RESUMEN
Imaging through scattering media is a classical inverse issue in computational imaging. In recent years, deep learning(DL) methods have excelled in speckle reconstruction by extracting the correlation of speckle patterns. However, high-performance DL-based speckle reconstruction also costs huge hardware computation and energy consumption. Here, we develop an opto-electronic DL method with low computation complexity for imaging through scattering media. We design the "end-to-end" optronic structure for speckle reconstruction, namely optronic fully convolutional neural network (OP-FCNN). In OP-FCNN, we utilize lens groups and spatial light modulators to implement the convolution, down/up-sampling, and skip connection in optics, which significantly reduces the computational complexity by two orders of magnitude, compared with the digital CNN. Moreover, the reconfigurable and scalable structure supports the OP-FCNN to further improve imaging performance and accommodate object datasets of varying complexity. We utilize MNIST handwritten digits, EMNIST handwritten letters, fashion MNIST, and MIT-CBCL-face datasets to validate the OP-FCNN imaging performance through random diffusers. Our OP-FCNN reveals a good balance between computational complexity and imaging performance. The average imaging performance on four datasets achieves 0.84, 0.91, 0.79, and 16.3dB for JI, PCC, SSIM, and PSNR, respectively. The OP-FCNN paves the way for all-optical systems in imaging through scattering media.
RESUMEN
The right sinus of the Valsalva aneurysm (SVA) rupturing into the right atrium (RA) and dissecting into the interventricular septum (IVS) is rare. The disease can be definitively diagnosed using two-dimensional (2D) echocardiography and color Doppler ultrasonography. Real-time biplane imaging and three-dimensional (3D) echocardiography offer new perspectives for viewing and diagnosing this disease.
Asunto(s)
Aneurisma Roto , Aneurisma de la Aorta , Disección Aórtica , Rotura de la Aorta , Seno Aórtico , Tabique Interventricular , Humanos , Aneurisma de la Aorta/complicaciones , Aneurisma de la Aorta/diagnóstico por imagen , Seno Aórtico/diagnóstico por imagen , Atrios Cardíacos/diagnóstico por imagenRESUMEN
The development of optical neural networks greatly slows the urgent demand of searching for fast computing approaches to solve big data processing. However, most optical neural networks following electronic training and optical inferencing do not really take full advantage of optical computing to reduce computational burden. Take the extensively used optronic convolutional neural networks (OPCNN) as an example, the convolutional operations still require vast computational operations in training stages on the computer. To address this issue, this study proposes the in-situ training algorithm to train the networks directly in optics. We derive the backpropagation algorithms of OPCNN hence the complicated gradient calculation in backward propagating processes can be obtained through optical computing. Both forward propagation and backward propagation are all executed on the same optical system. Furthermore, we successfully realize the introduction of optical nonlinearity in networks through utilizing photorefractive crystal SBN:60 and we also derive the corresponding backpropagation algorithm. The numerical simulation results of classification performance on several datasets validates the feasibility of the proposed algorithms. Through in-situ training, the reduction in performance resulting from the inconsistency of the plantform between training and inferencing stages can be eliminated completely. For example, we demonstrate that by using the optical training approach, OPCNN is capable of gaining a strong robustness under several misalignmed situations, which enhances the practicability of OPCNN and greatly expands its application range.
RESUMEN
Although deeper convolutional neural networks (CNNs) generally obtain better performance on classification tasks, they incur higher computation costs. To address this problem, this study proposes the optronic convolutional neural network (OPCNN) in which all computation operations are executed in optics, and data transmission and control are executed in electronics. In OPCNN, we implement convolutional layers with multi input images by the lenslet 4f system, downsampling layers by optical-strided convolution and obtaining nonlinear activation by adjusting the camera's curve and fully connected layers by optical dot product. The OPCNN demonstrates good performance on the classification tasks in simulations and experiments and achieves better performance than other current optical convolutional neural networks by comparison due to the more complex architecture. The scalability of OPCNN contributes to building deeper networks when facing complicated datasets.
RESUMEN
Imaging through scattering media is a common practice in many applications of biomedical imaging. Object image would deteriorate into unrecognizable speckle pattern when scattering media is presented. Many methods have been investigated to reconstruct the object image when only speckle pattern is available. In this paper, we demonstrate a method of single-shot imaging through scattering media. This method is based on classification and support vector regression of the measured speckle pattern. We prove the possibility of speckle pattern classification and related formulas are presented. The specified and limited imaging capability without speckle pattern classification is demonstrated. Our proposed approach, that is, speckle pattern classification based support vector regression method, makes up the deficiency. Experimental results show that, with our approach, speckle patterns could be utilized for classification when object images are unavailable, and object images can be reconstructed with high fidelity. The proposed approach for imaging through scattering media is expected to be applicable to various sensing schemes.
RESUMEN
In this paper, the problem of cognitive radar (CR) waveform optimization design for target detection and estimation in multiple extended targets situations is investigated. This problem is analyzed in signal-dependent interference, as well as additive channel noise for extended targets with unknown target impulse response (TIR). To address this problem, an improved algorithm is employed for target detection by maximizing the detection probability of the received echo on the promise of ensuring the TIR estimation precision. In this algorithm, an additional weight vector is introduced to achieve a trade-off among different targets. Both the estimate of TIR and transmit waveform can be updated at each step based on the previous step. Under the same constraint on waveform energy and bandwidth, the information theoretical approach is also considered. In addition, the relationship between the waveforms that are designed based on the two criteria is discussed. Unlike most existing works that only consider single target with temporally correlated characteristics, waveform design for multiple extended targets is considered in this method. Simulation results demonstrate that compared with linear frequency modulated (LFM) signal, waveforms designed based on maximum detection probability and maximum mutual information (MI) criteria can make radar echoes contain more multiple-target information and improve radar performance as a result.
RESUMEN
We report in this paper on the study of surface acoustic wave (SAW) resonators based on an AlN/titanium alloy (TC4) structure. The AlN/TC4 structure with different thicknesses of AlN films was simulated, and the acoustic propagating modes were discussed. Based on the simulation results, interdigital transducers with a periodic length of 24 µm were patterned by lift-off photolithography techniques on the AlN films/TC4 structure, while the AlN film thickness was in the range 1.5-3.5 µm. The device performances in terms of quality factor (Q-factor) and electromechanical coupling coefficient (k²) were determined from the measure S11 parameters. The Q-factor and k² were strongly dependent not only on the normalized AlN film thickness but also on the full-width at half-maximum (FWHM) of AlN (002) peak. The dispersion curve of the SAW phase velocity was analyzed, and the experimental results showed a good agreement with simulations. The temperature behaviors of the devices were also presented and discussed. The prepared SAW resonators based on AlN/TC4 structure have potential applications in integrated micromechanical sensing systems.
RESUMEN
High temperature characteristics of langasite surface acoustic wave (SAW) devices coated with an AlN thin film have been investigated in this work. The AlN films were deposited on the prepared SAW devices by mid-frequency magnetron sputtering. The SAW devices coated with AlN films were measured from room temperature to 600 °C. The results show that the SAW devices can work up to 600 °C. The AlN coating layer can protect and improve the performance of the SAW devices at high temperature. The SAW velocity increases with increasing AlN coating layer thickness. The temperature coefficients of frequency (TCF) of the prepared SAW devices decrease with increasing thickness of AlN coating layers, while the electromechanical coupling coefficient (K²) of the SAW devices increases with increasing AlN film thickness. The K² of the SAW devices increases by about 20% from room temperature to 600 °C. The results suggest that AlN coating layer can not only protect the SAW devices from environmental contamination, but also improve the K² of the SAW devices.
RESUMEN
Two Surface acoustic wave (SAW) resonators were fabricated on langasite substrates with Euler angle of (0°, 138.5°, 117°) and (0°, 138.5°, 27°). A dipole antenna was bonded to the prepared SAW resonator to form a wireless sensor. The characteristics of the SAW sensors were measured by wireless frequency domain interrogation methods from 20 °C to 600 °C. Different temperature behaviors of the sensors were observed. Strain sensing was achieved using a cantilever configuration. The sensors were measured under applied strain from 20 °C to 500 °C. The shift of the resonance frequency contributed merely by strain is extracted from the combined effects of temperature and strain. Both the strain factors of the two SAW sensors increase with rising ambient temperature, and the SAW sensor deposited on (0°, 138.5°, 117°) cut is more sensitive to applied strain. The measurement errors of the two sensors are also discussed. The relative errors of the two sensors are between 0.63% and 2.09%. Even at 500 °C, the hysteresis errors of the two sensors are less than 5%.
RESUMEN
The health of city inhabitants is considerably impacted by the urban park green spaces (UPGS). Existing studies lack an assessment of the UPGS equity from different perspectives and are limited to poor accuracy. This study reviews the definition and determination methods of UPGS, as well as UPGS accessibility and spatial equity related studies. Then, a spatial equity evaluation system is established from the dimensions of equity in providing UPGS services and equity in supply and demand matching. This study extrapolates from micro to macro analysis using network big data and census data to pinpoint the population down to the building level in the main urban area of Fuzhou City, China. The network analysis method, improved Gaussian floating catchment area method, Gini coefficient, and Lorenz curve measurement help to evaluate UPGS service euquity, explore the similarities and differences of UPGS fairness under different travel modes. It also helps to understand the service relationship between UPGS service and population demand under objective space, and to pinpoint the weak supply area using the locational entropy method. The results show that: (1) The overall accessibility and per capita accessibility show similar spatial distribution patterns under different travel modes, which are "high around and low in the middle" and "centered on the Minjiang River and dispersed to the north and south, respectively" in study area. (2) The supply of UPGS services in Fuzhou is relatively adequate and at a relatively equity level, while varies in the allocation of UPGS services among sub-districts. The spatial equity of different traveling modes varies widely. (3) The sub-districts on the periphery of the study area should be subject to increased UPGS and improvements to the road network and public infrastructure. Sub-district with high population density and old neighborhoods should implement micro-renewal and build community parks. Our study presents a new idea for spatial equity research.
Asunto(s)
Parques Recreativos , Viaje , Humanos , Población Urbana , Ciudades , ChinaRESUMEN
Rapid urbanization has gradually increased the contradiction between the demand and supply of urban resources. The quantitative optimization and adjustment of the infrastructure of the 15-min living circle is conducive to the scientific formulation of living circle planning guidelines, and also allows the evaluation of the effectiveness and practicality of policies. We investigate the spatial allocation pattern of infrastructure construction, the actual service capacity of facilities, and the spatial matching of facility service supply and residents' demand from the spatial dimension. Taking Fuzhou City as an example, this study uses multi-source network big data to accurately quantify the supply and demand, and constructs a 15-min living circle facility service supply evaluation system based on the kernel density analysis method, the network analysis method, and the supply and demand matching model. We propose infrastructure enhancement strategies in conjunction with the current status of Fuzhou's urban development. This study also further explores the factors influencing the spatial distribution of basic service facilities and the construction status of community living circles in China. The results show that (a) The distribution of infrastructure presents a different spatial distribution pattern from the actual service supply within the living circle. (b) The infrastructure service in the main area of Fuzhou can basically cope with the demand of residents, but there is still a mismatch between supply and demand. The areas with insufficient supply are mainly distributed in the periphery of the study area. (c) In order to further improve the construction of community living circle, we should first focus on the sub-districts with low service supply level and insufficient supply. According to the specific distribution of facilities, the number of public service facilities should be increased or decreased based on the demand of residents. This paper enriches the practical application of multi-source network big data in urban infrastructure construction, provides a guideline for the spatial layout and resource allocation of infrastructure in 15-min living circles in other cities.
RESUMEN
The spatial distribution of urban green spaces (UGS) is closely related to the health of residents and the ecological pattern of cities. Exploring the equity of UGSs plays an important role in urban planning and also provides guidance for urban development. Taking the main urban area of Fuzhou City as an example, this study uses network big data and census data to pinpoint the population demand, evaluates the accessibility and equity of UGS within the basic living circle, neighborhood living circle and daily living circle of residents at the scale of residential and sub-districts. Based on the G2SFCA model, we also quantify the actual effective UGS's service capacity. Then, using the scale and travel range as the entry point, we further discuss the similarities and differences under different scales and different travel ranges. Finally, optimization strategies are proposed for the construction status. The results show that: (1) The spatial allocation of urban green space resources varies significantly, and there is a serious inequity in the spatial distribution of urban green space under pedestrian conditions; (2) The results of UGS accessibility, equity, and service capacity in Fuzhou at both residential and sub-district scales are consistent; (3) Urban construction should be multi-level overall planning, combined with local economic and social development factors in accordance with local conditions to take measures. The results of the study can provide a scientific reference for the optimization of the spatial distribution of UGS.
Asunto(s)
Parques Recreativos , Caminata , Ciudades , China , ViajeRESUMEN
Immunotherapy using dendritic cell (DC)-based vaccination is an established approach for treating cancer and infectious diseases; however, its efficacy is limited. Therefore, targeting the restricted migratory capacity of the DCs may enhance their therapeutic efficacy. In this study, the effect of laponite (Lap) on DCs, which can be internalized into lysosomes and induce cytoskeletal reorganization via the lysosomal reprogramming-calcium flicker axis, is evaluated, and it is found that Lap dramatically improves the in vivo homing ability of these DCs to lymphoid tissues. In addition, Lap improves antigen cross-presentation by DCs and increases DC-T-cell synapse formation, resulting in enhanced antigen-specific CD8+ T-cell activation. Furthermore, a Lap-modified cocktail (Lap@cytokine cocktail [C-C]) is constructed based on the gold standard, C-C, as an adjuvant for DC vaccines. Lap@C-C-adjuvanted DCs initiated a robust cytotoxic T-cell immune response against hepatitis B infection, resulting in > 99.6% clearance of viral DNA and successful hepatitis B surface antigen seroconversion. These findings highlight the potential value of Lap as a DC vaccine adjuvant that can regulate DC homing, and provide a basis for the development of effective DC vaccines.
Asunto(s)
Calcio , Vacunas , Linfocitos T CD8-positivos , Antígenos , Adyuvantes Inmunológicos , Citocinas , Lisosomas , Antivirales , Células DendríticasRESUMEN
A key problem in code-division multiple access (CDMA) system is to mitigate the multiple access interference (MAI) from other users while detecting the desired user. The performance of the conventional minimum output energy (MOE) multiuser detector for CDMA system significantly degrades in the presence of signature waveform distortions induced by multipath propagation or timing asynchronism. In this paper, a robust linear programming (ROLP) algorithm for blind multiuser detection is proposed. Different from the existing MOE-based multiuser detection techniques, the proposed ROLP minimizes the â ∞ -norm of the output to exploit the non-Gaussianity of the communication signals. To achieve robustness against signature waveform mismatch, the proposed method constrains the magnitude response of any signature vector within a specified uncertainty set to exceed unity. The uncertainty set is modeled as a rhombus, which differs from the spherical uncertainty region widely taken in the existing robust multiuser detectors. The resulting optimization problem is reformulated into a linear programming program and hence can be solved efficiently. The proposed ROLP is computationally simpler than its robust counterparts that requires solving a second-order cone programming. Simulation results demonstrate the superiority of the ROLP over several robust detectors, which indicate that its performance approaches the optimal performance bound.
RESUMEN
The Minjiang River Basin is one of the first pilot areas for ecological conservation and the restoration of mountain-river-forest-farmland-lake-grass in China. Taking the Minjiang River Basin as an example, this paper selected the importance of ecosystem service functions and ecological sensitivity to evaluate the ecological environment and identify ecological sources. Furthermore, we constructed an ecological resistance surface using artificial and natural interference factors. Through a minimum cumulative resistance model (MCR), the ecological security pattern (ESP) of "two barriers, one belt, many corridors, and many spots" was constructed. Research shows that: (1) In total, 43 ecological sources were identified, with a total area of 523 km2, accounting for 0.6% of the total land area. These were mainly distributed in the southwest and northwest of the Minjiang River Basin, such as in Zhangping, covered forest land, and cultivated land. (2) The connectivity of the network was low, and the spatial distribution of the ecological pinch points was uneven. A total of 118 ecological corridors and 22 important ecological pinch points were identified. The total length of the ecological corridor is 3,732,051.88 km, which is dense on the left side and sparse on the right side. (3) The ecological restoration area was composed of a low ecological safety area and a lower ecological safety area; the ecological control area was composed of a medium ecological safety area and a higher ecological safety area; and the ecological conservation area was composed of a high ecological safety area, at 6.5%, 27.7%, and 65.8%, respectively. Constructing the ESP of the Minjiang River Basin is important for promoting harmonious socioeconomic development and ecological protection. In addition, it can provide a reference basis for other experimental areas of mountain-river-forest-farmland-lake-grass.
Asunto(s)
Ecosistema , Ríos , China , Conservación de los Recursos Naturales , BosquesRESUMEN
In highly fragmented urban areas, plant diversity of remnant vegetation may depend not only on community structure and topographical factors, but also on landscape heterogeneity. Different buffer radius settings can affect the relative importance of these factors to plant diversity. The aim of this study was to examine the relative importance of landscape heterogeneity, community structure, and topographical factors on plant diversity under different buffer radii in biodiversity hotspots. We established 48 plots of remnant vegetation in Guangzhou city, one of the biodiversity hotspots. A buffer radius of 500 m, 1000 m, and 2000 m was established around the center of each sample plot, and 17 landscape heterogeneity indices in each buffer were calculated by FRAGSTATS 4.2 software. Combined with the community structure and topographical factors, the impact factors of plant diversity under different buffer radii were analyzed by multiple regression analysis. We found the following: (1) The combined explanatory power of the three factors accounted for 43% of the species diversity indices and 62% of the richness index at its peak. The three impact factors rarely act independently and usually create comprehensive cumulative effects. (2) Scale does matter in urban landscape studies. At a 500 m buffer radius, community structure combined with road disturbance indices was strongly related to diversity indices in herb and shrub layers. The stand age was negatively correlated with the tree-layer richness index. As the scale increased, the diversity indices and richness index in the three layers decreased or increased under the influence of comprehensive factors. (3) The richness index in the herb layer was more responsive to impact factors than other biodiversity indices. Except for the herb layer, the interpretation of landscape heterogeneity for each plant diversity index was more stable than that for the other two factors. Road disturbance indices, combined with the other six landscape pattern metrics, can well indicate species diversity and richness. We suggest that the vegetation area of remnant patches within a radius of 500-2000 m should be appropriately increased to protect plant diversity, and the negative effects of road disturbance should also be considered.
Asunto(s)
Biodiversidad , Ecosistema , Plantas , CiudadesRESUMEN
Matrix and tensor completion aim to recover the incomplete two- and higher-dimensional observations using the low-rank property. Conventional techniques usually minimize the convex surrogate of rank (such as the nuclear norm), which, however, leads to the suboptimal solution for the low-rank recovery. In this paper, we propose a new definition of matrix/tensor logarithmic norm to induce a sparsity-driven surrogate for rank. More importantly, the factor matrix/tensor norm surrogate theorems are derived, which are capable of factoring the norm of large-scale matrix/tensor into those of small-scale matrices/tensors equivalently. Based upon surrogate theorems, we propose two new algorithms called Logarithmic norm Regularized Matrix Factorization (LRMF) and Logarithmic norm Regularized Tensor Factorization (LRTF). These two algorithms incorporate the logarithmic norm regularization with the matrix/tensor factorization and hence achieve more accurate low-rank approximation and high computational efficiency. The resulting optimization problems are solved using the framework of alternating minimization with the proof of convergence. Simulation results on both synthetic and real-world data demonstrate the superior performance of the proposed LRMF and LRTF algorithms over the state-of-the-art algorithms in terms of accuracy and efficiency.
RESUMEN
Tensor robust principal component analysis via tensor nuclear norm (TNN) minimization has been recently proposed to recover the low-rank tensor corrupted with sparse noise/outliers. TNN is demonstrated to be a convex surrogate of rank. However, it tends to over-penalize large singular values and thus usually results in biased solutions. To handle this issue, we propose a new definition of tensor logarithmic norm (TLN) as the nonconvex surrogate of rank, which can decrease the penalization on larger singular values and increase that on smaller ones simultaneously to preserve the low-rank structure of a tensor. Then, the strategy of tensor factorization is combined into the minimization of TLN to improve computational performance. To handle impulsive scenarios, we propose a nonconvex 'p-ball projection scheme with 0 < p < 1 instead of the conventional convex scheme with p = 1, which enhances the robustness against outliers. By incorporating the TLN minimization and the 'p-ball projection, we finally propose two low-rank recovery algorithms, whose resulting optimization problems are efficiently solved by the alternating direction method of multipliers (ADMM) with convergence guarantees. The proposed algorithms are applied to the synthetic data recovery and image and video restorations in real-world. Experimental results demonstrate the superior performance of the proposed methods over several state-ofthe- art algorithms in terms of tensor recovery accuracy and computational efficiency.
RESUMEN
A clear image of an observed object may deteriorate into unrecognizable speckle when encountering heterogeneous scattering media, thus it is necessary to recover the object image from the speckle. A method combining least square and semidefinite programming is proposed, which can be used for imaging through scattering media. The proposed method consists of two main stages, that is, media scattering characteristics (SCs) estimation and image reconstruction. SCs estimation is accomplished through LS concept after establishing a database of known object-and-speckle pairs. Image reconstruction is realized by solving an SDP problem to obtain the product of the unknown object image and its Hermitian transposition. Finally, the unknown object image can be reconstructed by extracting the largest rank-1 component of the product. Structural similarity (SSIM) index is employed as a performance indicator in speckle prediction and image reconstruction. Numerical simulations and physical experiments are performed to verify the feasibility and practicality of the proposed method. Compared with the existing phase shift interferometry mean square optimization method and the single-shot phase retrieval algorithm, the proposed method is the most precise to obtain the best reconstruction results with highest SSIM index value. The work can be used for exploring the potential applications of scattering media, especially for imaging through turbid media in biomedical, scattering property measurement, and optical image encryption.
Asunto(s)
Procesamiento de Imagen Asistido por Computador/métodos , Imagen Óptica/métodos , Algoritmos , Dispersión de RadiaciónRESUMEN
Recently, amorphous InGaZnO ultraviolet photo thin-film transistors have exhibited great potential for application in future display technologies. Nevertheless, the transmittance of amorphous InGaZnO (~80%) is still not high enough, resulting in the relatively large sacrifice of aperture ratio for each sensor pixel. In this work, the ultraviolet photo thin-film transistor based on amorphous InGaMgO, which processes a larger bandgap and higher transmission compared to amorphous InGaZnO, was proposed and investigated. Furthermore, the effects of post-deposition annealing in oxygen on both the material and ultraviolet detection characteristics of amorphous InGaMgO were also comprehensively studied. It was found that oxygen post-deposition annealing can effectively reduce oxygen vacancies, leading to an optimized device performance, including lower dark current, higher sensitivity, and larger responsivity. We attributed it to the combined effect of the reduction in donor states and recombination centers, both of which are related to oxygen vacancies. As a result, the 240-min annealed device exhibited the lowest dark current of 1.7 × 10-10 A, the highest photosensitivity of 3.9 × 106, and the largest responsivity of 1.5 × 104 A/W. Therefore, our findings have revealed that amorphous InGaMgO photo thin-film transistors are a very promising alternative for UV detection, especially for application in touch-free interactive displays.