Search

We propose a deep learning computational ghost imaging (CGI) scheme to achieve sub-Nyquist and high-quality image reconstruction. Unlike the second-order-correlation CGI and compressive-sensing CGI, which use lots of illumination patterns and a one-dimensional (1-D) light intensity sequence (LIS) for image reconstruction, a deep neural network (DAttNet) is proposed to restore the target image only using the 1-D LIS. The DAttNet is trained with simulation data and retrieves the target image from experimental data. The experimental results indicate that the proposed scheme can provide high-quality images with a sub-Nyquist sampling ratio and performs better than the conventional and compressive-sensing CGI methods in sub-Nyquist sampling ratio conditions (e.g., 5.45%). The proposed scheme has potential practical applications in underwater, real-time and dynamic CGI.

The interferometry technique is commonly used to obtain the phase information of an object in optical metrology. The obtained wrapped phase is subject to a 2π ambiguity. To remove the ambiguity and obtain the correct phase, phase unwrapping is essential. Conventional phase unwrapping approaches are time-consuming and noise sensitive. To address those issues, we propose a new approach, where we transfer the task of phase unwrapping into a multi-class classification problem and introduce an efficient segmentation network to identify classes. Moreover, a noise-to-noise denoised network is integrated to preprocess noisy wrapped phase. We have demonstrated the proposed method with simulated data and in a real interferometric system.

A snapshot multi-wavelength interference microscope is proposed for high-speed measurement of large vertical range discontinuous microstructures and surface roughness. A polarization CMOS camera with a linear micro-polarizer array and Bayer filter accomplishes snapshot multi-wavelength phase-shifting measurement. Four interferograms with 𝜋/2 phase shift are captured at each wavelength for phase measurement, the 2𝜋 ambiguities are removed by using two or three wavelengths.

To reduce the environmental errors, a snapshot phase-shifting interference microscope (SPSIM) has been developed for surface roughness measurement. However, fringe-print-through (FPT) error widely exists in the phase-shifting interferometry (PSI). To ensure the measurement accuracy, we analyze the sources which introduce the FPT error in the SPSIM. We also develop a FPT error correction algorithm which can be used in the different intensity distribution conditions. The simulation and experiment verify the correctness and feasibility of the FPT error correction algorithm.

To achieve high measurement accuracy with less computational time-in-phase shifting interferometry, a random phase-shifting algorithm based on principal component analysis and Lissajous ellipse fitting (PCA&LEF) is proposed. It doesn't need pre-filtering and can obtain relatively accurate phase distribution with only two phase shifted interferograms and less computational time and is suitable for different background intensity, modulation amplitude distributions and noises. Moreover, it can obtain absolutely accurate result when the background intensity and modulation amplitude are perfect and can partly suppress the effect of imperfect background intensity and modulation amplitude. Last but not least, it removes the restriction that PCA needs more than three interferograms with well-distributed phase shifts to subtract relatively accurate mean. The simulations and experiments verify the correctness and feasibility of PCA&LEF.

To overcome the phase shift error in phase shifting interferometry, a two-step random phase retrieval approach based on Gram-Schmidt (GS) orthonormalization and Lissajous ellipse fitting (LEF) method (GS&LEF) is proposed. It doesn’t need pre-filtering, and can obtain relatively accurate phase distribution with only two phase shifted interferograms and less computational time. It is suitable for different background intensity, modulation amplitude distributions and noises. Last but not least, this method is effective for circular, straight or complex fringes. The simulations and experiments verify the correctness and feasibility of GS&LEF.

To overcome the phase shift error in phase shifting interferometry, a three-step random phase retrieval approach based on difference map normalization and diamond diagonal vector normalization (DN&DDVN) is proposed. It does not need pre-filtering for the interferograms and can obtain relatively accurate phase distribution with a simple process and less computational time. This simulation and experiment verify the correctness and feasibility of DN&DDVN.

To accurately obtain the phase distribution of an optical surface under test, the accurate phase extraction algorithm is essential. To overcome the phase shift error, a random two-step phase shifting algorithm, which can be used in the fluctuating and non-uniform background intensity and modulation amplitude, Lissajous ellipse fitting, and least squares iterative phase shifting algorithm (LEFLSI PSA), is proposed; pre-filtering interferograms are not necessary, but they can get relatively accurate phase distribution and unknown phase shift value. The simulation and experiment verify the correctness and feasibility of the LEFLSI PSA.

We correct two figures showing the micro-polarizer arrangement on the RGB full-Stokes camera.