Search

Concrete architectural structures are widely used in urban construction, making the health diagnosis and maintenance of these structures increasingly essential and urgent. Crack identification is crucial for maintaining the structural integrity and safety of buildings. Traditional methods rely on manual inspection, which is plagued by low accuracy, inefficiency, and safety hazards. This paper proposes a technique combining an attention-based SqueezeNet network with Gradient-weighted Class Activation Mapping (Grad-CAM) for automatically recognizing and visually explaining building cracks. By integrating the Squeeze-and-Excitation (SE) attention mechanism with the lightweight SqueezeNet network, this method can adaptively adjust the importance of feature channels by learning global information, effectively improving the network’s accuracy and efficiency. The experimental results show that the Att-SqueezeNet model achieved a high precision of 0.995, a training time of only 133 seconds, and a model size of 4.9M, significantly outperforming models such as SqueezeNet, RF, CNN, VGG-19, and B-CNN. This demonstrates its robustness, rapid identification and suitability for practical applications and building crack identification. Moreover, the utilization of Grad-CAM for visualization not only offers an intuitive explanation of the model’s decision-making process but also provides a more comprehensible understanding of crack detection results. This is crucial for advancing building maintenance automation, reducing reliance on manual labor, and increasing the precision and reliability of detection tasks.

In the modern field of multi-view 3D detection, interest has seen a notable rise. Existing methods are predominantly focused on constructing dense BEV (Bird’s Eye View) features or utilizing sparse queries for detection. In this paper, we synergize the methodologies of sparse queries with dense BEV feature representation and integrate robust 2D detection capabilities to propose an innovative structure. Specifically, we leverage the results of 2D detection and their predicted depth to generate a series of sparse queries endowed with strong semantic and positional priors. These queries demonstrate enhanced capability in capturing challenging cases in 3D detection. Additionally, we employ 2D detection and depth information to discern critical foreground BEV queries. This strategy enables a efficient and noiseless feature aggregation, leveraging multi-view image features. Our methodology extends to a comprehensive global modeling of all BEV queries, ensuring the derivation of premium quality BEV features. Furthermore, we enhance BEV features through temporal modeling that differentiates between static and dynamic objects. Our proposed approach achieves state-of-the-art results on the nuScenes test set with 60.4%mAP and 51.7%NDS. Code will be available.

Stereo matching is an important component technology that constitutes the 3D perception capability of autonomous vehicles. On resource-constrained edge devices, it is very important to compute in real-time with very low time. However, most stereo matching networks focus on generating disparity maps on high-end GPUs, which do not meet the real-time requirements on edge devices. To solve this problem, a new stereo matching network is proposed in this paper to achieve real-time stereo matching on edge devices. The proposed network greatly improves the inference speed by constructing a low-resolution feature extractor, and by using multi-stage residual methods for stereo matching. In particular, we propose a method that combines the group-wise L1 distance & the group-wise correlation cost volume and an effective attention-based 3D cost aggregation method. Our network achieves a good balance between speed and accuracy on the KITTI 2012 and KITTI 2015 datasets. The proposed network achieves 2.77% and 3.44% accuracy (D1-all) on KITTI 2012 and KITTI 2015, respectively. With TensorRT, the proposed network achieves 31.8 FPS and outperforms the real-time results of most state-of-the-art networks on NVIDIA Jetson Nano edge devices.

Although the channel state information (CSI) is never perfect and varies in time, most studies on spatial modulation (SM) systems assume perfectly known CSI and time-invariant channel. The spatial correlations among multiple spatial subchannels, which have to be considered when CSI is imperfect, are also often neglected. In this paper, we release the above assumptions and take the CSI uncertainty along with the spatial-temporal selectivities into account. We derive the channel estimation error aware maximum likelihood (CEEA-ML) detectors as well as several low-complexity alternatives for PSK and QAM signals. As the CSI uncertainty depends on the channel estimator (CE) used, we consider two well-known CEs in our study. The error rate performance of the ML and some suboptimal detectors is analyzed. Numerical results obtained by simulations and analysis show that the CEEA-ML detectors offer clear performance gain against conventional mismatched SM detectors and, in many cases, the proposed suboptimal detectors incur only minor performance loss.

Modern multi-color RGB micro-light emitting diode (μLED) displays and digital micro-mirror laser projectors often require the use of both III-V and III-Nitride material systems for different pixel/laser colors. This is due primarily to the conventionally low efficiencies of red emitters based on the InGaN materials system, which is used to create green and blue emitters for RGB displays. The main challenges for InGaN red emitters are the quantum confined stark effect (QCSE) and the difficulty of incorporating high In-content into the active region. In this work, InGaN/InGaN/delta-InN quantum wells (QWs) on InGaN substrates are proposed and demonstrated to show significant enhancement in electron-hole wavefunction overlap (Te_hh) and spontaneous emission radiative recombination rate (Rsp) in the red emission regime. Analysis of InGaN/InGaN/delta-InN QWs with InGaN barriers emitting at 630 nm was performed using a self-consistent six-band k · p formalism. The Γe_hh was shown to increase by more than 230% compared to an InGaN/InGaN QW emitting at 630 nm, leading to significant increases in Rsp and internal quantum efficiency (ηIQE). With growth of InN monolayers on InGaN now readily achievable, this novel active region design could pave the way for high-efficiency, native red-emitting InGaN LEDs and lasers.

Point-of-care (POC) testing has demonstrated great transformative potential in personalized medicine. In particular, patients undergoing transplantation require POC testing to ensure appropriate serum immunosuppressant levels so as to maintain adequate graft function and survival. However, no suitable POC device for monitoring immunosuppressant levels is currently available. Exploiting the latest advances in metamaterials can lead to a breakthrough in POC testing. A high-contrast grating (HCG) biosensor is a low-cost, compact, simple-to-fabricate, and easy-to-operate structure. It is highly sensitive and robust in surface-based biomarker detection, which is favorable for the efficiency of a POC device. In this study, the feasibility of using an HCG as a POC sensor for therapeutic drug monitoring of immunosuppressants was evaluated. The detection efficiency of the most commonly prescribed immunosuppressive medication cyclosporine A by using this sensor was demonstrated to be comparable to those of conventional commercial kits, suggesting that the sensor has the potential to be used as a rapid detection and feedback platform for increasing drug compliance and improving new organ transplant survival.

In this paper, a phase retrieval method is proposed to reconstruct the high-quality image free from undesired terms in on-axis digital holography (DH) with modified coherent diffraction imaging (MCDI). A random phase plate (RPP) is applied to generate a modulated diffraction pattern which forms an on-axis digital hologram with a plane reference wave, and a proper iterative algorithm is designed to reconstruct high-quality object information from both on-axis digital hologram and modulated diffraction pattern. Numerical simulations and experiments prove that both modulus and phase distributions can be accurately reconstructed within 100 iterations while the undesired twin image and zero-order diffraction can be effectively removed, thus demonstrating the feasibility of the suggested method.

Cloud servers are very suitable for sharing, and storing image data, and meanwhile it is confronted with many security, and privacy problems. Optical chaos, and hardware implementation of image encryption possess distinct superiorities in secret image sharing. Hence, combining the advantages of advanced reduced instruction set computer machine (ARM)-embedded platform Exynos4412 model, together with vertical-cavity surface emitting laser (VCSEL) subject to positive optoelectronic feedback (POEF), a novel color image encryption, and transmission system is experimentally realized, and appraised. We investigate the dynamic behaviors of the VCSEL, and obtain the chaotic output. An analog to digital conversion (ADC) module is utilized for optical chaos data acquisition. The improved gravity model, and the double sine map are exploited to encrypt the image, and the encrypted image is securely shared by cloud services. The user can download the encrypted image, and decrypt it by the hardware board. The experiment results reveal that the system has an excellent resistance to common attacks. Compared to the numerical simulation, our hardware implementation for such a color image encryption, and transmission system is formidable to actualize, but we successfully implemented, and tested in a real-world environment. By comparing experimental, and simulation results, the consistent results are found. Therefore, in the case of similar encryption effect, the user's information privacy protection in the actual industrial or commercial systems can be securely guaranteed on our hardware platform.

Voice coil motor (VCM) is widely used in high precision position servo control for its merits of high linearity and no cogging-torque. There are two conventional driving modes: PWM chopper drive and analog drive. PWM chopper drive usually results in current ripples, which could affect position control precision. Acceptable current ripple limits the increase in the DC bus voltage and could not achieve faster response. On the other hand, analog drive uses power amplifiers, which does not suffer from current ripple but can result in high power loss. Such power loss could also affect precision or result in cooling difficulty, especially for thermal-sensitive applications such as semi-conductor lithography. In this paper, a VCM chopping drive with an LCL filter together with filter design method is proposed. Unlike conventional LCL filter design for grid inverter that only focuses on steady state current quality, the proposed LCL filter design in this paper aims at transient position control, which considers the current ripple, transient position tracking time and additional volume. With the designed filter, position tracking time can be shortened with acceptable current ripple. To compensate the delay caused by LCL filter, the DC drive voltage is increased. And under the same current ripple level, with LCL filter and higher DC voltage can get faster position response speed than without LCL filter and lower voltage. The results are validated by simulation and experimental results.

Phase-change materials, also well known as the Chalcogenide alloy, have received considerable attention during last two decades owing to its widespread applications in the field of the electrical storage market such as phase-change random access memory and phase-change probe memory. In addition to the storage devices, its unique electrical properties that can be dynamically tunable with respect to the electrical excitations lead to numerous novel applications represented by memristor and memristor-based neuromorphic electrical circuits. These emerging applications undoubtedly allows for a further exploitation of the potential of phase-change materials, and thus makes it advantageous over other storage medium like ferroelectric and magnetic materials. In order to help researchers understand the role of phase-change materials in these novel applications as well as their importance for citizen's daily life, a comprehensive review that not only covers the traditional storage applications, but also the applications on these exotic devices becomes imperative. In this review, the chemical structure of phase-change materials and their remarkable electrical properties are first reviewed, followed by an introduction of their applications on the storage fields. The physical principles of various emerging electrical devices using phase-change materials are subsequently overviewed in association with their state-of-the-art progress. The prospect of phase-change materials for the future non-volatile electrical applications that are yet to be unraveled is finally envisaged.

The main dielectric medium in converter transformers is the oil-pressboard. The valve windings of converter transformers are subjected to AC-DC combined voltage. Creepage discharge is a common defect of oil-pressboard insulation. The degree of the oil-immersed pressboard (OIP) damage is directly affected by different creepage discharge paths. In this study, a needle-plate model was developed to explore different creepage discharge paths in oil-pressboard insulation under AC-DC combined voltage. A high-speed camera was used to record the developmental process of white marks on the OIP. We explored reasons behind the different creepage discharge paths under different voltage types. Results revealed that the creepage discharge path and the degree of OIP damage were influenced by the DC component. Although the damaged OIP in converter transformers may cause catastrophic flashover inside the transformers, it is difficult to replace it. This implies that transformer insulation designers should pay attention to the influence of the DC component on OIP damage. In this study, we used pulse current method to detect discharge patterns. Discharge parameters were then extracted from discharge patterns. We observed that developmental process of white marks corresponded to developmental process of discharge parameters. This suggests that creepage discharges can be diagnosed based on discharge parameters. Based on these conclusions, important information was provided for the on-line monitoring design of transformer insulation status.

In image processing and computer vision, image segmentation plays a fundamental role since it can make images easier to analyze. However, noise is easily introduced into images and brings great challenges to image segmentation. This paper focuses on the segmentation problem of noisy images and proposes an efficient variational level set model based on adaptive local fitted image to handle it. By utilizing normalized local entropy and local means, an adaptive local fitted image is proposed and introduced into the data term to enhance the robustness of the model against noise. Then a penalty term is proposed to reduce the deviation of the adaptive local fitted image from the original image by punishing the difference between them, so as to guarantee the accuracy of segmentation results. Later, the total variational regularization term is introduced into the model, so the level set function can be smoothed and the effect of noise on the active contour can be further reduced. The energy functional of the whole model is convex rigorously, which can reach the minimum and should have good properties in noisy image segmentation. Numerous experiments on synthetic, natural, synthetic aperture radar and oil spill images demonstrate that the proposed model is strongly robust to different types and levels of noise, which indicates its good performance in noisy image segmentation.

Although semi-supervised clustering ensemble methods have achieved satisfactory performance, they fail to effectively utilize the constrained knowledge such as cannot-link and must-link when generating diverse ensemble members. In addition, they ignore negative effects brought about by redundancies and noisy data. To address the above shortcomings, in this paper we propose an approach to combine multiple semi-supervised clustering solutions via adaptively regularizing the weights of clustering ensemble members, which is referred to as ARSCE. First, we generate a series of feature subspaces by randomly selecting feature without replacement to avoid the scenario where there are two identical feature subspaces. Second, we conduct feature transformation on the above obtained feature subspaces while considering the pairwise constraints to find new clustering-friendly spaces, where clustering methods are exploited to generate various clustering solutions. Finally, we design a novel fusion strategy to integrate multiple clustering solutions into a unified clustering partition, where weights are designated for each clustering ensemble member. Extensive experiments are conducted on multiple real-world benchmarks, and experimental results demonstrate the effectiveness and superiority of our proposed method ARSCE over other counterparts.

In order to provide high transmission efficiency and visual quality for the color image/video in internet, effective estimation of its inherent just noticeable distortion (JND) is continuously an important topic. In this paper, a discrete cosine transform (DCT)-based perceptual model for estimating spatio-temporal JNDs of color videos is presented. Firstly, the compound masking adjustment integrated with different spatial masking factors is proposed. Based on the base detection threshold for DCT coefficients in luminance and chrominance components of the color image, the adjustment combined by luminance masking, pattern-based contrast masking, and cross masking is exploited to measure visibility thresholds for luminance component, while the adjustment induced by the variance-based statistical properties is utilized to measure visibility thresholds for chrominance components. Then, the local temporal statistics of luminance component are considered to design the structure-based temporal masking adjustment for further estimating visibility thresholds of video signals. To verify the proposed method, a subjective viewing test is designed and a fair rating process of evaluating the visual quality is carried out under the specified viewing condition. Experimental results show that the proposed method is able to measure more visual distortion tolerance than the existing color-based methods at high visual quality.

Based on the first-order Taylor expansion, an efficient Rigorous Coupled-Wave Analysis (RCWA) for one-dimensional ultrathin periodic structures is proposed in this paper. The derivation of the ultrathin form RCWA method is completed by using the first-order Taylor expansion to rearrange the matrix in the equations of boundary conditions. Then, the reliability of the proposed algorithm is verified by two examples. Finally, it is concluded that the proposed algorithm can reduce the CPU time of TE polarization and TM polarization by more than 50%. Meanwhile, compared with the conventional algorithm, the proposed algorithm also needs less memory.

As a non-interferometric method, the transport of intensity equation (TIE) method suits for quantitative phase imaging with the commercial microscope platform, especially those higher order TIE approaches which can realize precise phase retrieval by eliminating undesired higher order derivatives are widely used. However, these approaches are mostly adopted separately without considering their phase retrieval precision in various noise levels. In this paper, we first compared these classical higher order TIE approaches through theoretical analysis, numerical simulations, and experiments. Then, based on the quantitative comparisons mainly focusing on their phase retrieval accuracy and noise suppression capability, we determine the application scope corresponding to different noise levels of each higher order TIE approach. Finally, in order to deal with different noisy cases, we design the hybrid higher order TIE application, in which the specific higher order TIE approach is selected for phase retrieval according to the precise noise estimation, and the performance of the hybrid higher order TIE application is tested by both the numerical simulations and the experiments, proving it can perform high-quality phase imaging by balancing the tradeoff between the phase retrieval accuracy and the noise influence. The paper not only provides a systematic reference for analysis and comparisons on different higher order TIE approaches, but also proposes the hybrid application for noise adaptive phase imaging, which can be a potential tool in biological observations and medical diagnostics.

A new method for optical multiple-image encryption based on polarization-resolved diffraction imaging is proposed in this study. A linearly polarized beam is split into two beams of orthogonal polarization to illuminate different images to be encrypted, and these images are then encrypted into a single white-noise-like intensity image in a single shot. A multi-mode iterative phase-retrieval algorithm is used in the decryption to reconstruct the amplitudes and phases of the secret images simultaneously. Both computer simulations and optical experiments are performed to demonstrate the feasibility and effectiveness of the proposed method.

The development of mobile computing systems is an inherently collaborative activity, and the nature of the collaboration likely affects systems success. Collaboration can lead to faster development of systems and increased creativity, but it may also adversely affect the quality of systems. As one of the most popular and active Web services for developing open source software, Github has been a place of origin of many novel mobile computing systems with the Internet of Things, such as OpenEdge and CloudSim. Meanwhile, Github provides a place for people to collaborate with each other on developing novel mobile service computing systems. In this paper, we explore the effects of collaboration in developing systems hosted by Github and use a model of collaboration to detect bugs of systems aiming to enhance their quality. We conduct an at-scale study of more than five million projects on GitHub to reveal collaboration's effects on the popularity of software systems and speed of development. We find that having more collaborators results in higher popularity but slower development. Projects that use Objective-C are at an even more significant risk for lower development speeds than other projects. We define collaboration networks and co-commit networks, which capture the projects' collaboration patterns and use structural features of these networks to identify files within commits with bugs. By examining 18 large popular projects, we find that involving these features statistically significantly improves bug detection by 6.7% on average.

To get better control performance in motor control, more and more researches tend to apply non-linear control laws in the field of motor control. However, most conventional non-linear control theory is based on explicit model of controlled object and often resulting in complexity. Besides, the control parameters tuning is mainly aiming at stability of the system. No valid direct performance-oriented non-linear control theory has been proposed. Facing the limitations, this paper presents a direct motor position control in an implicit data-driven manner. Unlike conventional non-linear motor controls that are based on explicit models and with stability-based parameters tuning, this study gives performance-oriented non-linear control by mastering non-linear discrete optimal control law in an implicit data-learning manner. Firstly, optimal data of position tracking problem is obtained by solving optimization problem. Secondly, the implicit discrete optimal control law hidden in data is learned by a BP neural network. Finally, the learned control law is implemented in real-time control to reproduce optimal control performance. Simulation and experiment results validated the feasibility of the data-driven controller, which could be helpful for performance-oriented non-linear control designs. The merits and further improvements are also discussed.

Ensuring multimedia content security is one of the main targets to solve multimedia information security problems and k-anonymous model is a classic and effective model to protect personal information when we publish multimedia data. The k-anonymous security with different k values has always been a hot topic of research, and there is no specific quantitative method for k-anonymous model security without considering any application scenarios or assumptions. In this paper, we give a quantitative method for the capability of attacker and defender under k-anonymous model. More specifically, we present a new general notion of limitation to provide a precise bound of attack and defense capability in a k-anonymous model. Firstly, we establish the attack channel and the defense channel from the point of the attacker's and the defender's view respectively. Moreover, we describe the capability of both attacking and defending sides by introducing the average mutual information, as well as the maximum value of the the average mutual information is the limitation of the capability of both sides, i.e. the channel capacity. Finally, according to the variation curve of the average mutual information amount, we compare the capability of the attacker and defender and analyze the effect of k value on it and get the best interval of k, that is, the value of k should be greater than 8 and less than 40.

Icicle will cause serious distortion of insulator electric field, weaken its voltage strength, and easily cause insulator flashover. Previous studies have shown that the longer the length of icicles on insulator, the greater the possibility of electric flashover. Then, it is of great theoretical significance and practical value to study the image identification technology of icicle length of iced insulators. Firstly, the growth mechanism and structural characteristics of insulator icicles are discussed in this paper, and the image acquisition method of iced insulator is discussed. Secondly, the methods of saliency detection and color feature analysis are used to extract the region of icicle and insulator sheds respectively, then, the icicle structure on insulator is obtained by the difference calculation of above two images. Meanwhile, the pixel curve of the iced insulator image is processed by Fourier transform, and the reference size of insulators icicle is obtained by calculating the length of the shed spacing. Finally, a practical case about iced insulators in substations is analyzed in detail, which proves the effectiveness of the method proposed in this paper.

According to the DNA encryption algorithm and the double-chaotic system which contains the optical chaos and the coupled map lattice chaotic system, a novel image encryption-then-transmission system is proposed. In the system, with identical chaotic injection from a master laser with two optical feedbacks, two slave lasers (SL1 and SL2) can output similar chaotic signals served as chaotic carrier to transmit image and used to generate the core part of the encryption scheme. A 128-b key is selected to generate the original value of the double-chaotic system, which decides the DNA complementary rule, hence, the key is hypersensitive in encryption and decryption process. The security analysis demonstrates the effectiveness of the proposed encryption system. The simulation results verify that the cryptosystem is enough against the traditional attacks, such as statistical attack, differential attack, brute force attack, and entropy attack. Moreover, the encrypted image can be the optical message and transmitted in 10 km single-mode fiber channel from SL1 to SL2. In order to ensure the security, we use the chaos masking technique to modulate and demodulate the optical message. Through numerical simulations of the cross-correlation function, the chaos synchronization between SL1 and SL2 is desired. The Q-factor is 9.559 and the bit error rate is 5.771 × 10-22.

A multiple Fano resonance based optical refractive index sensor composed of micro-cavity and micro-structure is designed. The multiple Fano resonances, generated by the interference between the multiple resonant peaks produced by the micro-cavity and the continuous background produced by the micro-structure, can reach high sensitivity in refractive index measurement. Besides, the micro-cavity configuration obviously reduces the energy loss, thus effectively increasing the figure of merit. Adjusted by the micro-cavity thickness and the micro-structure configuration, multiple Fano resonances can be generated potential for refractive index sensing: the sensitivity of 831 nm/RIU and the figure of merit of ~600 can be reached according to a single Fano resonance; moreover, spectral interval between double Fano resonances can also be used for refractive index sensing with the sensitivity of 194 nm/RIU but significantly reducing the error due to the environmental changes. Considering the advantages as high sensitivity, narrow Fano resonant peak, high figure of merit and ease of integration, as well as its potential in environmental error reduction, the proposed multiple Fano resonance based optical refractive index sensor is expected to provide new tactics for optical refractive index sensing.

Condition-based maintenance is an effective method for deciding when to maintain deteriorating system. In this paper, a non-periodic maintenance model with adaptive inspection intervals is proposed by the consideration of system stability and deterioration. Different maintenance plans and monitoring strategies are adopted in distinct stages of life cycle of the deteriorating system to reduce costs, which is a complex multiple optimization parameter problem. And particle swarm optimization algorithm combining heuristic rules is designed to solve this multi-objective problem. Finally, a numerical example is implemented to illustrate the effectiveness and rationality of the proposed model. The comparison between the proposed model and other maintenance models illustrates the economy of the proposed model and the importance of considering system stability. Sensitivity analysis is also performed to investigate the effect of seven cost factors.

This paper investigates the polarization-dependent light extraction efficiency (ηextraction) and the effect of passivation layer for AlGaN-based nanowire (NW) deep-ultraviolet (DUV) light-emitting diodes (LEDs) emitted at 230 nm using three-dimensional finite-difference time-domain method. Our results show that the use of NW structure for 230-nm LEDs can result in higher transverse-magnetic (TM) polarized ηextraction than transverseelectric (TE) polarized ηextraction. Specifically, ηextraction up to ~48% for TM polarization and ~41% for TE polarization can be achieved by the investigated 230-nm NW LEDs as compared to conventional planar LEDs (~0.2% and ~2% for TM- and TE-polarizations, respectively) attributed to the large surface-to-volume ratio and strong sidewall emissions. In addition, the analysis on the effect of various passivation layer materials suggests the use of SiO2, which has smaller refractive index than the NW core, could extract more photons out of the NW core and lead to TM-polarized ηextraction up to ~47%. Therefore, the use of the AlGaN NW structure is expected to lead to high external quantum efficiency DUV LEDs due to the large TM-polarized ηextraction in combination with the dominant TM-polarized spontaneous emissions at 230 nm.

To realize active light control with a compact structure, a nanoscale optical switch-based plasmonic demultiplexer is proposed. The device is designed according to the metal-Kerr nonlinear material-metal waveguide with multiple resonant cavities, providing the capability of active optical regulation for dual wavelengths: Signals of 1310 and 1550 nm can be guided into either output with high directional extinction ratios larger than 30 dB controlled by another pump light of 730 nm with low power. Considering its light control capability and compact configuration, it is believed that the proposed demultiplexer can be adopted in the future logic plasmonic circuits.