Search

With the rapid growth of the PCB manufacturing industry, there is an increasing demand for computer vision inspection to detect defects during production. Improving the accuracy and generalization of PCB defect detection models remains a significant challenge. This paper proposes a high-precision, robust, and real-time end-to-end method for PCB defect detection based on deep Convolutional Neural Networks (CNN). Traditional methods often suffer from low accuracy and limited applicability. We propose a novel approach combining YOLOv5 and multiscale modules for hierarchical residual-like connections. In PCB defect detection, noise can confuse the background and small targets. The YOLOv5 model provides a strong foundation with its real-time processing and accurate object detection capabilities. The multi-scale module extends traditional approaches by incorporating hierarchical residual-like connections within a single block, enabling multiscale feature extraction. This plug-and-play module significantly enhances performance by extracting features at multiple scales and levels, which are useful for identifying defects of varying sizes and complexities. Our multi-scale architecture integrates feature extraction, defect localization, and classification into a unified network. Experiments on a large-scale PCB dataset demonstrate significant improvements in precision, recall, and F1-score compared to existing methods. This work advances computer vision inspection for PCB defect detection, providing a reliable solution for high-precision, robust, real-time, and domain-adaptive defect detection in the PCB manufacturing industry.

Functionalization and volatilization are competing reactions during the oxidation of carbonaceous materials and are important processes in many different areas of science and technology. Here we present a combined ambient pressure X-ray photoelectron spectroscopy (APXPS) and grazing incidence X-ray scattering (GIXS) investigation of the oxidation of oleic acid ligands surrounding NaYF4 nanoparticles (NPs) deposited onto SiOx/Si substrates. While APXPS monitors the evolution of the oxidation products, GIXS provides insight into the morphology of the ligands and particles before and after the oxidation. Our investigation shows that the oxidation of the oleic acid ligands proceeds at O2 partial pressures of below 1 mbar in the presence of X-rays, with the oxidation eventually reaching a steady state in which mainly CHx and -COOH functional groups are observed. The scattering data reveal that the oxidation and volatilization reaction proceeds preferentially on the side of the particle facing the gas phase, leading to the formation of a chemically and morphologically asymmetric ligand layer. This comprehensive picture of the oxidation process could only be obtained by combining the X-ray scattering and APXPS data. The investigation presented here lays the foundation for further studies of the stability of NP layers in the presence of reactive trace gasses and ionizing radiation, and for other nanoscale systems where chemical and morphological changes happen simultaneously and cannot be understood in isolation., Comment: 18 pages, 8 figures

Mechanical force is an essential feature for many physical and biological processes.1-12 Remote measurement of mechanical signals with high sensitivity and spatial resolution is needed for diverse applications, including robotics,13 biophysics,14-20 energy storage,21-24 and medicine.25-27 Nanoscale luminescent force sensors excel at measuring piconewton forces,28-32 while larger sensors have proven powerful in probing micronewton forces.33,34 However, large gaps remain in the force magnitudes that can be probed remotely from subsurface or interfacial sites, and no individual, non-invasive sensor is capable of measuring over the large dynamic range needed to understand many systems.35,36 Here, we demonstrate Tm3+-doped avalanching nanoparticle37 force sensors that can be addressed remotely by deeply penetrating near-infrared (NIR) light and can detect piconewton to micronewton forces with a dynamic range spanning more than four orders of magnitude. Using atomic force microscopy coupled with single-nanoparticle optical spectroscopy, we characterize the mechanical sensitivity of the photon avalanching process and reveal its exceptional force responsiveness. By manipulating the Tm3+ concentrations and energy transfer within the nanosensors, we demonstrate different optical force-sensing modalities, including mechanobrightening and mechanochromism. The adaptability of these nanoscale optical force sensors, along with their multiscale sensing capability, enable operation in the dynamic and versatile environments present in real-world, complex structures spanning biological organisms to nanoelectromechanical systems (NEMS).

The optical Ramsey technique is used to obtain precise measurements of the hyperfine splittings in the $2\,^3\!S_1$ and $2\,^3\!P_J$ states of $^7$Li$^+$. Together with bound-state quantum electrodynamic theory, the Zemach radius and quadrupole moment of the $^7$Li nucleus are determined to be $3.35(1)$~fm and $-3.86(5)$~fm$^2$ respectively, with the quadrupole moment deviating from the recommended value of $-4.00(3)$~fm$^2$ by $1.75\sigma$. Furthermore, we determine the quadrupole moment ratio of $^6$Li to $^7$Li as $0.101(13)$, exhibiting a $6\sigma$ deviation from the previous measured value of $0.020161(13)$ by LiF molecular spectroscopy. The results taken together provide a sensitive test of nuclear structure models.

Optically bistable materials respond to a single input with two possible optical outputs, contingent upon excitation history. Such materials would be ideal for optical switching and memory, yet limited understanding of intrinsic optical bistability (IOB) prevents development of nanoscale IOB materials suitable for devices. Here, we demonstrate IOB in Nd3+-doped KPb2Cl5 avalanching nanoparticles (ANPs), which switch with high contrast between luminescent and non-luminescent states, with hysteresis characteristic of bistability. We elucidate a nonthermal mechanism in which IOB originates from suppressed nonradiative relaxation in Nd3+ ions and from the positive feedback of photon avalanching, resulting in extreme, >200th-order optical nonlinearities. Modulation of laser pulsing tunes hysteresis widths, and dual-laser excitation enables transistor-like optical switching. This control over nanoscale IOB establishes ANPs for photonic devices in which light is used to manipulate light.

We investigate forward signal propagation and gradient back propagation in deep, randomly initialized transformers, yielding simple necessary and sufficient conditions on initialization hyperparameters that ensure trainability of deep transformers. Our approach treats the evolution of the representations of $n$ tokens as they propagate through the transformer layers in terms of a discrete time dynamical system of $n$ interacting particles. We derive simple update equations for the evolving geometry of this particle system, starting from a permutation symmetric simplex. Our update equations show that without MLP layers, this system will collapse to a line, consistent with prior work on rank collapse in transformers. However, unlike prior work, our evolution equations can quantitatively track particle geometry in the additional presence of nonlinear MLP layers, and it reveals an order-chaos phase transition as a function of initialization hyperparameters, like the strength of attentional and MLP residual connections and weight variances. In the ordered phase the particles are attractive and collapse to a line, while in the chaotic phase the particles are repulsive and converge to a regular $n$-simplex. We analytically derive two Lyapunov exponents: an angle exponent that governs departures from the edge of chaos in this particle system, and a gradient exponent that governs the rate of exponential growth or decay of backpropagated gradients. We show through experiments that, remarkably, the final test loss at the end of training is well predicted just by these two exponents at the beginning of training, and that the simultaneous vanishing of these two exponents yields a simple necessary and sufficient condition to achieve minimal test loss.

The mutual information characterizes correlations between spatially separated regions of a system. Yet, in experiments we often measure dynamical correlations, which involve probing operators that are also separated in time. Here, we introduce a space-time generalization of mutual information which, by construction, satisfies several natural properties of the mutual information and at the same time characterizes correlations across subsystems that are separated in time. In particular, this quantity, that we call the \emph{space-time mutual information}, bounds all dynamical correlations. We construct this quantity based on the idea of the quantum hypothesis testing. As a by-product, our definition provides a transparent interpretation in terms of an experimentally accessible setup. We draw connections with other notions in quantum information theory, such as quantum channel discrimination. Finally, we study the behavior of the space-time mutual information in several settings and contrast its long-time behavior in many-body localizing and thermalizing systems., Comment: 33 pages, 12 figures

We construct a class of nonlinear coherent states (NLCSs) by introducing a more general nonlinear function and study their non-classical properties, specifically the second-order correlation function $g^{(2)}(0)$, Mandel parameter $Q$, squeezing, amplitude squared squeezing and Wigner function of the optical field. The results indicate that the non-classical properties of the new types of even and odd NLCSs crucially depend on nonlinear functions. More concretely, we find that the new even NLCSs could exhibit the photon-bunching effect whereas the new odd NLCSs could show photon-antibunching effect. The degree of squeezing is also significantly affected by the parameter selection of these NLCSs. By employing various forms of nonlinear functions, it becomes possible to construct NLCSs with diverse properties, thereby providing a theoretical foundation for corresponding experimental investigations.

The integration of Autonomous Vehicles (AVs) into existing human-driven traffic systems poses considerable challenges, especially within environments where human and machine interactions are frequent and complex, such as at unsignalized intersections. To deal with these challenges, we introduce a novel framework predicated on dynamic and socially-aware decision-making game theory to augment the social decision-making prowess of AVs in mixed driving environments. This comprehensive framework is delineated into three primary modules: Interaction Orientation Identification, Mixed-Strategy Game Modeling, and Expert Mode Learning. We introduce 'Interaction Orientation' as a metric to evaluate the social decision-making tendencies of various agents, incorporating both environmental factors and trajectory characteristics. The mixed-strategy game model developed as part of this framework considers the evolution of future traffic scenarios and includes a utility function that balances safety, operational efficiency, and the unpredictability of environmental conditions. To adapt to real-world driving complexities, our framework utilizes a dynamic optimization framework for assimilating and learning from expert human driving strategies. These strategies are compiled into a comprehensive strategy library, serving as a reference for future decision-making processes. The proposed approach is validated through extensive driving datasets and human-in-loop driving experiments, and the results demonstrate marked enhancements in decision timing and precision.

Autonomous driving technology is poised to transform transportation systems. However, achieving safe and accurate multi-task decision-making in complex scenarios, such as unsignalized intersections, remains a challenge for autonomous vehicles. This paper presents a novel approach to this issue with the development of a Multi-Task Decision-Making Generative Pre-trained Transformer (MTD-GPT) model. Leveraging the inherent strengths of reinforcement learning (RL) and the sophisticated sequence modeling capabilities of the Generative Pre-trained Transformer (GPT), the MTD-GPT model is designed to simultaneously manage multiple driving tasks, such as left turns, straight-ahead driving, and right turns at unsignalized intersections. We initially train a single-task RL expert model, sample expert data in the environment, and subsequently utilize a mixed multi-task dataset for offline GPT training. This approach abstracts the multi-task decision-making problem in autonomous driving as a sequence modeling task. The MTD-GPT model is trained and evaluated across several decision-making tasks, demonstrating performance that is either superior or comparable to that of state-of-the-art single-task decision-making models., Comment: Accepted by ITSC 2023

With the integration of Autonomous Vehicles (AVs) into our transportation systems, their harmonious coexistence with Human-driven Vehicles (HVs) in mixed traffic settings becomes a crucial focus of research. A vital component of this coexistence is the capability of AVs to mimic human-like interaction intentions within the traffic environment. To address this, we propose a novel framework for Unprotected left-turn trajectory planning for AVs, aiming to replicate human driving patterns and facilitate effective communication of social intent. Our framework comprises three stages: trajectory generation, evaluation, and selection. In the generation stage, we use real human-driving trajectory data to define constraints for an anticipated trajectory space, generating candidate motion trajectories that embody intent expression. The evaluation stage employs maximum entropy inverse reinforcement learning (ME-IRL) to assess human trajectory preferences, considering factors such as traffic efficiency, driving comfort, and interactive safety. In the selection stage, we apply a Boltzmann distribution-based method to assign rewards and probabilities to candidate trajectories, thereby facilitating human-like decision-making. We conduct validation of our proposed framework using a real trajectory dataset and perform a comparative analysis against several baseline methods. The results demonstrate the superior performance of our framework in terms of human-likeness, intent expression capability, and computational efficiency. Limited by the length of the text, more details of this research can be found at https://shorturl.at/jqu35

We introduce a general approach to realize quantum states with holographic entanglement structure via monitored dynamics. Starting from random unitary circuits in $1+1$ dimensions, we introduce measurements with a spatiotemporally-modulated density. Exploiting the known critical properties of the measurement-induced entanglement transition, this allows us to engineer arbitrary geometries for the bulk space (with a fixed topology). These geometries in turn control the entanglement structure of the boundary (output) state. We demonstrate our approach by giving concrete protocols for two geometries of interest in two dimensions: the hyperbolic half-plane and a spatial section of the BTZ black hole. We numerically verify signatures of the underlying entanglement geometry, including a direct imaging of entanglement wedges by using locally-entangled reference qubits. Our results provide a concrete platform for realizing geometric entanglement structures on near-term quantum simulators., Comment: 5 pages, 3 figures

Photon avalanching nanoparticles (ANPs) exhibit extremely nonlinear upconverted emission valuable for sub-diffraction imaging, nanoscale sensing, and optical computing. Avalanching has been demonstrated with Tm3+, Nd3+ or Pr3+-doped nanocrystals, but their emission is limited to 600 and 800 nm, restricting applications. Here, we utilize Gd3+-assisted energy migration to tune the emission wavelengths of Tm3+-sensitized ANPs and generate highly nonlinear emission of Eu3+, Tb3+, Ho3+, and Er3+ ions. The upconversion intensities of these spectrally discrete ANPs scale with the nonlinearity factor s = 10-17 under 1064 nm excitation at power densities as low as 6 kW/cm2. This strategy for imprinting avalanche behavior on remote emitters can be extended to fluorophores adjacent to ANPs, as we demonstrate with CdS/CdSe/CdS core/shell/shell quantum dots. ANPs with rationally designed energy transfer networks provide the means to transform conventional linear emitters into a highly nonlinear ones, expanding the use of photon avalanching in biological, chemical, and photonic applications., Comment: 13 pages, 5 figures

Under the global COVID-19 crisis, accurate diagnosis of COVID-19 from Chest X-ray (CXR) images is critical. To reduce intra- and inter-observer variability, during the radiological assessment, computer-aided diagnostic tools have been utilized to supplement medical decision-making and subsequent disease management. Computational methods with high accuracy and robustness are required for rapid triaging of patients and aiding radiologists in the interpretation of the collected data. In this study, we propose a novel multi-feature fusion network using parallel attention blocks to fuse the original CXR images and local-phase feature-enhanced CXR images at multi-scales. We examine our model on various COVID-19 datasets acquired from different organizations to assess the generalization ability. Our experiments demonstrate that our method achieves state-of-art performance and has improved generalization capability, which is crucial for widespread deployment., Comment: Accepted for publication at the Journal of Machine Learning for Biomedical Imaging (MELBA) https://melba-journal.org/2023:008

We investigate the $2\,^3\!S_1$--$2\,^3\!P_J$ ($J = 0, 1, 2$) transitions in $\rm^6Li^+$ using the optical Ramsey technique and achieve the most precise values of the hyperfine splittings of the $2\,^3\!S_1$ and $2\,^3\!P_J$ states, with smallest uncertainty of about 10~kHz. The present results reduce the uncertainties of previous experiments by a factor of 5 for the $2\,^3\!S_1$ state and a factor of 50 for the $2\,^3\!P_J$ states, and are in better agreement with theoretical values. Combining our measured hyperfine intervals of the $2\,^3\!S_1$ state with the latest quantum electrodynamic (QED) calculations, the improved Zemach radius of the $\rm^6Li$ nucleus is determined to be 2.44(2)~fm, with the uncertainty entirely due to the uncalculated QED effects of order $m\alpha^7$. The result is in sharp disagreement with the value 3.71(16) fm determined from simple models of the nuclear charge and magnetization distribution. We call for a more definitive nuclear physics value of the $\rm^6Li$ Zemach radius., Comment: 6 pages, 6 figures

In the present work, we investigate the charmed meson loops contributions to the pionic transitions from $Z_c(4020)^+$ to the $D$ wave triplets charmonia by using an effective Lagrangian approach. Our estimations indicate that the predicted branching fraction of $Z_c(4020)^+ \to \pi^+ \psi(1^3D_J) , \ J=(1,2,3)$ are much smaller than the one of $Z_c(4020)^+ \to \pi^+ h_c $. Thus, searching $Z_c(4020)^\pm$ in the $\pi^\pm \psi(1^3D_J) $ invariant mass distributions is impossible. Thus, the observed peak structures at 4.04 and 4.13 GeV in the $\pi^\pm \psi(3770)$ invariant mass distributions should not come from the contributions of $Z_c(4020)^\pm$, and further precise experimental measurements of the $e^+ e^- \to \pi^+ \pi^- \psi(3770)$ process are needed to decode the nature of these two peak structures., Comment: 9 pages, 5 figures, accepted for publication in EPJC

Endometriosis is a non-malignant disorder that affects 176 million women globally. Diagnostic delays result in severe dysmenorrhea, dyspareunia, chronic pelvic pain, and infertility. Therefore, there is a significant need to diagnose patients at an early stage. Our objective in this work is to investigate the potential of deep learning methods to classify endometriosis from ultrasound data. Retrospective data from 100 subjects were collected at the Rutgers Robert Wood Johnson University Hospital (New Brunswick, NJ, USA). Endometriosis was diagnosed via laparoscopy or laparotomy. We designed and trained five different deep learning methods (Xception, Inception-V4, ResNet50, DenseNet, and EfficientNetB2) for the classification of endometriosis from ultrasound data. Using 5-fold cross-validation study we achieved an average area under the receiver operator curve (AUC) of 0.85 and 0.90 respectively for the two evaluation studies., Comment: Accepted to 2023 SPIE Medical Imaging Conference

Graph sampling is a technique to pick a subset of vertices and/ or edges from original graph. Among various graph sampling approaches, Traversal Based Sampling (TBS) are widely used due to low cost and feasibility for many cases, in which Simple Random Walk (SRW) and its variants share a large proportion in TBS. We illustrate the foundation SRW and presents the problems of SRW. Based on the problems, we provide a taxonomy of different Random Walk (RW) based graph sampling methods and give an insight to the reason why and how they revise SRW. our summary includes classical methods and state-of-art RW-based methods. There are 3 ways to propose new algorithms based on SRW, including SRW and its combinations, modified selection mechanisms, and the graph topology modification. We explained the ideas behind those algorithms, and present detailed pseudo codes. In addition, we add the mathematics behind random walk, and the essence of random walk variants, which is not mentioned in detail in many research papers and literature reviews. Apart from RW-based methods, SRW also has related with the non-RW and non-TBS methods, we discuss the relationships between SRW and non-RW methods, and the relationships between SRW and non-TBS methods. The relations between these approaches are formally argued and a general framework to bridge theoretical analysis and practical implementation is provided.

Random walk sampling methods have been widely used in graph sampling in recent years, while it has bias towards higher degree nodes in the sample. To overcome this deficiency, classical methods such as MHRW design weighted walking by repeating low-degree nodes while rejecting high-degree nodes, so that the long-term behavior of Markov chain can achieve uniform distribution. This modification, however, may make the sampler stay in the same node for several times, leading to undersampling. To address this issue, we propose a sampling framework that only need current and candidate node degree to improve the performance of graph sampling methods. We also extend our original idea to a more general framework. Our extended IDRW method finds a balance between the large deviation problem of SRW and sample rejection problem in MHRW. We evaluate our technique in simulation by running extensive experiments on various real-world datasets, and the result show that our method improves the accuracy compared with the state of art techniques. We also investigate the effect of the parameter and give the suggested range for a better usage in application.

Random walk based sampling methods have been widely used in graph sampling in recent years, while it has bias towards higher degree nodes in the sample. To overcome this deficiency, classical methods such as GMD modify the topology of target graphs so that the long-term behavior of Markov chain can achieve uniform distribution. This modification, however, reduces the conductance of graphs, thus makes the sampler stay in the same node for long time, resulting in undersampling. To address this issue, we propose a new way of modifying target graph, thus propose Weighted Jump Random Walk (WJRW) with parameter C to improve the performance. We prove that WJRW can unify Simple Random Walk and uniform distribution through C, and we also conduct extensive experiments on real-world dataset. The experimental results show WJRW can promote the accuracy significantly under the same budget. We also investigate the effect of the parameter C, and give the suggested range for a better usage in application.

Random tensor network states are toy models for holographic duality, which have entanglement properties determined by graph geometry. In this paper, we propose a generalization of the random tensor network states which describe an ensemble of states preserving a given global symmetry. We show that Renyi entropy for this family of states can be described by a quantum extremal surface formula, with corrections to the area law term determined by a bulk gauge theory wavefunction. This provides a toy model of the correspondence between boundary global symmetry and bulk gauge symmetry in holographic duality. We discuss the boundary physical consequences of the bulk deconfined and confined phases., Comment: A paper contributed to "A Festschrift in Honor of the C. N. Yang Centenary". 17 pages. 3 figures

The role of chest X-ray (CXR) imaging, due to being more cost-effective, widely available, and having a faster acquisition time compared to CT, has evolved during the COVID-19 pandemic. To improve the diagnostic performance of CXR imaging a growing number of studies have investigated whether supervised deep learning methods can provide additional support. However, supervised methods rely on a large number of labeled radiology images, which is a time-consuming and complex procedure requiring expert clinician input. Due to the relative scarcity of COVID-19 patient data and the costly labeling process, self-supervised learning methods have gained momentum and has been proposed achieving comparable results to fully supervised learning approaches. In this work, we study the effectiveness of self-supervised learning in the context of diagnosing COVID-19 disease from CXR images. We propose a multi-feature Vision Transformer (ViT) guided architecture where we deploy a cross-attention mechanism to learn information from both original CXR images and corresponding enhanced local phase CXR images. We demonstrate the performance of the baseline self-supervised learning models can be further improved by leveraging the local phase-based enhanced CXR images. By using 10\% labeled CXR scans, the proposed model achieves 91.10\% and 96.21\% overall accuracy tested on total 35,483 CXR images of healthy (8,851), regular pneumonia (6,045), and COVID-19 (18,159) scans and shows significant improvement over state-of-the-art techniques. Code is available https://github.com/endiqq/Multi-Feature-ViT, Comment: Accepted to the 2022 MICCAI Workshop on Medical Image Learning with Limited and Noisy Data