Wyniki wyszukiwania - BazTech

1

Application of a deep-learning neural network for image reconstruction from a single-pixel infrared camera

Urbaś Sebastian, Więcek Bogusław

Opto-Electronics Review

|

2024

|

Vol. 32, No. 1

art. no. e148877

EN

The article presents the simulation results of a single-pixel infrared camera image reconstruction obtained by using a convolutional neural network (CNN). Simulations were carried out for infrared images with a resolution of 80 × 80 pixels, generated by a low-cost, low-resolution thermal imaging camera. The study compares the reconstruction results using the CNN and the ℓ₁ reconstruction algorithm. The results obtained using the neural network confirm a better quality of the reconstructed images with the same compression rate expressed by the peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM).

2

The Development of a Generative Approach for Joint Super-Resolution Image Reconstruction from Highly Sparse Raw Data in the Context of MR-PET Imaging

Malczewski Krzysztof

Machine Graphics and Vision

|

2023

|

Vol. 32, No. 3/4

161--191

EN

The present study introduces a rapid and efficient approach for reconstructing high-resolution images in hybrid MRI-PET scanners. The application of sparsity, compressed sensing (CS), and super-resolution reconstruction (SRR) methodologies can significantly decrease the demands of data acquisition while concurrently attaining high-resolution output. G-guided generative multilevel networks for sparsely sampled MR-PET input are shown here. Compressed Sensing using conjugate symmetry and Partial Fourier methodology speeds up data collection over k-space sampling methods. GANs and k-space adjustments are used in this image domain technique. The employed methodology utilizes discrete preprocessing stages to effectively tackle the challenges associated with the deblurring, reducing motion artifacts, and denoising of layers. Initial trials offer contextual details and accelerate evaluations. Preliminary experiments provide contextual information and expedite assessments.

3

Compressed optical image encryption in the diffractive-imaging-based scheme by input plane and output plane random sampling

Wan Shujia, Gong Qiong, Wang Hongjuan, Ma Shibang, Qin Yi

Optica Applicata

|

2022

|

Vol. 52, nr 1

51--66

EN

The successful recovery of the plaintext in the simplified diffractive-imaging-based encryption (S-DIBE) scheme needs to record one intact axial intensity map as the ciphertext. By aid of compressive sensing, we propose here a new image encryption approach, referred to as compressed DIBE (C-DIBE), which allows further compression of the intensity map. The plaintext is sampled before being sent to DIBE. Afterwards, the intensity map recorded by the CCD camera is also processed by such sampling operation to generate the ciphertext. For decryption, we first obtain the sparse plaintext using the proposed phase retrieval algorithm, and then reobtain the primary plaintext from it via compressive sensing. Numerical results show that a proper proportion of the intensity map (e.g. 50%) is enough to totally recover a grayscale image. We achieve multiple-image encryption by space multiplexing without enlarging the size of the ciphertext. The robustness of C-DIBE against brute-force attack evidently outperforms S-DIBE due to the extended key space. Numerical simulation has been presented to confirm the proposal.

4

A Dai-Liao-type projection method for monotone nonlinear equations and signal processing

Ibrahim Abdulkarim Hassan, Kumam Poom, Abubakar Auwal Bala, Abdullahi Muhammad Sirajo, Mohammad Hassan

Demonstratio Mathematica

|

2022

|

Vol. 55, nr 1

978--1013

EN

In this article, inspired by the projection technique of Solodov and Svaiter, we exploit the simple structure, low memory requirement, and good convergence properties of the mixed conjugate gradient method of Stanimirović et al. [New hybrid conjugate gradient and broyden-fletcher-goldfarbshanno conjugate gradient methods, J. Optim. Theory Appl. 178 (2018), no. 3, 860–884] for unconstrained optimization problems to solve convex constrained monotone nonlinear equations. The proposed method does not require Jacobian information. Under monotonicity and Lipschitz continuity assumptions, the global convergence properties of the proposed method are established. Computational experiments indicate that the proposed method is computationally efficient. Furthermore, the proposed method is applied to solve the ℓ1 -norm regularized problems to decode sparse signals and images in compressive sensing.

5

Compressive sensing aided seismic geometry design for offshore acquisition

Huang Xiaogang, Zhang Jinmiao, Du Xiangdong, Zhu Zhenyu, Li Weixin, Zhang Yiming, Ding Jicai, Liu Zhipeng, Zhang Jianfeng

Acta Geophysica

|

2022

|

Vol. 70, no 2

547--562

EN

Seismic acquisition guided by the compressive sensing theory can significantly improve seismic data acquisition efficiency and reduce the cost. After reviewing the basic principles of compressive sensing, we propose an optimized random sampling method that can control the maximum sampling interval and improve the design flexibility. We analyze several factors that can introduce reconstruction errors from compressive sensed data and learn that besides sampling method, reconstruction errors increase with decimation degree and the complexity of structures and also depend on the reconstruction workflow. In addition, we provide a basic workflow of the geometry design of compressive sensing acquisition. We analyze the feasibility of the three types of receiving equipment that are widely used in marine environment and discuss the potential cost reduction and efficiency gain. Our field example demonstrates the detailed working process and the feasibility of the combination of random sailing line intervals and random shot intervals and verifies the effect of cost saving and efficiency increasing.

6

Color image encryption scheme based on quaternion discrete multi-fractional random transform and compressive sensing

Ye Huo-Sheng, Dai Jing-Yi, Wen Shun-Xi, Gong Li-Hua, Zhang Wen-Quan

Optica Applicata

|

2021

|

Vol. 51, nr 3

349--364

EN

A color image compression-encryption algorithm by combining quaternion discrete multi-fractional random transform with compressive sensing is investigated, in which the chaos-based fractional orders greatly improve key sensitivity. The original color image is compressed and encrypted with the assistance of compressive sensing, in which the partial Hadamard matrix adopted as a measurement matrix is constructed by iterating Chebyshev map instead of utilizing the entire Guassian matrix as a key. The sparse images are divided into 12 sub-images and then represented as three quaternion signals, which are modulated by the quaternion discrete multi-fractional random transform. The image blocking and the quaternion representation make the proposed cryptosystem avoid additional data extension existing in many transform-based methods. To further improve the level of security, the plaintext-related key streams generated by the 2D logistic-sine-coupling map are adopted to diffuse and confuse the intermediate results simultaneously. Consequently, the final ciphertext image is attained. Simulation results reveal that the proposed cryptosystem is feasible with high security and has strong robustness against various attacks.

7

Modified Block Sparse Bayesian Learning-Based Compressive Sensing Scheme For EEG Signals

Upadhyaya Vivek, Salim Mohammad

International Journal of Electronics and Telecommunications

|

2021

|

Vol. 67, No. 3

331--336

EN

Advancement in medical technology creates some issues related to data transmission as well as storage. In real-time processing, it is too tedious to limit the flow of data as it may reduce the meaningful information too. So, an efficient technique is required to compress the data. This problem arises in Magnetic Resonance Imaging (MRI), Electrocardiogram (ECG), Electroencephalogram (EEG), and other medical signal processing domains. In this paper, we demonstrate Block Sparse Bayesian Learning (BSBL) based compressive sensing technique on an Electroencephalogram (EEG) signal. The efficiency of the algorithm is described using the Mean Square Error (MSE) and Structural Similarity Index Measure (SSIM) value. Apart from this analysis we also use different combinations of sensing matrices too, to demonstrate the effect of sensing matrices on MSE and SSIM value. And here we got that the exponential and chi-square random matrices as a sensing matrix are showing a significant change in the value of MSE and SSIM. So, in real-time body sensor networks, this scheme will contribute a significant reduction in power requirement due to its data compression ability as well as it will reduce the cost and the size of the device used for real-time monitoring.

8

Sparse representation of a non-stationary signal in compressive sensing technique

Pałczyńska Beata, Świsulski Dariusz

Przegląd Elektrotechniczny

|

2019

|

R. 95, nr 11

66--68

EN

The paper presents the application of the compressive sensing technique to reconstruct a non-stationary signal based on compressed samples in the time-frequency domain. A greedy algorithm with different dictionaries to seek sparse atomic decomposition of the signal was applied. The results of the simulation confirm that the use of compressive sensing allows reconstruction of the non-stationary signal from a reduced number of randomly acquired samples, with slight loss of reconstruction quality.

PL

Przedstawiono zastosowanie techniki oszczędnego próbkowania do rekonstrukcji sygnału niestacjonarnego na podstawie skompresowanych próbek w dziedzinie czas-częstotliwość. Zastosowano nadmiarowy algorytm z różnymi słownikami aby znaleźć rzadką reprezentację sygnału. Wyniki symulacji potwierdzają, że zastosowanie oszczędnego próbkowania pozwala na rekonstrukcję sygnału niestacjonarnego z małej liczby losowo pobranych próbek, z niewielką utratą jakości rekonstrukcji.

9

Fast harmonics identification based on a compressive sensing approach

Pałczyńska Beata

Przegląd Elektrotechniczny

|

2019

|

R. 95, nr 11

69--71

EN

The paper presents the application of a fast reconstruction algorithm, based on the theory of compressive sensing that can detect harmonics in an input signal. The problem of signal reconstruction is solved using a convex optimization by the linear programming algorithm. Additionally, to accelerate the convergence, a K-rank-order filter is applied in the signal's sparse domain. The numerical simulation carried out confirms the effectiveness of the algorithm used.

PL

W pracy przedstawiono implementację szybkiego algorytmu rekonstrukcji sygnału, opartego na teorii oszczędnego próbkowania, który może wykrywać harmoniczne w sygnale wejściowym. Zagadnienie rekonstrukcji sygnału jest problemem optymalizacyjnym rozwiązywanym za pomocą algorytmu programowania liniowego. Dodatkowo, aby przyspieszyć zbieżność rozwiązania zastosowano w rzadkiej dziedzinie sygnału filtr typu K-rank-order. Przeprowadzona symulacja numeryczna potwierdza skuteczność zastosowanego algorytmu.

10

Image compression and encryption algorithm with wavelet-transform-based 2D compressive sensing

Fan Jing-Hui, Liu Xian-Bao, Chen Yan-Bin

Optica Applicata

|

2019

|

Vol. 49, nr 3

461--472

EN

By combining a wavelet transform with chaos scrambling, an image compression and encryption algorithm based on 2D compressive sensing is designed. The wavelet transform is employed to obtain the sparse representation of a plaintext image. The sparse image is measured in two orthogonal directions by compressive sensing. Then, the result of 2D compressive sensing is confused by the Arnold transform and the random pixel scrambling. The combination of four-dimensional chaos and logistic map is exploited to generate the first row of the key-controlled circulant matrix. The proposed algorithm not only carries out image compression and encryption simultaneously, but also reduces the consumption of the key by controlling the generation of measurement matrix. Experimental results reveal that the proposed image compression and encryption algorithm is resistant to noise attacks with good compression performance and high key sensitivity.

11

Compressive-sensing-based double-image encryption algorithm combining double random phase encoding with Josephus traversing operation

Jiang Hao, Nie Zhe, Zhou Nanrun, Zhang Wenquan

Optica Applicata

|

2019

|

Vol. 49, nr 3

445--459

EN

A double-image encryption scheme based on compressive sensing is designed by combining a double random phase encoding technique with Josephus traversing operation. Two original images are first compressed and encrypted by compressive sensing in the discrete wavelet domain and then connected into a complex image according to the order of the alternate rows. Moreover, the resulting image is re-encrypted into stationary white noise by a double random phase encoding technique. Lastly, Josephus traversing method is utilized to scramble the transformed image. The initial states of the Henon chaotic map are the secret keys of this double-image encryption algorithm, which can be used to control the construction of the measurement matrix in compressive sensing and generation of the random-phase mask in double random phase encoding. Simulation results show that the proposed double-image encryption algorithm is effective and secure.

12

Image compression-encryption algorithm combining compressive sensing with log operation

Chen R.-L., Zhou Y., Luo M., Zhang A.-D., Gong L.-H.

Optica Applicata

|

2018

|

Vol. 48, nr 4

563--573

EN

Based on compressive sensing and log operation, a new image compression-encryption algorithm is proposed, which accomplishes encryption and compression simultaneously. The proposed image compression-encryption algorithm takes advantage of not only the physical realizability of partial Hadamard matrix, but also the resistance of the chosen-plaintext attack since all the elements in the partial Hadamard matrix are 1, –1 or log 1 = 0. The proposed algorithm is sensitive to the key and it can resist various common attacks. The simulation results verify the validity and reliability of the proposed image compression-encryption algorithm.

13

Pilot Design for Sparse Channel Estimation in Orthogonal Frequency Division Multiplexing Systems

Vimala P., Yamuna G.

Journal of Telecommunications and Information Technology

|

2018

|

nr 2

60--68

EN

Orthogonal Frequency Division Multiplexing (OFDM) is a well-known technique used in modern wide band wireless communication systems. Coherent OFDM systems achieve its advantages over a multipath fading channel, if channel impulse response is estimated precisely at the receiver. Pilot-aided channel estimation in wide band OFDM systems adopts the recently explored compressive sensing technique to decrease the transmission overhead of pilot subcarriers, since it exploits the inherent sparsity of the wireless fading channel. The accuracy of compressive sensing techniques in sparse channel estimation is based on the location of pilots among OFDM subcarriers. A sufficient condition for the optimal pilot selection from Sylow subgroups is derived. A Sylow subgroup does not exist for most practical OFDM systems. Therefore, a deterministic pilot search algorithm is described to select pilot locations based on minimizing coherence, along with minimum variance. Simulation results reveal the effectiveness of the proposed algorithm in terms of bit error rate, compared to the existing solutions.

14

Sparsity-based methods for processing of radar signals

Misiurewicz J.

Prace Naukowe Politechniki Warszawskiej. Elektronika

|

2013

|

z. 186

5--136

EN

This work summarizes the author's research on radar applications of methods resulting from the assumption of signal sparsity. The term sparsity means that a signal under investigation may be modeled with a small number of components taken from a large dictionary. This property makes it possible to employ a new class of mathematical methods, recently made known as Compressive Sensing framework, for recovering the signal from the measured samples. The main feature of sparsity-based methods is that they can recover a signal uniquely from much fewer samples than methods derived from the classical sampling theory. However, this is possible only if me sparse model is adequate and if the model dictionary and measurement process conform to the specific requirements of the mathematical framework. In the present work, the author demonstrates how the mathematical theory of sparse representation and recovery may be applied to practical problems arising in radar signal processing. An overall purpose of radar signal processing is to acquire the knowledge of the radar scene from the received echo of a radio frequency signal which illuminates the investigated area. This is a problem generally belonging to the class of inverse problems, which may be ill-conditioned and ambiguous. The assumption of the sparse model of the received signal is an innovative idea that opens new possibilities of resolving ambiguities. The aim of this work was to demonstrate by means of practical examples that sparse reconstruction methods are capable of solving a series of important problems in different areas of radar signal processing. Also, more detailed research was done on these cases, including the study on sampling requirements as well as simulations of the algorithms used. The ideas and methods were verified with the use of live recorded signals wherever possible. In the examples presented in this work, sparsity of the signal model is the key assumption to enable the solution of relevant inverse problems. The application areas described here are closely related to the author's experience with existing radar systems, including those currently under research or development at the Warsaw University of Technology. They cover a wide range of radar types and processing modes, including active and passive radars as well as surveillance and imaging ones. The author proposed applications of sparsity-based methods for active radars with a noise waveform, for classical MTI radars, and for imaging radars, using either die synthetic aperture (SAR) technique with noise illumination, or the inverse synthetic aperture (ISAR) technique with passive illumination from a GSM transmitter. The ideas and methods were verified with the use of live recorded signals wherever possible. In the examples presented in this work, sparsity of the signal model is the key assumption to enable the solution of relevant inverse problems. The application areas described here are closely related to the author's experience with existing radar systems, including those currently under research or development at the Warsaw University of Technology. They cover a wide range of radar types and processing modes, including active and passive radars as well as surveillance and imaging ones. The author proposed applications of sparsity-based methods for active radars with a noise waveform, for classical MTI radars, and for imaging radars, using either die synthetic aperture (SAR) technique with noise illumination, or the inverse synthetic aperture (ISAR) technique with passive illumination from a GSM transmitter. The ideas and methods were verified with the use of live recorded signals wherever possible. In the examples presented in this work, sparsity of the signal model is the key assumption to enable the solution of relevant inverse problems. The application areas described here are closely related to the author's experience with existing radar systems, including those currently under research or development at the Warsaw University of Technology. They cover a wide range of radar types and processing modes, including active and passive radars as well as surveillance and imaging ones. The author proposed applications of sparsity-based methods for active radars with a noise waveform, for classical MTI radars, and for imaging radars, using either die synthetic aperture (SAR) technique with noise illumination, or the inverse synthetic aperture (ISAR) technique with passive illumination from a GSM transmitter. The ideas and methods were verified with the use of live recorded signals wherever possible. In the examples presented in this work, sparsity of the signal model is the key assumption to enable the solution of relevant inverse problems. The application areas described here are closely related to the author's experience with existing radar systems, including those currently under research or development at the Warsaw University of Technology. They cover a wide range of radar types and processing modes, including active and passive radars as well as surveillance and imaging ones. The author proposed applications of sparsity-based methods for active radars with a noise waveform, for classical MTI radars, and for imaging radars, using either die synthetic aperture (SAR) technique with noise illumination, or the inverse synthetic aperture (ISAR) technique with passive illumination from a GSM transmitter. The ideas and methods were verified with the use of live recorded signals wherever possible. In the examples presented in this work, sparsity of the signal model is the key assumption to enable the solution of relevant inverse problems. The application areas described here are closely related to the author's experience with existing radar systems, including those currently under research or development at the Warsaw University of Technology. They cover a wide range of radar types and processing modes, including active and passive radars as well as surveillance and imaging ones. The author proposed applications of sparsity-based methods for active radars with a noise waveform, for classical MTI radars, and for imaging radars, using either die synthetic aperture (SAR) technique with noise illumination, or the inverse synthetic aperture (ISAR) technique with passive illumination from a GSM transmitter.The ideas and methods were verified with the use of live recorded signals wherever possible. In the examples presented in this work, sparsity of the signal model is the key assumption to enable the solution of relevant inverse problems. The application areas described here are closely related to the author's experience with existing radar systems, including those currently under research or development at the Warsaw University of Technology. They cover a wide range of radar types and processing modes, including active and passive radars as well as surveillance and imaging ones. The author proposed applications of sparsity-based methods for active radars with a noise waveform, for classical MTI radars, and for imaging radars, using either die synthetic aperture (SAR) technique with noise illumination, or the inverse synthetic aperture (ISAR) technique with passive illumination from a GSM transmitter. The ideas and methods were verified with the use of live recorded signals wherever possible. In the examples presented in this work, sparsity of the signal model is the key assumption to enable the solution of relevant inverse problems. The application areas described here are closely related to the author's experience with existing radar systems, including those currently under research or development at the Warsaw University of Technology. They cover a wide range of radar types and processing modes, including active and passive radars as well as surveillance and imaging ones. The author proposed applications of sparsity-based methods for active radars with a noise waveform, for classical MTI radars, and for imaging radars, using either die synthetic aperture (SAR) technique with noise illumination, or the inverse synthetic aperture (ISAR) technique with passive illumination from a GSM transmitter. vThe ideas and methods were verified with the use of live recorded signals wherever possible. In the examples presented in this work, sparsity of the signal model is the key assumption to enable the solution of relevant inverse problems. The application areas described here are closely related to the author's experience with existing radar systems, including those currently under research or development at the Warsaw University of Technology. They cover a wide range of radar types and processing modes, including active and passive radars as well as surveillance and imaging ones. The author proposed applications of sparsity-based methods for active radars with a noise waveform, for classical MTI radars, and for imaging radars, using either die synthetic aperture (SAR) technique with noise illumination, or the inverse synthetic aperture (ISAR) technique with passive illumination from a GSM transmitter. The ideas and methods were verified with the use of live recorded signals wherever possible. In the examples presented in this work, sparsity of the signal model is the key assumption to enable the solution of relevant inverse problems. The application areas described here are closely related to the author's experience with existing radar systems, including those currently under research or development at the Warsaw University of Technology. They cover a wide range of radar types and processing modes, including active and passive radars as well as surveillance and imaging ones. The author proposed applications of sparsity-based methods for active radars with a noise waveform, for classical MTI radars, and for imaging radars, using either die synthetic aperture (SAR) technique with noise illumination, or the inverse synthetic aperture (ISAR) technique with passive illumination from a GSM transmitter. The ideas and methods were verified with the use of live recorded signals wherever possible. In the examples presented in this work, sparsity of the signal model is the key assumption to enable the solution of relevant inverse problems. The application areas described here are closely related to the author's experience with existing radar systems, including those currently under research or development at the Warsaw University of Technology. They cover a wide range of radar types and processing modes, including active and passive radars as well as surveillance and imaging ones. The author proposed applications of sparsity-based methods for active radars with a noise waveform, for classical MTI radars, and for imaging radars, using either die synthetic aperture (SAR) technique with noise illumination, or the inverse synthetic aperture (ISAR) technique with passive illumination from a GSM transmitter.The ideas and methods were verified with the use of live recorded signals wherever possible. In the examples presented in this work, sparsity of the signal model is the key assumption to enable the solution of relevant inverse problems. The application areas described here are closely related to the author's experience with existing radar systems, including those currently under research or development at the Warsaw University of Technology. They cover a wide range of radar types and processing modes, including active and passive radars as well as surveillance and imaging ones. The author proposed applications of sparsity-based methods for active radars with a noise waveform, for classical MTI radars, and for imaging radars, using either die synthetic aperture (SAR) technique with noise illumination, or the inverse synthetic aperture (ISAR) technique with passive illumination from a GSM transmitter. The ideas and methods were verified with the use of live recorded signals wherever possible. In the examples presented in this work, sparsity of the signal model is the key assumption to enable the solution of relevant inverse problems. The application areas described here are closely related to the author's experience with existing radar systems, including those currently under research or development at the Warsaw University of Technology. They cover a wide range of radar types and processing modes, including active and passive radars as well as surveillance and imaging ones. The author proposed applications of sparsity-based methods for active radars with a noise waveform, for classical MTI radars, and for imaging radars, using either die synthetic aperture (SAR) technique with noise illumination, or the inverse synthetic aperture (ISAR) technique with passive illumination from a GSM transmitter.

PL

Praca podsumowuje badania autora na temat radiolokacyjnego zastosowania nowatorskich metod wynikających z założenia o rzadkim modelu sygnału. Pojęcie to oznacza, że dany sygnał może być modelowany jako liniowa kombinacja niewielkiej liczby składowych należących do pewnego, z założenia pojemnego, słownika. Przyjęcie lego założenia otwiera możliwość zastosowania nowej klasy metod matematycznych, znanych od niedawna pod wspólna nazwą "Compressive Sensing" (po polsku ostatnio używa się określenia "oszczędne próbkowanie"), do odtworzenia sygnału ze zmierzonych próbek. Najważniejsza cechą tych metod jest możliwość odtworzenia sygnału ze znacznie mniejszej liczby próbek niż wynikałoby to z założeń metod klasycznych (opartych na twierdzeniu o próbkowaniu). Należy jednak podkreślić, że jest to możliwe tylko pod warunkiem adekwatności modelu rzadkiego oraz spełnienia pewnych wymagań przez słownik oraz przez proces pomiaru, zgodnie z teorią matematyczną oszczędnego próbkowania. W przedstawionej pracy autor pokazuje, w jaki sposób rzadki model sygnału i matematyczna teoria oszczędnego próbkowania mogą być użyte w zagadnieniach praktycznie występujących w radiolokacji. Ogólnym celem przetwarzania sygnałów radiolokacyjnych jest uzyskanie wiedzy o oświetlanej scenie poprzez badanie sygnału radiowego od tej sceny odbitego. Jest to problem z dziedziny zagadnień odwrotnych (inverse problems), i jako taki może być źle uwarunkowany i nie mieć jednoznacznego rozwiązania. Ograniczenie swobody szukanego rozwiązania poprzez przyjęcie rzadkiego modelu otwiera nowe możliwości usunięcia niejednoznaczności wyniku. Autor zaproponował w pracy wykorzystanie metod opartych na rzadkiej reprezentacji sygnału do modelowania silnych ech w celu usunięcia ich wpływu na proces detekcji cech słabych, zwanego efektem maskowania. Ma to zastosowanie w aktywnych radarach z oświetleniem szumowym i w radarach pasywnych. Autor badał kilka algorytmów w aspekcie modelowania ech złożonych, wskazał przyczyny niepowodzenia algorytmu kolejnego usuwania składowych i zaproponował w zamian algorytm modelowania łącznego ograniczonej liczby składowych. Kolejne zaproponowane przez autora rozwiązanie przeznaczone jest dla szumowego radaru z syntetyczną aperturą (SAR). Pozwala ono zmniejszyć znacznie liczbę pobieranych próbek w wymiarze przestrzennym a zatem i łączny czas akwizycji sygnału. Zastosowanie rzadkiego modelu sygnału pozwoliło rozwikłać niejednoznaczność odtworzenia obrazu sceny radarowej powstałą wskutek zmniejszenia częstości próbkowania poniżej granicy Nyquista. Rozwiązanie zostało przebadane w symulacjach i w eksperymentach z wykorzystaniem zarejestrowanych rzeczywistych sygnałów. W zastosowaniu do klasycznych radarów MTI, autor zaproponował zastosowanie rzadkiego modelu sygnału w dziedzinie częstotliwości w celu zwiększenia zdolności pomiaru prędkości kilku obiektów znajdujących się w tej samej odległości od radaru. Następne zaproponowane przez autora rozwiązanie dotyczy estymacji wysokości w radarze pasywnym. Na przykładzie radaru pasywnego pracującego z wykorzystaniem nadajnika telewizji cyfrowej DVB-T autor pokazał, że metody oszczędnego próbkowania pozwalają na rozdzielenie ech obiektu propagujących z i bez odbicia od ziemi. Dysponując pomiarami opóźnienia ech wzdłuż różnych dróg propagacji można określić wysokość obiektu. Jest to istotna innowacja wobec faktu, że określenie wysokości poprzez pomiar kąta przyjścia sygnału jest bardzo trudne przy typowych rozmiarach anten dla pasma telewizji cyfrowej. Przy pracy radaru pasywnego z wykorzystaniem niewielkiej liczby nadajników uzyskanie niezależnego pomiaru wysokości obiektu znacząco poprawia jakość lokalizacji obiektu w przestrzeni trójwymiarowej. Rozpatrując zagadnienie obrazowania obiektów ruchomych w radarze pasywnym wykorzystującym nadajnik GSM, autor zaproponował metodę uzyskiwania wyraźnego obrazu w technice ISAR (odwrotnej syntetycznej apertury), korzystając z faktu, że ruchomy obiekt przy obrazowaniu ISAR może być uważany za zbiór niewielu punktów odbijających, poruszających się w jednolity sposób. Zastosowanie tej metody pozwoliło z rzeczywistych, zarejestrowanych sygnałów uzyskać obraz ISAR jadącego pojazdu. Klasyczne metody oparte na filtracji dopasowanej w tej samej sytuacji zawiodły, gdyż cały obserwowany obiekt był mniejszy niż rozmiar komórki rozdzielczości odległościowej, który wynika z szerokości pasma sygnału. W opisanych przez autora przykładach zastosowań rzadkość modelu jest kluczowym założeniem przy rozwiązywaniu postawionych zagadnień odwrotnych. Przykłady dotyczą obszarów zastosowań ściśle związanych z długoletnim doświadczeniem autora w projektowaniu i konstruowaniu urządzeń radarowych, w tym eksperymentalnych urządzeń aktualnie opracowywanych w Politechnice Warszawskiej. Zastosowania przedstawione w pracy obejmują szerokie spektrum różnych typów radarów i sposobów ich wykorzystania, włączając w to radary pasywne i aktywne czy też radary przeznaczone do wykrywania obiektów lub tworzenia ich obrazów.

15

Semantic Sparse Representation of Disease Patterns

Przelaskowski A.

International Journal of Electronics and Telecommunications

|

2010

|

Vol. 56, No. 3

273-280

EN

Sparse data representation is discussed in a context of useful fundamentals led to semantic content description and extraction of information. Disease patterns as semantic information extracted from medical images were underlined because of discussed application of computer-aided diagnosis. Compressive sensing rules were adjusted to the requirements of diagnostic pattern recognition. Proposed methodology of sparse disease patterns considers accuracy of sparse representation to estimate target content for detailed analysis. Semantics of sparse representation were modeled by morphological content analysis. Subtle or hidden components were extracted and displayed to increase information completeness. Usefulness of sparsity was verified for computer-aided diagnosis of stroke based on brain CT scans. Implemented method was based on selective and sparse representation of subtle hypodensity to improve diagnosis. Visual expression of disease signatures was fixed to radiologist requirements, domain knowledge and experimental analysis issues. Diagnosis assistance suitability was proven by experimental subjective rating and automatic recognition.