Textile-reinforced mortars; an experimental comparative study of tensile strength improvement methods

In this article, an experimental study is conducted to compare eight improvement methods for the tensile strength of textile-reinforced mortars (TRM). 12 series of samples with different modification methods are compared to determine the most effective factors on crack initiation force and tensile strength of TRM. Eight modification methods are categorized under three main groups of mortar modification, fabric modification, and fabric-mortar interface modification. TRM's first crack force and ultimate force are considered as indices of method performance. One-way ANOVA and factorial analysis were also conducted to statically determine the most significant methods for improving TRM tensile behavior. The results showed that the modification of mortar by short fiber is the most effective method for the enhancement of TRM's first crack force. Also, the methods which led to the transfer of failure mode from mortar to fabrics were the most effective methods on TRM ultimate force improvement. The result showed that coating fabrics with epoxy affects TRM tensile strength more than all other methods. Extra enhancement of TRM ultimate force is achieved by adding silica fume to epoxy before coating the fabrics and spreading the sand and short fibers on impregnated fabrics.

Semantic Graph Neural Network: A Conversion from Spam Email Classification to Graph Classification

In this study, we propose a method named Semantic Graph Neural Network (SGNN) to address the challenging task of email classification. This method converts the email classification problem into a graph classification problem by projecting email into a graph and applying the SGNN model for classification. The email features are generated from the semantic graph; hence, there is no need of embedding the words into a numerical vector representation. The method performance is tested on the different public datasets. Experiments in the public dataset show that the presented method achieves high accuracy in the email classification test against a few public datasets. The performance is better than the state-of-the-art deep learning-based method in terms of spam classification.

Developing Deep Learning Models for the Classification of Pediatric Elbow Radiographic Abnormalities of Comparable Performance to Physicians: Strategies for Model Optimization With Small Sized Development Sets

BackgroundTo compare the performance of an AI model based on strategies designed to overcome small sized development sets to pediatric ER physicians at a classification triage task of pediatric elbow radiographs. Methods1,314 pediatric elbow lateral radiographs (mean age: 8.2 years) were retrospectively retrieved, binomially classified based on their annotation as normal or abnormal (with pathology), and randomly partitioned into a development set (993 images), tuning set (109 images), second tuning set (100 images) and test set (112 images). The AI model was trained on the development set and utilized the EfficientNet B1 compound scaling network architecture and online augmentations. Its performance on the test set was compared to a group of five physicians (inter-rater agreement: fair). Statistical analysis: AUC of AI model - DeLong method. Performance of AI model and physician groups - McNemar test. ResultsAccuracy of the model on the test set - 0.804 (95% CI, 0.718 - 0.873), AUROC - 0.872 (95% CI, 0.831 - 0.947). AI model performance compared to the physician group on the test set - sensitivity 0.790 (95% CI 0.684 to 0.895) vs 0.649 (95% CI 0.525 to 0.773), p value 0.088; specificity 0.818 (95% CI 0.716 to 0.920) vs 0.873 (95% CI 0.785 to 0.961), p value 0.439.ConclusionsThe AI model for elbow radiograph triage designed with strategies to optimize performance for a small sized development set showed comparable performance to physicians.

Fuzzy Logic-Based Direct Power Control Method for PV Inverter of Grid-Tied AC Microgrid without Phase-Locked Loop

A voltage source inverter (VSI) is the key component of grid-tied AC Microgrid (MG) which requires a fast response, and stable, robust controllers to ensure efficient operation. In this paper, a fuzzy logic controller (FLC)-based direct power control (DPC) method for photovoltaic (PV) VSI was proposed, which was modelled by modulating MG’s point of common coupling (PCC) voltage. This paper also introduces a modified grid synchronization method through the direct power calculation of PCC voltage and current, instead of using a conventional phase-locked loop (PLL) system. FLC is used to minimize the errors between the calculated and reference powers to generate the required control signals for the VSI through sinusoidal pulse width modulation (SPWM). The proposed FLC-based DPC (FLDPC) method has shown better tracking performance with less computational time, compared with the conventional MG power control methods, due to the elimination of PLL and the use of a single power control loop. In addition, due to the use of FLC, the proposed FLDPC exhibited negligible steady-state oscillations in the output power of MG’s PV-VSI. The proposed FLDPC method performance was validated by conducting real-time simulations through real time digital simulator (RTDS). The results have demonstrated that the proposed FLDPC method has a better reference power tracking time of 0.03 s along with reduction in power ripples and less current total harmonic distortion (THD) of 1.59%.

Robust PPG Peak Detection Using Dilated Convolutional Neural Networks

<div>Accurate peak determination from noise-corrupted photoplethysmogram (PPG) signal is the basis for further analysis of physiological quantities such as heart rate and heart rate variability. In the past decades, many methods have been proposed to provide reliable peak detection. These peak detection methods include rule-based algorithms, adaptive thresholds, and signal processing techniques. However, they are designed for noise-free PPG signals and are insufficient for PPG signals with low signal-to-noise ratio (SNR). This paper focuses on enhancing PPG noise-resiliency and proposes a robust peak detection algorithm for noise and motion artifact corrupted PPG signals. Our algorithm is based on Convolutional Neural Networks (CNN) with dilated convolutions. Using dilated convolutions provides a large receptive field, making our CNN model robust at time series processing. In this study, we use a dataset collected from wearable devices in health monitoring under free-living conditions. In addition, a data generator is developed for producing noisy PPG data used for training the network. The method performance is compared against other state-of-the-art methods and tested in SNRs ranging from 0 to 45 dB. Our method obtains better accuracy in all the SNRs, compared with the existing adaptive threshold and transform-based methods. The proposed method shows an overall precision, recall, and F1-score 80%, 80%, and 80% in all the SNR ranges. However, these figures for the other methods are below 78%, 77%, and 77%, respectively. The proposed method proves to be accurate for detecting PPG peaks even in the presence of noise.</div>

An assessment of reference method selective broths and plating media using 19 Listeria spp. highlights the importance of including diverse species in Listeria spp. method evaluations

Reference methods developed for L. monocytogenes are commonly used for Listeria spp. detection. Improved method performance data are needed, since the genus Listeria has expanded from 6 to 26 species and now includes several Listeria sensu lato species, which can show phenotypes distinct from Listeria sensu stricto . Here, we evaluated growth of 19 Listeria spp., including 12 recently described sensu lato species, using the media specified by (i) the U.S. Food and Drug Administration (FDA) Bacteriological Analytical Manual , (ii) the U.S. Department of Agriculture (USDA) Microbiology Laboratory Guidebook , and (iii) the International Organization for Standardization (ISO). The FDA enrichment procedure allowed all species to grow to detectable levels (≥ 4 log 10 ), yielded the highest mean growth (7.58 log 10 ), and was the only procedure where no sensu lato species yielded significantly higher bacterial growth than a sensu stricto species. With the USDA or ISO enrichment procedures several sensu lato species yielded significantly higher bacterial growth than either L. seeligeri or L. ivanovii , suggesting that these two sensu stricto species could be outgrown by sensu lato species. On selective and differential agars, L. seeligeri, L. ivanovii, and L. grayi yielded atypical colony morphologies and/or showed inhibited growth (which may lead to incorrect classification of a sample as negative), while several newly described sensu lato species grew well and showed typical morphologies. Overall, our study shows that the ability to detect different Listeria spp. can be impacted by the specific broth and selective and differential agars used. Our data will aid with selection of media and detection methods for environmental Listeria monitoring programs and facilitate selection of methods that are most likely to detect the targeted Listeria groups (e.g., Listeria sensu stricto, which appear to be the most appropriate index organisms for the pathogen L. monocytogenes ).

Application of Artificial Intelligence Technology in Decision-Making of Mechanical Manufacturing Process

The traditional mechanical manufacturing process is to transform all raw materials into the final materials and products and directly into the international market all the production process, in this process we involved a lot of problems about decision-making methods, decision-making process is a most basic production technology activity, it is widely exists in the whole social life and each link of enterprise production. This paper studies the decision-making method of mechanical manufacturing process based on artificial intelligence, optimizes the process parameters of plastic integrated mechanical manufacturing process, and compares it with the traditional decision-making method. Finally, the experimental results are obtained that the traditional decision-based method is reduced by more than 10% in size error. But several experiments, the AI decision-making method appeared deviation, the error results are higher than the traditional decision-making method, which may be objective factors, but also reflects the possibility of instability, in the result of deformation. AI-based decision method performance is higher than the traditional decision-making method, reduce the deformation amount by 3.5%

Multi-Drone 3D Building Reconstruction Method

In the recent decade, the rapid development of drone technologies has made many spatial problems easier to solve, including the problem of 3D reconstruction of large objects. A review of existing solutions has shown that most of the works lack the autonomy of drones because of nonscalable mapping techniques. This paper presents a method for centralized multi-drone 3D reconstruction, which allows performing a data capturing process autonomously and requires drones equipped only with an RGB camera. The essence of the method is a multiagent approach—the control center performs the workload distribution evenly and independently for all drones, allowing simultaneous flights without a high risk of collision. The center continuously receives RGB data from drones and performs each drone localization (using visual odometry estimations) and rough online mapping of the environment (using image descriptors for estimating the distance to the building). The method relies on a set of several user-defined parameters, which allows the tuning of the method for different task-specific requirements such as the number of drones, 3D model detalization, data capturing time, and energy consumption. By numerical experiments, it is shown that method parameters can be estimated by performing a set of computations requiring characteristics of drones and the building that are simple to obtain. Method performance was evaluated by an experiment with virtual building and emulated drone sensors. Experimental evaluation showed that the precision of the chosen algorithms for online localization and mapping is enough to perform simultaneous flights and the amount of captured RGB data is enough for further reconstruction.

An Efficient Approach Using Knowledge Distillation Methods to Stabilize Performance in a Lightweight Top-Down Posture Estimation Network

Multi-person pose estimation has been gaining considerable interest due to its use in several real-world applications, such as activity recognition, motion capture, and augmented reality. Although the improvement of the accuracy and speed of multi-person pose estimation techniques has been recently studied, limitations still exist in balancing these two aspects. In this paper, a novel knowledge distilled lightweight top-down pose network (KDLPN) is proposed that balances computational complexity and accuracy. For the first time in multi-person pose estimation, a network that reduces computational complexity by applying a “Pelee” structure and shuffles pixels in the dense upsampling convolution layer to reduce the number of channels is presented. Furthermore, to prevent performance degradation because of the reduced computational complexity, knowledge distillation is applied to establish the pose estimation network as a teacher network. The method performance is evaluated on the MSCOCO dataset. Experimental results demonstrate that our KDLPN network significantly reduces 95% of the parameters required by state-of-the-art methods with minimal performance degradation. Moreover, our method is compared with other pose estimation methods to substantiate the importance of computational complexity reduction and its effectiveness.

A Spatial-Temporal Feature Extraction and Unsupervised Recognition of Video Anomalies in Real Time

Anomaly detection is an area of video analysis has a great importance in automated surveillance. Although it has been extensively studied, there has been little work started using CNN networks. Hence, in this thesis we presented a novel approach for learning motion features and modeling normal Spatio-temporal dynamics for anomaly detection. In our technique, we capture variations in scale of the patterns of motion in an image object by using optical flow dense estimation technique and train our auto encoder model using convolution long short term memories (ConvLSTM2D) as we are processing video frames and we predict the anomaly in real time using Euclidean distance between the generated and the ground truth frame and we achieved a real time accuracy of nearly 98% for the youtube videos which are not used for either testing or training. Error between the network’s output and the target output is used to classify a video volume as normal or abnormal. In addition to the use of reconstruction error, we also use prediction error for anomaly detection. The prediction models show comparable performance with state of the art methods. In comparison with the proposed method, performance is improved in one dataset. Moreover, running time is significantly faster.