Dual-wavelength digital holography has advantages over single-wavelength digital holography in resolving phase discontinuities at high aspect-ratio. However, the operations are very time-consuming and cannot achieve real-time processing. We realized the phase reconstruction of dual-wavelength off-axis holograms on Java platform, and used GPU to accelerate the computation-intensive part. Preliminary experiments show we can reconstruct 1 mega pixel holograms continuously at a speed of 41 fps, which can satisfy the stable video-rate. Through Java, the system can be easily combined with numerous plugins of ImageJ, such as filters, LUT for pseudo-color, 3D tools, etc. This is of great help to the subsequent image analysis and processing.
Off-axis dual-wavelength digital holography (oaDWDH) can enable quantitative phase imaging on thickness samples without numerical phase unwrapping in a single shot. However, the traditional oaDWDH is huge and unstable owing to its separated-path geometries. In this paper, we presented a compact oaDWDH using wavefront-splitting in the quasi common-path. In our approach, a dual-wavelength spherical wave is split into two parts to act as the reference wave and the object wave, respectively. Only a few such optical elements as a mirror and a beam splitter are employed to adjust and recombine the two waves, and a hologram containing two-wavelength information is then captured by a monochromatic CCD camera. The information of a specimen, including phase and height, can be reconstructed through a division algorithm with the help of a specimen-free multiplexed interferogram. In order to verify the feasibility of the system, observations were performed on the step samples. The height of the sample is obtained quantitatively, and finally compared with the measured height result of the step sample by AFM to prove the accuracy of the measurement result.
An autofocusing method using correlation coefficient (CC) is proposed for dual-wavelength off-axis digital holography. The complex amplitudes of the object wave relative to the two wavelengths are first retrieved at different reconstruction distances, and the correlation degrees are then calculated between the two complex amplitudes. Considered the diffraction independency between the two wavelengths, the maximum CC is employed to automatically determine the focus plane. Our method can be applicable for the amplitude, phase, or both mixed sample. The experimental results demonstrate that the proposed method can enable automatically autofocusing with higher resolution in contrast to the state-of-the-art method.
The existing infrared thermal imaging detection methods usually process the whole video stream data collected by a thermal camera, which involve large amounts of data and have a negative effect on the efficiency of defect detection. In this paper, we propose an infrared thermal imaging detection method which considers the spatial correlation of the adjacent images in the video stream data. By extracting the edge information and analyzing the correlation between two adjacent frames, the defect area and the non-defect area show different correlation coefficients, and only part of the video data is required for defect detection. Furthermore, the fusion method is introduced to enhance the image quality. The experiment results demonstrate that the proposed method can not only reflect the change of heat in the defect area during the heating process but also reduce computation time involved in the subsequent processing.
We build a two-wavelength off-axis quasi-common-path digital holography for quantitative phase imaging (QPI) using polarization-multiplexing and flipping. The interference is performed by flipping the relative position of a sample and reference beam, and the dual-wavelength information is spatially multiplexed onto a monochromatic CCD camera simultaneously using polarization-multiplexing. Due to orthogonal interference fringes of two-wavelengths, the unwrapped information on the phase and thickness for the sample can be extracted from a single interferogram. Our setup requires no pinholes, gratings or dichroic mirror with straightforward alignment. Additionally, a division algorithm for dual-wavelength off-axis digital holography with the help of a specimen-free multiplexed interferogram is proposed to extract the phase of a specimen. We demonstrate the operation of the setup with step target and circular pillar.
An extrinsic Fabry-Perot interferometric fibre microphone based on polydimethylsiloxane, or PDMS, diaphragm was proposed. The large free-standing PDMS diaphragm, with a diameter of 4 mm, is prepared by a simple “spin-strip” process. The experimental result shows that the fabricated sensor has a high dynamic pressure sensitivity of about -136 dB re 1 rad/μPa in the range of 100~2000Hz. The noise equivalent acoustic signal level of the microphone, limited by the environmental noise, is about 1000 μPa/Hz1/2. And the dynamic range is tested to be more than 47.48 dB. The proposed microphone is expected to be used in the field of weak acoustic pressure testing.
We propose a security-enhanced optical interference-based multiple-image encryption (IBMIE) using a modified multiplane phase retrieval algorithm (MPPRA) in the Fresnel transform domain. In this IBMIE scheme, while a phase-only mask (POM) distributing randomly between [ 0 , 2π ] is fixed, the other POM is iteratively extracted based on a modified mMPPRA, and thus, multiple plaintexts are simultaneously encrypted into two POMs with different distances to the image planes. At last, the retrieved POM is pixel scrambled by chaotic pixel scrambling (CPS). During image decryption, the decrypted images can be obtained at their preset positions by an intensity detector directly. The silhouette of the original images cannot be obtained using either of the two POMs. The parameters of both CPS and Fresnel transform can serve as security keys to enhance the security. Numerical simulation is presented to demonstrate the validity of the proposed mMPPRA-based IBMIE.