Statistical imputation is a field of study that attempts to fill missing data. It is commonly applied to population statistics whose data have no correlation with running time. For a time series, data is typically analyzed using the autocorrelation function (ACF), the Fourier transform to estimate power spectral densities (PSD), the Allan deviation (ADEV), trend extensions, and basically any analysis that depends on uniform time indexes. We explain the rationale for an imputation algorithm that fills gaps in a time series by applying a backward, inverted replica of adjacent live data. To illustrate, four intentional massive gaps that exceed 100% of the original time series are recovered. The L(f) PSD with imputation applied to the gaps is nearly indistinguishable from the original. Also, the confidence of ADEV with imputation falls within 90% of the original ADEV with mixtures of power-law noises. The algorithm in Python is included for those wishing to try it.

The method is designed for presentation attack detection. Both classical descriptor such as SURF and deep learning is used to build the model.

Temperature monitoring during medical interventions such as hyperthermia or radio frequency ablation is a demanding task. An alternative to expensive MRI measurements are ultrasound measurements exploiting the dependence of sound velocity on temperature. In this paper, a new measurement technique is presented that determines the sound velocity spatially resolved in tissue phantoms with the aim of temperature monitoring. The measurements are performed with annular arrays operating element-wise in pulse-echo mode to allow synthetic focusing. Unlike speckle techniques, which also evaluate echoes from scatterers, this method allows to determine the true position of scatterers and absolute values of local sound velocity, rather than evaluating only relative changes. This is achieved by varying the set of delay times used for focusing and determining the set that gives the maximum amplitude of the focused echo. The measurement technique is described in detail and measurement results are shown. The achievable resolution (1mm axial and 0.4mm lateral) and accuracy of 98% are determined. Sound field simulations are used to model the measurement process and vary measurement parameters. Thus, the effect of parameter variations on the achievable accuracy can be evaluated very precisely.

Event cameras are novel, bio-inspired, imagers that output per-pixel brightness changes at ~1us latency. Besides their high response rates, they offer numerous advantages over conventional cameras, such as no motion blur, high dynamic range, and low power consumption. However, event cameras suffer from some limitations such as lacking the intensity information that regular cameras provide. In this paper, we present a hybrid object tracking algorithm that leverages both images and events, thus, providing a complementary approach that utilizes some of the advantages of both imaging types. Our tracking algorithm detects the objects in the image frames, then tracks objects in the blind time between consecutive frames using per-object event masks extracted from the event data. Moreover, we set up a data collection experiment to evaluate and analyze our algorithm’s performance using Dynamic and Active-Pixel Vision Sensor (DAVIS), which combines a monochrome camera as well as an event-based sensor using the same pixel array. Results show that our tracking algorithm can reach up to 500 Hz tracking rates based on a standard image framerate of 24 Hz and asynchronous event-data data collected by the hybrid camera.

Developing a computer-aided diagnostic system for detecting various skin malignancies from images has attracted many researchers. Unlike many machine learning approaches such as Artificial Neural Networks, Genetic Programming (GP) automatically evolves models with flexible representation. GP successfully provides effective solutions using its intrinsic ability to select prominent features (i.e. feature selection) and build new features (i.e. feature construction). Existing approaches have utilized GP to construct new features from the complete set of original features and the set of operators. However, the complete set of features may contain redundant or irrelevant features that do not provide useful information for classification. This study aims to develop a two-stage GP method where the first stage selects prominent features, and the second stage constructs new features from these selected features and operators such as multiplication in a wrapper approach to improve the classification performance. To include local, global, texture, color, and multi-scale image properties of skin images, GP selects and constructs features extracted from local binary patterns and pyramid-structured wavelet decomposition. The accuracy of this GP method is assessed using two real-world skin image datasets captured from the standard camera and specialized instruments, and compared with commonly used classification algorithms, three state-of-the-art, and an existing embedded GP method. The results reveal that this new approach of feature selection and feature construction effectively helps improve the performance of the machine learning classification algorithms. Unlike other black-box models, the evolved models by GP are interpretable, therefore the proposed method can assist dermatologists to identify prominent features, which has been shown by further analysis on the evolved models.

Multi-access Edge Computing (MEC) has recently emerged as a potential technology to serve the needs of mobile devices (MDs) in 5G and 6G cellular networks. By offloading tasks to high-performance servers installed at the edge of the wireless networks, resource-limited MDs can cope with the proliferation of the recent computationally-intensive applications. In this paper, we study the computation offloading problem in a massive multiple-input multiple-output (MIMO)-based MEC system where the base stations are equipped with a large number of antennas. Our objective is to minimize the power consumption and offloading delay at the MDs under the stochastic system environment. To this end, we formulate the problem as a Markov Decision Process (MDP) and propose two Deep Reinforcement Learning (DRL) strategies to learn the optimal offloading policy without any prior knowledge of the environment dynamics. First, a Deep Q-Network (DQN) strategy to solve the curse of the state space explosion is analyzed. Then, a more general Proximal Policy Optimization (PPO) strategy to solve the problem of discrete action space is introduced. Simulation results show that the proposed DRL-based strategies outperform the baseline and state-of-the-art algorithms. Moreover, our PPO algorithm exhibits stable performance and efficient offloading results compared to the benchmark DQN strategy.

How to reduce dose radiation while preserving the image quality as when using standard dose is one of the most important topics in the Computed Tomography (CT) imaging domain due to quality of low dose CT (LDCT) images is often strongly affected by noise and artifacts. Recently there has been considerable interest in using deep learning as a post-processing step to improve the quality of reconstructed LDCT images. This paper, first, gives an overview of learning-based LDCT image denoising methods from patch-based early learning methods to state-of-the-art CNN-based ones, and then a novel CNN-based method is presented. In the proposed method, preprocessing and post-processing techniques are integrated into a dilated convolutional neural network to extend receptive fields. Hence, large distance pixels in input images will participate in enriching feature maps of the learned model and thus effectively denoises. Experimental results showed that the proposed method is light while its denoising effectiveness is competitive with well-known CNN-based models.

Numerical simulations are carried out on the IEEE 13 node test feeder.

General artificial intelligence is a trade-off between the inductive bias of an algorithm and its out-of-distribution generalization performance. The conspicuous impact of inductive bias is an unceasing trend of improved predictions in various problems in computer vision like object detection. Although a recently introduced object detection technique, based on transformers (DETR), shows results competitive to the conventional and modern object detection models, its accuracy deteriorates for detecting small-sized objects (in perspective). This study examines the inductive bias of DETR and proposes a normalized inductive bias for object detection using a transformer (SOF-DETR). It uses a lazy-fusion of features to sustain deep contextual information of objects present in the image. The features from multiple subsequent deep layers are fused with element-wise-summation and input to a transformer network for object queries that learn the long and short-distance spatial association in the image by the attention mechanism. SOF-DETR uses a global set-based prediction for object detection, which directly produces a set of bounding boxes. The experimental results on the MS COCO dataset show the effectiveness of the added normalized inductive bias and feature fusion techniques by detecting more small-sized objects than DETR.

The operation of energy storage systems (ESSs) in power systems where variable renewable energy sources (VRESs) and ESSs must contribute to securing the supply, can be considered as an arbitrage against scarcity. The value of using stored energy instantly must be balanced against its potential future value and future risk of scarcity. This paper proposes a multi-stage stochastic programming model for the operation of microgrids with VRESs, ESSs and thermal generators that is divided into a short- and a long-term model. The short-term model utilizes information from forecasts updated every six hours, while the long-term model considers the value of stored energy beyond the forecast horizon. The model is solved using stochastic dual dynamic programming and Markov chains, and the results show that the significance of accounting for short- and long-term uncertainty increases for systems with a high degree of variable renewable generation and ESSs and limited dispatchable generation capacity.

Today, many large-scale software projects have members working from home, which has changed the way teams coordinate work. To better understand coordination in this setting, we conducted a case study through which we examined two teams in a large-scale agile project by observing meetings and conducting 17 interviews. Through the lens of Relational Coordination Theory (RCT), we analyzed the use of the goal-setting framework Objectives and Key Results (OKRs) and the collaboration tool Slack. Slack was used for frequent, timely, and problem-solving communication and, and its use decreased the number of planned meetings. However, discussions often started on Slack and continued in virtual ad-hoc meetings. The use of OKRs facilitated knowledge sharing, helped the teams align their goals, and provided inter-team insights. The main implication of our research is that projects using OKRs need to support project members, especially in formulating the key results that align and motivate the teams to work toward the same mission.

Wastewater got much of our intention these days because wastewater is polluting our lakes, pounds and even sea have a lot of contaminated amount of waste. This water is hazardous for the acute life, dangerous for living things. Wastewater polluted the natural reservoirs. Over the past, the knowledge of the mechanisms of electrochemical wastewater treatment has progressively evolved. A comprehensive understanding of the types of methods and mechanisms of treatment of wastewater is a prerequisite to the understanding of their relativities and elucidation of intermediate products generated during the oxidation process and degradation pathways. The type, nature, and quantity of reactive species generated in electrochemical treatment processes are controlled by many factors, including the type of the treatment technique, electrode/electro catalyst materials, water/wastewater composition, water pH conditions, and operating parameters are to be considered. Multiple methods such as separation, conversion and combined methods are used for treatment. However, basic principle works on the electrochemical mechanism. This article gives the basic idea of electrochemical methods working principles, techniques being considered. It will also help us understand the byproducts recovery of different metal ions and how they converted into useful form. Best methods based on the efficiency and economic value. Feasibility of long term and short term methods for the treatment of wastewater.

Operating at reduced voltages promises substantial energy efficiency improvement, however the downside is significant down-scaling of clock frequency. This paper propose vision chips as excellent fit for low-voltage operation. Low-level sensory data processing in many Internet-of-Things (IoT) devices pursue energy efficiency by utilizing sleep modes or slowing the clocking to the minimum. To curb the share of stand-by power dissipation in those designs, near-threshold/sub-threshold operational points or ultra-low-leakage processes in fabrication are employed. Those limit the clocking rates significantly, reducing the computing throughputs of individual processing cores. In this contribution we explore compensating for the performance loss of operating in near-threshold region ($V_{dd}=$0.6V) through massive parallelization. Benefits of near-threshold operation and massive parallelism are optimum energy consumption per instruction operation and minimized memory round-trips, respectively. The Processing Elements (PE) of the design are based on Transport Triggered Architecture. The fine grained programmable parallel solution allows for fast and efficient computation of learnable low-level features (e.g. local binary descriptors and convolutions). Other operations, including Max-pooling have also been implemented. The programmable design achieves excellent energy efficiency for Local Binary Patterns computations. Our results demonstrates that the inherent properties of chip processor and vision applications allow voltage and clock frequency aggressively without having to compromise performance.

Background: Commonly used items like wallets, keys, and phones are both tricky and impractical to disinfect from the dangers of harmful microbes. This study demonstrates the efficacy of UVC-LED technology in creating an efficient, useful, and practical solution. Methods: As a demonstration of the efficacy of the UVC-LED light (275 nm), a panel of UVC LEDs was fabricated and was driven with a constant current electronic driver. Staphylococcus aureus and Escherichia coli were placed on Petri dishes, and placed 38 cm away from the UVC-LED panel. UVC flux measured at the petri dishes was 0.093 mW/cm2 . The method involved exposing both the bacteria to UVC treatment for 4 and 8 minutes. For each petri dish, the number of colony forming units were compared before and after the treatment and compared to the control. Results: A significant reduction in colony forming unit (cfu) counts was found in all samples for both sets of bacteria: 97.9% in the 4 minutes treat- ment(22.3 mJ/cm2), and 99.9% in the 8 minutes treatment(44.6 mJ/cm2). Conclusion: UVC-LED technology offers an effective, simple and inexpensive approach for disinfection.

*This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible. In this article, we propose a 3-D synthetic aperture imaging method with spherical antenna scanning to identify scatterers located close to an antenna array, such as fixtures or support of the antenna. Previous studies have shown that 2-D and 3-D synthetic aperture imaging techniques with planar, circular, and cylindrical scanning can successfully reconstruct spatial images of near-field scatterers. The spherical scanning approach considered in this article is expected to improve the 3-D image resolution because more angular diversity can be achieved in the elevation direction. However, as we show in this study, simple extension of the previous techniques to the spherical case results in undesired blur artifacts in the reconstructed image. To overcome this problem, we introduce a correction factor in the image reconstruction. The proposed imaging algorithm is validated by numerical electromagnetic simulation based on the method of moments.

Unmanned Aerial Vehicles (UAVs, also called drones) have been widely deployed in our living environments for a range of applications such as healthcare, agriculture, and logistics. Despite their unprecedented advantages, the increased number of UAVs and their growing threats demand high-performance management and emergency control strategies. To accurately detect a UAV’s working state including hovering and flying, data collection from Radio Frequency (RF) signals is a key step of these strategies and has thus attracted significant research interest. Deep neural networks (DNNs) have been applied for UAV state detection and shown promising potentials. While existing work mostly focuses on improving the DNN structures, we discover that RF signals’ pre-processing before sending them to the classification model is as important as improving the DNN structures. Experiments on a dataset show that, after applying proposed pre-processing methods, the 10-time average accuracy is improved from 46.8% to 91.9%, achieving nearly 50% gain comparing with the benchmark work using the same DNN structure. This work also outperforms the state-of-the-art CNN models, confirming the great potentials of data pre-processing for RF-based UAV state detection.

Scene text recognition is a popular topic and can benefit various tasks. Although many methods have been proposed for the close-set text recognition challenges, they cannot be directly applied to open-set scenarios, where the evaluation set contains novel characters not appearing in the training set. Conventional methods require collecting new data and retraining the model to handle these novel characters, which is an expensive and tedious process. In this paper, we propose a label-to-prototype learning framework to handle novel characters without retraining the model. In the proposed framework, novel characters are effectively mapped to their corresponding prototypes with a label-to-prototype learning module. This module is trained on characters with seen labels and can be easily generalized to novel characters. Additionally, feature-level rectification is conducted via topology-preserving transformation, resulting in better alignments between visual features and constructed prototypes while having a reasonably small impact on model speed. A lot of experiments show that our method achieves promising performance on a variety of zero-shot, close-set, and open-set text recognition datasets.

Deep learning’s rapid gains in automation are making it more popular in a variety of complex jobs. The self-driving object is an emerging technology that has the potential to transform the entire planet. The steering control of an automated item is critical to ensuring a safe and secure voyage. Consequently, in this study, we developed a methodology for predicting the steering angle only by looking at the front images of a vehicle. In addition, we used an Internet of Things-based system for collecting front images and steering angles. A Raspberry Pi (RP) camera is used in conjunction with a Raspberry Pi (RP) processing unit to capture images from vehicles, and the RP processing unit is used to collect the angles associated with each image. Apart from that, we’ve made use of deep learning-based algorithms such as VGG16, ResNet-152, DenseNet-201, and Nvidia’s models, all of which were trained using labeled training data. Our models are End-to-End CNN models, which do not require extracting elements from data such as roads, lanes, or other objects before predicting steering angle. As a result of our comparative investigation, we can conclude that the Nvidia model’s performance was satisfactory, with a Mean Squared Error (MSE) value of 0. But the Nvidia model outperforms the other pre-trained models, even though other models work well.