Papers by Anna Pastuszczak
Optical single pixel detection with sampling functions utilizing prior knowledge
Single pixel detectors find applications ranging from multispectral imaging, through polarimetric... more Single pixel detectors find applications ranging from multispectral imaging, through polarimetric, 3d, and holographic imaging, up to optical encryption and imaging through scattering media, which are however all hindered by the high measurement and reconstruction times. In this paper we propose to reduce the required signal acquisition time by using non-ergodic stationary sampling correlated with the measurement. The proposed sampling scheme after binarization can be used with binary spatial light modulators such as the digital micro-mirror devices. We report respective optical results obtained with a single-pixel detector with a state-of-the-art 21 kHz DMD. The proposed sampling method is based on a random selection of Morlet wavelet functions convolved with white noise. Our choice of sampling functions provides a trade-off between two contradictory objectives. One is the use sampling functions incoherent with the basis in which the image has a sparse representation, as is required in the compressive sensing framework. This is an argument for using random sampling. The other is to use sampling functions with the highest possible correlation with the image, which give a fast learning rate when some a priori information on the image is available. Here, wavelets are clearly preferable with respect to uncorrelated random patterns, and what we propose is a a trade-off between these two approaches.
Imaging in the visible wavelength range through anisotropic layered flat lens operating in the canalization regime
Proceedings of SPIE, Apr 30, 2010
We study the propagation of light through silver-dielectric layered structures operating in the c... more We study the propagation of light through silver-dielectric layered structures operating in the canalization regime. These structures have an extremely large value of the effective permittivity in the propagation direction. Therefore they are able to couple a broad spectrum of incident spatial frequencies, including evanescent waves, into propagating modes. As a result, subwavelength resolution at the back interface of the
Sub-wavelength imaging using silver-dielectric metamaterial layered prism
Proceedings of SPIE, Sep 24, 2010
ABSTRACT Export Date: 6 April 2013, Source: Scopus, Art. No.: 774613
Obrazowanie nadrozdzielcze za pomocą nanoukładów plazmonicznych
Obrazowanie nadrozdzielcze za pomocą nanoukładów plazmonicznych
Optics Express
The usually reported pixel resolution of single pixel imaging (SPI) varies between 32 × 32 and 25... more The usually reported pixel resolution of single pixel imaging (SPI) varies between 32 × 32 and 256 × 256 pixels falling far below imaging standards with classical methods. Low resolution results from the trade-off between the acceptable compression ratio, the limited DMD modulation frequency, and reasonable reconstruction time, and has not improved significantly during the decade of intensive research on SPI. In this paper we show that image measurement at the full resolution of the DMD, which lasts only a fraction of a second, is possible for sparse images or in a situation when the field of view is limited but is a priori unknown. We propose the sampling and reconstruction strategies that enable us to reconstruct sparse images at the resolution of 1024 × 768 within the time of 0.3s. Non-sparse images are reconstructed with less details. The compression ratio is on the order of 0.4% which corresponds to an acquisition frequency of 7Hz. Sampling is differential, binary, and non-adap...
Visualization4.avi
Real-time differential single-pixel imaging with Fourier domain regularization at the resolution ... more Real-time differential single-pixel imaging with Fourier domain regularization at the resolution of 256x256, at 5.8fps at the compression ratio of 6%. The object contains a robotic toy constructed with Lego bricks moving at a varying speed.
Visualization5.avi
Real-time differential single-pixel imaging with Fourier domain regularization at the resolution ... more Real-time differential single-pixel imaging with Fourier domain regularization at the resolution of 256x256, at 11.5fps at the compression ratio of 3%. The object contains a robotic toy constructed with Lego bricks moving at a varying speed.
Visualization2.avi
Real-time differential single-pixel imaging with Fourier domain regularization. Polarization imag... more Real-time differential single-pixel imaging with Fourier domain regularization. Polarization imaging at the resolution of 256x256, at 5.5fps at the compression ratio of 6%. Orthogonal linear polarizations are shown in pseudocolor (in red and in blue). The scene contains a birefirngent object (a ruler) moved in front of a linear polarizer.
Visualization6.avi
Real-time differential single-pixel imaging with Fourier domain regularization at the resolution ... more Real-time differential single-pixel imaging with Fourier domain regularization at the resolution of 256x256, at 17.2fps at the compression ratio of 2%. The object contains a robotic toy constructed with Lego bricks moving at a varying speed.
Visualization3.avi
Real-time differential single-pixel imaging with Fourier domain regularization. Combined visible ... more Real-time differential single-pixel imaging with Fourier domain regularization. Combined visible and near infrared imaging at the resolution of 256x256, at 5.5fps at the compression ratio of 6%. The object consists of a backround scene with a resolutioin test image illuminated with white light and of a hot soldering iron tip (400 deg C) emiting infrared light (shown in orange).
Visualization 1 (without simplex encoding)
Exemplary video captured with a single-pixel camera (complementary sensing mode) using 2000 binar... more Exemplary video captured with a single-pixel camera (complementary sensing mode) using 2000 binary DCT sampling without simplex encoding. The video is recorded at 11.3 Hz frame rate with real-time reconstructions using fast Fourier domain regularized inverse algorithm. No post-processing was applied.
Differential real-time single-pixel imaging with Fourier domain regularization: applications to VIS-IR imaging and polarization imaging
Optics Express, 2021
The speed and quality of single-pixel imaging (SPI) are fundamentally limited by image modulation... more The speed and quality of single-pixel imaging (SPI) are fundamentally limited by image modulation frequency and by the levels of optical noise and compression noise. In an approach to come close to these limits, we introduce a SPI technique, which is inherently differential, and comprises a novel way of measuring the zeroth spatial frequency of images and makes use of varied thresholding of sampling patterns. With the proposed sampling, the entropy of the detection signal is increased in comparison to standard SPI protocols. Image reconstruction is obtained with a single matrix-vector product so the cost of the reconstruction method scales proportionally with the number of measured samples. A differential operator is included in the reconstruction and following the method is based on finding the generalized inversion of the modified measurement matrix with regularization in the Fourier domain. We demonstrate 256 × 256 SPI at up to 17 Hz at visible and near-infrared wavelength ranges...
Single-pixel video imaging with DCT sampling
Imaging and Applied Optics 2019 (COSI, IS, MATH, pcAOP), 2019
Discrete cosine transform elements are highly efficient sampling patterns for compressive single-... more Discrete cosine transform elements are highly efficient sampling patterns for compressive single-pixel video imaging. We discuss the influence of the ordering of these patterns on the quality of the reconstruction of a moving scene.
Balanced single-pixel camera with noiselet sampling
2017 19th International Conference on Transparent Optical Networks (ICTON), 2017
Single-pixel cameras (SPC) are image capturing devices, which use only a single detector to colle... more Single-pixel cameras (SPC) are image capturing devices, which use only a single detector to collect information about the entire image, by making use of its spatial modulation. Although a single detector is enough to obtain images of good quality in idealistic conditions, in practice the SPC systems are prone to all sorts of noise, including variations of light intensity over time, ambient illumination and signal oscillations induced by electronic components. We address this problem by studying a modification of a SPC set-up, in which two detectors are used to simultaneously measure the modulation of the image with the sensing patterns and their negations. We choose discrete noiselets for the sampling patterns and we apply the theory of compressive sensing to reduce the number of samples taken by the balanced SPC. Low mutual coherence of the noiselet functions and wavelets, as well as other properties such as fast calculation algorithm and possibility of transforming discrete noiselets into binary patterns (ready to display with binary SLM) makes them a solid choice for compressive image acquisition. We show that the combination of noiselet-based sampling with balanced SPC set-up leads to much better signal-to-noise ratio of the recovered image than it is possible in other configurations with the same equipment and measurement conditions.
ArXiv, 2016
We introduce a compressive pattern recognition method for non-adaptive WalshHadamard or discrete ... more We introduce a compressive pattern recognition method for non-adaptive WalshHadamard or discrete noiselet-based compressive measurements and show that images measured at extremely high compression rates may still contain sufficient information for pattern recognition and target localization. We report on a compressive pattern recognition experiment with a single-pixel detector with which we validate the proposed method. The correlation signals produced with the phaseonly matched filter or with the pure-phase correlation are obtained from the compressive measurements through lasso optimization without the need to reconstruct the original image. This is possible owing to the two properties of phase-only filtering: such filtering is a unitary circulant transform, and the correlation plane it produces in pattern recognition applications is usually sparse.
Scientific Reports, 2018
Single-pixel imaging is an indirect imaging technique which utilizes simplified optical hardware ... more Single-pixel imaging is an indirect imaging technique which utilizes simplified optical hardware and advanced computational methods. It offers novel solutions for hyper-spectral imaging, polarimetric imaging, three-dimensional imaging, holographic imaging, optical encryption and imaging through scattering media. The main limitations for its use come from relatively high measurement and reconstruction times. In this paper we propose to reduce the required signal acquisition time by using a novel sampling scheme based on a random selection of Morlet wavelets convolved with white noise. While such functions exhibit random properties, they are locally determined by Morlet wavelet parameters. The proposed method is equivalent to random sampling of the properly selected part of the feature space, which maps the measured images accurately both in the spatial and spatial frequency domains. We compare both numerically and experimentally the image quality obtained with our sampling protocol against widely-used sampling with Walsh-Hadamard or noiselet functions. The results show considerable improvement over the former methods, enabling single-pixel imaging at low compression rates on the order of a few percent.
Optics Express, 2018
We present a closed-form image reconstruction method for single pixel imaging based on the genera... more We present a closed-form image reconstruction method for single pixel imaging based on the generalized inverse of the measurement matrix. Its numerical cost scales linearly with the number of measured samples. Regularization is obtained by minimizing the norms of the convolution between the reconstructed image and a set of spatial filters, and the final reconstruction formula can be expressed in terms of matrix pseudoinverse. At high compression this approach is an interesting alternative to the methods of compressive sensing based on l1norm optimization, which are too slow for real-time applications. For instance, we demonstrate experimental single-pixel detection with real-time reconstruction obtained in parallel with the measurement at the frame rate of 11 Hz for highly compressive measurements with the resolution of 256 × 256. For this purpose, we preselect the sampling functions to match the average spectrum obtained with an image database. The sampling functions are selected from the Walsh-Hadamard basis, from the discrete cosine basis, or from a subset of Morlet wavelets convolved with white noise. We show that by incorporating the quadratic criterion into the closed-form reconstruction formula, we are able to use binary rather than continuous sampling reaching similar reconstruction quality as is obtained by minimizing the total variation. This makes it possible to use cosine or Morlet-based sampling with digital micromirror devices without advanced binarization methods.
Applied Optics, 2016
Minimal mutual coherence of discrete noiselets and Haar wavelets makes this pair of bases an esse... more Minimal mutual coherence of discrete noiselets and Haar wavelets makes this pair of bases an essential choice for the measurement and compression matrices in compressed-sensing-based single-pixel detectors. In this paper we propose an efficient way of using complex-valued and non-binary noiselet functions for object sampling in single-pixel cameras with binary spatial light modulators and incoherent illumination. The proposed method allows to determine m complex noiselet coefficients from m + 1 binary sampling measurements. Further, we introduce a modification to the complex fast noiselet transform, which enables computationally-efficient real-time generation of the binary noiselet-based patterns using efficient integer calculations on bundled patterns. The proposed method is verified experimentally with a single-pixel camera system using a binary spatial light modulator.
Uploads
Papers by Anna Pastuszczak