Pavel A. Cheremkhin

Digital holography is a powerful imaging technique for quantitative analysis in various fields of science and technology, including biomedical applications. However, the development and validation of neural network-based methods in this field are often limited by the unavailability of large, well-annotated experimental data. For example, in biological studies, the typical size of an experimental dataset is around 1000 images. In this work, we present a procedural framework for generating synthetic datasets of biological cell phase images and digital holograms. The generated datasets were tested in several key tasks of neural network applications, including cell classification, detection, and phase reconstruction from in-line holograms. High classification accuracy is achieved for both phase images and holograms, while phase reconstruction reaches high structural similarity indices. The applicability of the generated datasets is further demonstrated in complex scenarios involving multiple moving cells and multiple object planes. The proposed approach provides a flexible and scalable platform for method development, performance evaluation, and future integration with experimental holographic imaging systems.

Digital cameras are widely used in a variety of applications, including medicine, security, material characterization, and metrology. To reduce cost, increase dynamic range, and minimize noise, accurate knowledge of camera parameters, including noise characteristics, is required. The most promising camera noise estimation methods use a single image. However, most methods estimate the noise of the acquired image rather than the intrinsic sensor noise of the camera. In this paper, a method is proposed to estimate main noise components from a single image: light temporal, light spatial and dark noise The method is based on segmentation, multi-interval approximation, and weighted averaging. The method was experimentally validated on a dataset obtained from a camera whose noise parameters were independently measured. The noise estimation error was reduced to single-digit or several tens of percent, which is more than 5 times better than using patch-based approximation. The proposed method can be used in signal and image processing, for camera selection, noise reduction, image authenticity, and so on.

This article addresses a pressing issue in microwave photonics: the development and measurement of the characteristics of ultra-high-frequency optical modulation systems based on Mach-Zehnder modulators. Using ultrahigh-resolution optical spectrometers to measure the spectral composition of a modulator during operation allows us to identify features that cannot be measured after photodetection. Three types of control signal deviations are investigated: operating point setting accuracy, attenuation inequality in the modulator arms, and inequality in the absolute values of signal voltages on different arms of the modulator. The presented spectral compositions of the optical and radio-frequency spectra, obtained through numerical simulation and experiments on modulating radiation with a sinusoidal signal in different operating modes of a standard Mach-Zehnder modulator, allow us to evaluate the infl uence of these effects. It is shown that for deviations less than 1 %, the greatest contribution to spectral distortion is due to the accuracy of the operating point setting. Experimental evaluation methods are proposed for measuring deviation. The obtained results and conclusions provide new opportunities for assessing the manufacturing quality of standard modulators, as well as the accuracy of operating point control systems.

High-speed realization of computer-generated holograms (CGHs) is a crucial problem in the field of modern 3D visualization and optical image processing system development. Binary CGHs can be realized using high-resolution, high-speed spatial light modulators such as ferroelectric liquid crystals on silicon devices or digital micro-mirror devices providing the high throughput of optoelectronic systems. However, the quality of holographic images restored by binary CGHs often suffers from distortions, background noise, and speckle noise caused by the limitations and imperfections of optical system components. The present manuscript introduces a method based on the optimization of CGH models directly in the optical system with a camera-in-the-loop configuration using effective direct search with a random trajectory algorithm. The method was experimentally verified. The results demonstrate a significant enhancement in the quality of the holographic images optically restored by binary-phase CGH models optimized through this method compared to purely digitally generated models.

Neural networks are a state-of-the-art technology for fast and accurate holographic image reconstruction. However, at present, neural network-based reconstruction methods are predominantly applied to objects with simple, homogeneous spatial structures: blood cells, bacteria, microparticles in solutions, etc. However, in the case of objects with high contrast details, the reconstruction needs to be as precise as possible to successfully extract details and parameters. In this paper we investigate the use of neural networks in holographic reconstruction of spatially inhomogeneous binary data containers (QR codes). Two modified lightweight convolutional neural networks (which we named HoloLightNet and HoloLightNet-Mini) with an encoder–decoder architecture have been used for image reconstruction. These neural networks enable high-quality reconstruction, guaranteeing the successful decoding of QR codes (both in demonstrated numerical and optical experiments). In addition, they perform reconstruction two orders of magnitude faster than more traditional architectures. In optical experiments with a liquid crystal spatial light modulator, the obtained bit error rate was equal to only 1.2%. These methods can be used for practical applications such as high-density data transmission in coherent systems, development of reliable digital information storage and memory techniques, secure optical information encryption and retrieval, and real-time precise reconstruction of complex objects.

This review explores the current landscape of neural network-based methods for digital image noise processing. Digital cameras have become ubiquitous in fields like forensics and medical diagnostics, and image noise remains a critical factor for ensuring image quality. Traditional noise suppression techniques are often limited by extensive parameter selection and inefficient handling of complex data. In contrast, neural networks, particularly convolutional neural networks, autoencoders, and generative adversarial networks, have shown significant promise for noise estimation, suppression, and analysis. These networks can handle complex noise patterns, leverage context-specific data, and adapt to evolving conditions with minimal manual intervention. This paper describes the basics of camera and image noise components and existing techniques for their evaluation. Main neural network-based methods for noise estimation are briefly presented. This paper discusses neural network application for noise suppression, classification, image source identification, and the extraction of unique camera fingerprints through photo response non-uniformity. Additionally, it highlights the challenges of generating reliable training datasets and separating image noise from photosensor noise, which remains a fundamental issue.

Digital holography enables the reconstruction of both 2D and 3D object information from interference patterns captured by digital cameras. A major challenge in this field is speckle noise, which significantly degrades the quality of the reconstructed images. We propose a novel speckle noise reduction method based on 3D adaptive filtering. Our technique processes a stack of holograms, each with an uncorrelated speckle pattern, using an adapted 3D Frost filter. Unlike conventional filtering techniques, our approach exploits statistical adaptivity to enhance noise suppression while preserving fine image details in the reconstructed holograms. Both numerical simulations and optical experiments confirm that our 3D filtering technique significantly enhances reconstruction quality. Specifically, it reduces the normalized standard deviation by up to 40% and improves the structural similarity index by up to 60% compared to classical 2D, 3D median, BM3D, and BM4D filters. Optical experiments validate the method’s effectiveness in practical digital holography scenarios by local and global image quality estimation metrics. These results highlight adaptive 3D filtering as a promising approach for mitigating speckle noise while maintaining structural integrity in digital holography reconstructions.

При передаче аналоговых сигналов по оптическим трактам качество сигнала ухудшается вследствие потерь, нелинейных искажений и шумов, что приводит к уменьшению эффективного числа бит в переданном сигнале, подвергнутом аналого-цифровому преобразованию. В настоящей статье представлены результаты экспериментального исследования функционирования сверхширокополосного сверхвысокочастотного аналогового оптического тракта с цифровым выходом. Целью работы является оценка качества передачи сигнала исследуемым трактом с точки зрения цифровых метрик. Выполнено сравнение результатов оцифровки сигнала генератора и этого же сигнала, переданного аналоговым оптическим трактом. При помощи анализа радиочастотных спектров оцифрованных сигналов оценен вклад оптической части системы в изменение качества сигнала. Измерением получено, что уровень нелинейных искажений и шума, вносимый в сигнал оптическим трактом, составил 14.7 дБ, что определяет уменьшение эффективного числа бит в оцифрованном сигнале тракта на 2.44.