic phase-encoded encryption system based on digital holography. The original image is made into an encrypted hologram and used as ciphertext. The CNN model can be trained by using the ciphertext and its corresponding original image. The trained CNN model is employed to predict the original image. Experimental results show that the trained CNN model can generate plaintext from ciphertext. According to the simulation results of this study, CNN is a feasible method for evaluating information security.

  Preview abstract

YOLO (You Only Look Once) neural network is an object detection-like neural network roadmap algorithm implemented in darknet architecture, which has the advantages of light weight and high efficiency. This research is applied to the identification of the remaining parking spaces in the parking lot and the intelligent car search system. The results show that yolo can identify the position of the vehicle with accuracy of 99.72%.

  Preview abstract