Accuracy Error Research Articles

Object and Pedestrian Detection on Road in Foggy Weather Conditions by Hyperparameterized YOLOv8 Model

Connected cooperative and automated (CAM) vehicles and self-driving cars need to achieve robust and accurate environment understanding. With this aim, they are usually equipped with sensors and adopt multiple sensing strategies, also fused among them to exploit their complementary properties. In recent years, artificial intelligence such as machine learning- and deep learning-based approaches have been applied for object and pedestrian detection and prediction reliability quantification. This paper proposes a procedure based on the YOLOv8 (You Only Look Once) method to discover objects on the roads such as cars, traffic lights, pedestrians and street signs in foggy weather conditions. In particular, YOLOv8 is a recent release of YOLO, a popular neural network model used for object detection and image classification. The obtained model is applied to a dataset including about 4000 foggy road images and the object detection accuracy is improved by changing hyperparameters such as epochs, batch size and augmentation methods. To achieve good accuracy and few errors in detecting objects in the images, the hyperparameters are optimized by four different methods, and different metrics are considered, namely accuracy factor, precision, recall, precision–recall and loss.

FLIGHTED: Inferring Fitness Landscapes from Noisy High-Throughput Experimental Data.

Machine learning (ML) for protein design requires large protein fitness datasets generated by high-throughput experiments for training, fine-tuning, and benchmarking models. However, most models do not account for experimental noise inherent in these datasets, harming model performance and changing model rankings in benchmarking studies. Here we develop FLIGHTED, a Bayesian method of accounting for uncertainty by generating probabilistic fitness landscapes from noisy high-throughput experiments. We demonstrate how FLIGHTED can improve model performance on two categories of experiments: single-step selection assays, such as phage display and SELEX, and a novel high-throughput assay called DHARMA that ties activity to base editing. We then compare the performance of standard machine-learning models on fitness landscapes generated with and without FLIGHTED. Accounting for noise significantly improves model performance, especially of CNN architectures, and changes relative rankings on numerous common benchmarks. Based on our new benchmarking with FLIGHTED, data size, not model scale, currently appears to be limiting the performance of protein fitness models, and the choice of top model architecture matters more than the protein language model embedding. Collectively, our results indicate that FLIGHTED can be applied to any high-throughput assay and any machine learning model, making it straightforward for protein designers to account for experimental noise when modeling protein fitness.

Development of a 3D Hydrogel SERS Chip for Noninvasive, Real-Time pH and Glucose Monitoring in Sweat.

Traditional diagnostic methods, such as blood tests, are invasive and time-consuming, while sweat biomarkers offer a rapid physiological assessment. Surface-enhanced Raman spectroscopy (SERS) has garnered significant attention in sweat analysis because of its high sensitivity, label-free nature, and nondestructive properties. However, challenges related to substrate reproducibility and interference from the biological matrix persist with SERS. This study developed a novel ratio-based 3D hydrogel SERS chip, providing a noninvasive solution for real-time monitoring of pH and glucose levels in sweat. Encapsulating the probe molecule (4-MBN) in nanoscale gaps to form gold nanoflower-like nanotags with internal standards and integrating them into an agarose hydrogel to create a 3D flexible SERS substrate significantly enhances the reproducibility and stability of sweat analysis. Gap-Au nanopetals modified with probe molecules are uniformly dispersed throughout the porous hydrogel structure, facilitating the effective detection of the pH and glucose in sweat. The 3D hydrogel SERS chip demonstrates a strong linear relationship in pH and glucose detection, with a pH response range of 5.5-8.0 and a glucose detection range of 0.01-5 mM, with R2 values of 0.9973 and 0.9923, respectively. In actual sweat samples, the maximum error in pH detection accuracy is only 1.13%, with a lower glucose detection limit of 0.25 mM. This study suggests that the ratio-based 3D hydrogel SERS chip provides convenient, reliable, and rapid detection capabilities with substantial application potential for analyzing body fluid pH and glucose.

Unveiling the short and faint X-ray transient nature of IGR J17419-2802

We report new X-ray results from the INTErnational Gamma-Ray Astrophysics Laboratory ( and observations of the hitherto poorly studied unidentified X-ray transient We studied in detail the temporal, spectral, and energetic properties of three hard X-ray outbursts detected above 20 keV by They are all characterized by an average X-ray luminosity of 3times 1035 erg $ and a constrained duration of a few days. This marks a peculiarly short and faint X-ray transient nature for From archival unpublished soft X-ray observations, we found that the source spends most of the time undetected at very low X-ray fluxes (down to $<4.7 $ erg cm $^ $ s$^ $) for a dynamic range $>$2,000 when in outburst. We provided an accurate arcsecond-sized source error circle. Inside it, we pinpointed the best candidate near-infrared counterpart whose photometric properties are compatible with a late-type spectral nature. Based on our new findings, we suggest that is a new member of the very faint X-ray transients (VFXTs) class. Detailed investigations of VFXT outbursts above 20 keV are particularly rare. In this respect, our reported outbursts are among the best studied to date; in particular, their constrained duration of a few days is among the shortest ever measured for a VFXT. X-rays: binaries: individual: src

Improved a posteriori error bounds for reduced port-Hamiltonian systems

Projection-based model order reduction of dynamical systems usually introduces an error between the high-fidelity model and its counterpart of lower dimension. This unknown error can be bounded by residual-based methods, which are typically known to be highly pessimistic in the sense of largely overestimating the true error. This work applies two improved error bounding techniques, namely (a) a hierarchical error bound and (b) an error bound based on an auxiliary linear problem, to the case of port-Hamiltonian systems. The approaches rely on a secondary approximation of (a) the dynamical system and (b) the error system. In this paper, these methods are adapted to port-Hamiltonian systems. The mathematical relationship between the two methods is discussed both theoretically and numerically. The effectiveness of the described methods is demonstrated using a challenging three-dimensional port-Hamiltonian model of a classical guitar with fluid–structure interaction.

Short-term PM2.5 forecasting using a unique ensemble technique for proactive environmental management initiatives

Particulate matter with a diameter of 2.5 microns or less (PM2.5) is a significant type of air pollution that affects human health due to its ability to persist in the atmosphere and penetrate the respiratory system. Accurate forecasting of particulate matter is crucial for the healthcare sector of any country. To achieve this, in the current work, a new time series ensemble approach is proposed based on various linear (autoregressive, simple exponential smoothing, autoregressive moving average, and theta) and nonlinear (nonparametric autoregressive and neural network autoregressive) models. Three ensemble models are also developed, each employing distinct weighting strategies: equal distribution of weight among all single models (ESME), weight assignment based on training average accuracy errors (ESMT), and weight assignment based on validation mean accuracy measures (ESMV). This technique was applied to daily PM2.5 concentration data from 1 January 2019, to 31 May 2023, in Pakistan’s main cities, including Lahore, Karachi, Peshawar, and Islamabad, to forecast short-term PM2.5 concentrations. When compared to other models, the best ensemble model (ESMV) demonstrated mean errors ranging from 3.60% to 25.79% in Islamabad, 0.81%–13.52% in Lahore, 1.08%–7.06% in Karachi, and 1.09%–12.11% in Peshawar. These results indicate that the proposed ensemble approach is more efficient and accurate for short-term PM2.5 forecasting than existing models. Furthermore, using the best ensemble model, a forecast was made for the next 15 days (June 1 to 15 June 2023). The forecast showed that in Lahore, the highest PM2.5 value (236.00 μg/m3) was observed on 8 June 2023. Other days also displayed higher and poor air quality throughout the 15 days. Conversely, Karachi experienced moderate PM2.5 concentration levels between 50 μg/m3 and 80 μg/m3. In Peshawar, the PM2.5 concentration levels were consistently unhealthy, with the highest peak (153.00 μg/m3) observed on 9 June 2023. This forecasting experience can assist environmental monitoring organizations in implementing cost-effective planning to minimize air pollution.

High-resolution SAR optoelectronic processor based on sensor-less adaptive optics

Studying phase error is central to synthetic aperture radar (SAR) research. Phase error due to the real motion status of the SAR platform and propagation effects can reduce the utility of high-resolution SAR images even with theoretical estimation-based phase error correction. Adaptive optics (AO) can be used to correct optical aberrations. This study proposes an advanced, high-resolution SAR optoelectronic processor by integrating conventional processors with sensor-less AO techniques. This processor provides accurate adaptive phase error correction capabilities. The processing results of SAR echo data demonstrate the effectiveness of the proposed system in adaptive phase error correction and imaging improvement.

Power quality optimization technology for photovoltaic nanoscale electronic grid-connected systems based on FNN algorithm

With the continuous maturity and widespread application of photovoltaic power generation technology, the power quality of photovoltaic grid-connected systems has become a significant factor affecting the power grid’s stability and the lifespan of equipment. This research aims to improve the power quality in photovoltaic nanoscale electronic grid-connected systems. Through in-depth analysis of harmonic problems in grid-connected systems, a harmonic prediction and governance means using a fuzzy neural network algorithm was put forward. To improve the accuracy and adaptability of predictions, the long short-term memory network was combined with a fuzzy neural network to form an improved deep learning model. Comparative experiments confirmed that the research prediction model exhibited advantages in training convergence, prediction accuracy, and various error evaluation indicators, with an accuracy of up to 96.24% and an R2 value as high as 0.9998. It effectively reduced harmonic content by 2.4% and significantly improved the harmonic compensation effect. Based on the above-mentioned model, dynamic compensation governance for harmonics was implemented. By predicting the harmonic content in grid-connected electricity in real-time and taking corresponding measures for compensation, the adverse effects of harmonics on the entire grid-connected system were effectively reduced. The deep learning-based power quality optimization technology developed in this research can accurately predict harmonic problems in photovoltaic nanoscale electronic grid-connected systems. Moreover, it can provide efficient compensation governance solutions and help improve the entire power grid’s energy efficiency and operational stability.

Variable coefficient-informed neural network for PDE inverse problem in fluid dynamics

Variable-coefficient equations are crucial in the field of fluid dynamics as they accurately capture the spatial and temporal variations of fluid properties. In many cases, there exist some constraints among multiple coefficients and embedding these constraints into neural networks poses a challenge. In this paper, we design a variable coefficient-informed neural network (VCINN) to address the inverse problem of variable-coefficients partial differential equation in fluid dynamics. The VCINN framework integrates the physics-informed neural network (PINN) with the constraints among multiple coefficients, encoding both constraints and physics information into the neural networks. Compared to classical PINN, VCINN enjoys such several advantages as parallelization capacity, embedding constraint information and efficient hyperparameter adjustment. Through series of examples, the capability of the approach to recover coefficients from observations has been demonstrated. Numerical results indicate that the present method achieves higher accuracy and lower training error compared to classical PINN.

Prediction model of maximum stress for concrete pipes based on XGBoost-PSO algorithm

Due to varying burial conditions and service environments, concrete pipes exhibit complex mechanical behavior. Existing theoretical, simulation, and test methods for determining the maximum stress of pipes have significant limitations, including excessive assumptions, low efficiency, and restricted applicable conditions. Therefore, there is an urgent need to propose a multi-factor associated method for pipe stress prediction. This paper proposed an innovative approach to finely simulate the mechanics response for concrete pipes under the coupling action of stress-seepage-fluid fields. The accuracy of this numerical simulation method was validated through full-scale test. Sensitivity analysis based on Morris sensitivity analysis theory was conducted on traffic load magnitude and speed, buried depth, groundwater table, fluid height and velocity, bedding strength, backfill strength, and pipe diameter. A dataset of “physical variables - maximum stress” for concrete pipes was established. A multi-factor-correlated prediction model of maximum stress for concrete pipe was proposed using XGBoost machine learning method optimized by Particle Swarm Optimization (PSO) algorithm. The results indicate that pipe diameter is highly sensitive; buried depth and traffic load magnitude are sensitive; groundwater table, bedding strength, and backfill strength are moderately sensitive; while fluid height, traffic speed, and flow velocity are insensitive. The XGBoost-PSO model demonstrates the highest accuracy and lowest error compared to BP, RF, and XGBoost models, with improvements in prediction accuracy of 59.6 %, 23.8 %, and 8.6 %, respectively. The model achieves RMSE of 0.118 and MAE of 0.213, demonstrating the suitability of the XGBoost-PSO model for predicting maximum stress for concrete pipes.

OCLVerifer: Automated Verfication of OCL Contracts in Requirements Models

Object Constraint Language (OCL) is one lightweight formal specification. Integrated within the Unified Modeling Language (UML) standard, it serves as a cornerstone in requirements modeling, enjoying widespread adoption across various domains. OCL can precisely define the pre- and post-condition of system operations and system invariants. While OCL provides a simple yet expressive syntax, it lacks clarity in mapping Object-Oriented (OO) concepts, such as object states, object links, and object attributes. This ambiguity makes it challenging for OO developers to identify errors in requirements. In this paper, we propose an approach named OCLVerifier, which can automatically detect the requirements errors of OCL, such as conflict, redundancy, and failure error. OCLVerifier first transforms OO contracts and detection patterns into SMT formulas and then proves them by using a SMT solver. Finally, the results are mapped to the original OCL contracts to display detailed error type and location information. To evaluate OCLVerifier, we conducted a comprehensive evaluation of four case studies. Experimental results indicate that OCLVerifier successfully identifies 65.5% of error cases, with each identified case offering accurate error location information. Compared with human experts, OCLVerifier can reduce evaluation time by 80.8% while enhancing repair accuracy by 18%. The results are satisfactory, and the proposed approach can be further extended to the software industry for requirements verification.

Enhancing operational efficiency of insurance companies: a fuzzy time series approach to loss ratio forecasting in the Egyptian market

ABSTRACT This article analyses the crucial significance of loss ratio (LR) evaluation in assessing the operational efficiency of insurance companies. Fuzzy time series (FTS) models are effective in modelling non-linear with uncertainties and providing adequate performance with limited data availability, which overcomes the shortage of conventional models used in Egyptian insurance market literature. This paper aims to introduce a comprehensive analysis of LR of the property and casualty sectors. Moreover, review and compare various forecasting FTS techniques to discuss the effectiveness of using FTS methods in forecasting LR. The suggested method is based on results from tests done with four models that had Huarng 465 partitions and interval length parts of 5, 10, 50, and 100. The results show LR prediction improved significantly. Yu and Cheng’s models using a Huarng 465 partition for training data and the Yu model with 100 partitions for testing data had high accuracy and low error.

A Method for Determining the Fundamental Site Period and the Average Shear Wave Velocity

The soil–structure interaction plays a crucial role in determining the displacement and internal forces of multi-story buildings subjected to strong ground motion. One of the critical dynamic characteristics influencing soil–structure interaction is the fundamental site period and the average shear wave velocity associated with it. This study introduces an original equation to determine these parameters. In addition, for the first time in the literature, the version of the Rayleigh method used for finding the fundamental periods of buildings is used to find the fundamental site period. The soil is modeled as an equivalent shear beam to obtain the proposed equation. The peak displacement is obtained by acting the soil mass as an external load on the equivalent shear beam. For single-layer soil, the fundamental site period is proportional to the square root of the peak displacement of the equivalent shear beam. The least squares method generalizes the proposed relation for single-layer soils to multi-layer soil profiles. Modified Finite element Transfer matrix method is used for calibration in the least squares method. The equations used in the literature and earthquake codes for determining the fundamental site period and average shear velocity are tested on various examples, and it is shown that the method proposed in this study, along with the Rayleigh method, gives better results than these equations. The performances of these two methods and the five commonly used equations are tested and compared on different soil profiles. Transfer functions, Finite Element Method (SAP200) and Modified Finite Element Transfer Matrix Method are used for verification. For all soil profiles, the results obtained from the transfer function, Finite Element Method (SAP200) and Modified Finite Element Transfer Matrix Method are found to be in agreement. The true percent relative error found in the results obtained with the proposed method is 4.47%.

Additive Manufacturing of Ceramic Reference Spheres by Stereolithography (SLA)

Additive Manufacturing (AM) is advancing technologically towards the production of components for high-demand mechanical applications with stringent dimensional accuracy, leveraging metallic and ceramic raw materials. The AM process for ceramic components, known as Ultraviolet Laser Stereolithography (SLA), enables the fabrication of unique parts or small batches without substantial investments in molds and dies, and avoids the problems associated with traditional manufacturing, which involves multiple stages and final machining for precision. This study addresses the need to produce reference elements or targets for metrological applications, including verification, adjustment, or calibration of 3D scanners and mid- to high-range optical sensors. Precision spheres are a primary geometry in this context due to their straightforward mathematical definition, facilitating rapid and accurate error detection in equipment. Our objective is to exploit this novel SLA process along with the advantageous optical properties of technical ceramics (such as being white, matte, lightweight, and corrosion-resistant) to materialize these reference objects. Specifically, this work involves the fabrication of alumina hemispheres using SLA. The manufacturing process incorporates four design variables (wall thickness, support shape, fill type, and orientation) and one manufacturing variable (the arrangement of spheres on the printing tray). To evaluate the impact of the design variables, dimensional and geometric parameters (GD&T), including diameters, form errors, and their distribution on the surface of the sphere, have been characterized. These measurements are conducted with high accuracy using a Coordinate Measuring Machine (CMM). The study also examines the influence of these variables in the dimensional and geometric accuracy of the spheres. Correlations between various parameters were identified, specifically highlighting critical factors affecting process precision, such as the position of the piece on the print tray and the wall thickness value. The smallest diameter errors were recorded at the outermost positions of the tray (rear and front), while the smallest shape errors were found at the central position, in both cases with errors in the range of tens of micrometers. In any case, the smallest deformations were observed with the highest wall thickness (2 mm).

An adaptive detection framework based on artificial immune for IoT intrusion detection system

Given the continual evolution of new network attack methodologies, defenders face the imperative of constantly upgrading security defenses. Current security technologies, albeit effective against known threats, often fall short in handling the intricacies of diverse and novel attacks. Artificial immunity-based network anomaly detection offers a promising avenue by dynamically adapting to evolving threats. However, prevailing algorithms in this domain suffer from low detection rates, limited adaptability, and extended detector generation times. This study aims to tackle these challenges by introducing a high-efficiency network anomaly detection framework, emphasizing both high-dimensional feature selection and adaptive detector generation. Our approach begins with an enhanced dual-module hybrid high-dimensional feature selection method, leveraging evolutionary principles. Furthermore, we introduce a self-sample clustering algorithm based on fuzzy clustering during the tolerance stage, enhancing detector tolerance efficiency. Additionally, an adaptive detector generation scheme is devised. It divides the non-boundary sub-population based on non-self differences and evolution, while employing the red fox optimization algorithm in the boundary region. This adaptive approach dynamically adjusts detector positions and radii to derive optimal detectors. Through comprehensive validation using well-established IoT and network anomaly datasets, our proposed artificial immunity-based IoT intrusion detection framework exhibits superior performance. It achieves higher classification accuracy and lower error rates compared to current state-of-the-art machine learning and artificial immunity algorithms.

Internet of things and optimized knn based intelligent transportation system for traffic flow prediction in smart cities

The rapid expansion of urban areas and the increasing number of vehicles on the road have resulted in accidents, traffic congestion, economic repercussions, environmental deterioration, and excessive fuel consumption. A dependable traffic management system is necessary to anticipate and regulate urban traffic patterns. Traffic forecast aids in the prevention of traffic issues. Urban traffic predictions often utilise historical and current traffic flow data to forecast road conditions. This article presents a traffic flow prediction system that utilises the Internet of Things (IoT), machine learning, and feature selection. Internet of Things (IoT) devices located on highways or within cars gather sensor data in real-time. The input data set comprises both real-time Internet of Things (IoT) data and historical traffic statistics. The input data is stored in a centralized cloud. The data is subjected to preprocessing in order to eliminate any unwanted interference and identify any exceptional values. The accuracy and root mean square error are contingent upon the process of feature selection. Particle swarm optimization identifies and extracts crucial features from input data. The classification model is constructed using K Nearest Neighbor, Multi layer Perceptron, and Bayes network approaches. The UCI traffic data is used for conducting experiments. The dataset has 47 attributes and 2102 occurrences. The accuracy of traffic flow prediction using PSO KNN is 96 %. The PSO KNN algorithm achieved a Mean Square Error (MSE) of 0.00289 and a Root Mean Square Error (RMSE) of 0.0595.

Photoidentification as a potential tool for individual recognition of Dendropsophus elegans (Anura: Hylidae)

Marking-recapture techniques are essential in the study of wild animal populations, and the use of accessible, easy to apply, and less invasive methods is preferable. Several studies have reported negative effects of more invasive or inaccurate techniques when studying animal behaviour and survival, which raised the alert for the encouragement of affordable and less invasive techniques, such as photoidentification using natural markings. The aim of this study was to compare the performance of the promising use of photoidentification with the application of visible elastomers in the field, verifying application time and presence of stress behaviours in the model anuran species Dendropsophus elegans (Wied-Neuwied, 1824), which has an evident dorsal spotting pattern. In addition, we checked the subsequent performance of individual recognition by software-assisted photoidentification, code recognition of visible elastomers, and manual photoidentification, considering accuracy (error types) and recognition time. We found that software photoidentification had a similar performance to the manual one and elastomer combination for identifying recaptures, but its execution time was the shortest among the three and it was more accurate. There was no association between the presence of stress behaviours and any technique applied in the field. These results suggest that software-assisted photoidentification combined with a field protocol may be an efficient technique in the study of this species, and its applicability may be extended to other species with natural markings.

Optimal Spectral Performance on Pediatric Photon-Counting CT: Investigating Phantom-Based Size-Dependent kV Selection for Spectral Body Imaging.

The comprehensive evaluation of kV selection on photon-counting computed tomography (PCCT) has yet to be performed. The aim of the study is to evaluate and determine the optimal kV options for variable pediatric body sizes on the PCCT unit. In this study, 4 phantoms of variable sizes were utilized to represent abdomens of newborn, 5-year-old, 10-year-old, and adult-sized pediatric patients. One solid water and 4 solid iodine inserts with known concentrations (2, 5, 10, and 15 mg I/mL) were inserted into phantoms. Each phantom setting was scanned on a PCCT system (Siemens Alpha) with 4 kV options (70 and 90 kV under Quantum Mode, 120 and 140 kV under QuantumPlus Mode) and clinical dual-source (3.0 pitch) protocol. For each phantom setting, radiation dose (CTDIvol) was determined by clinical dose settings and matched for all kV acquisitions. Sixty percent clinical dose images were also acquired. Reconstruction was matched across all acquisitions using Qr40 kernel and QIR level 3. Virtual monoenergetic images (VMIs) between 40 and 80 keV with 10 keV interval were generated on the scanner. Low-energy and high-energy images were reconstructed from each scan and subsequently used to generate an iodine map (IM) using an image-based 2-material decomposition method. Image noise of VMIs from each kV acquisition was calculated and compared between kV options. Absolute percent error (APE) of iodine CT number accuracy in VMIs was calculated and compared. Root mean square error (RMSE) and bias of iodine quantification from IMs were compared across kV options. At the newborn size and 50 keV VMI, noise is lower at low kV acquisitions (70 kV: 10.5 HU, 90 kV: 10.4 HU), compared with high kV acquisitions (120 kV: 13.8 HU, 140 kV: 13.9 HU). At the newborn size and 70 keV VMI, the image noise from different kV options is comparable (9.4 HU for 70 kV, 8.9 HU for 90 kV, 9.7 HU for 120 kV, 10.2 HU for 140 kV). For APE of VMI, high kV (120 or 140 kV) performed overall better than low kV (70 or 90 kV). At the 5-year-old size, APE of 90 kV (median: 3.6%) is significantly higher (P < 0.001, Kruskal-Wallis rank sum test with Bonferroni correction) than 140 kV (median: 1.6%). At adult size, APE of 70 kV (median: 18.0%) is significantly higher (P < 0.0001, Kruskal-Wallis rank sum test with Bonferroni correction) than 120 kV (median: 1.4%) or 140 kV (median: 0.8%). The high kV also demonstrated lower RMSE and bias than the low kV across all controlled conditions. At 10-year-old size, RMSE and bias of 120 kV are 1.4 and 0.2 mg I/mL, whereas those from 70 kV are 1.9 and 0.8 mg I/mL. The high kV options (120 or 140 kV) on the PCCT unit demonstrated overall better performance than the low kV options (70 or 90 kV), in terms of image quality of VMIs and IMs. Our results recommend the use of high kV for general body imaging on the PCCT.

Differential Signal-Amplitude-Modulated Multi-Beam Remote Optical Touch Based on Grating Antenna.

As screen sizes are becoming larger and larger, exceeding human physical limitations for direct interaction via touching, remote control is inevitable. However, among the current solutions, inertial gyroscopes are susceptible to positional inaccuracies, and gesture recognition is limited by cameras' focus depths and viewing angles. Provided that the issue of ghost points can be effectively addressed, grating antenna light-trapping technology is an ideal candidate for multipoint inputs. Therefore, we propose a differential amplitude modulation scheme for grating antenna-based multi-beam optical touch, which can recognize different incidence points. The amplitude of the incident beams was first coded with different pulse widths. Then, following the capture of incident beams by the grating antenna and their conversion into electrical currents by the aligned detector arrays, the incident points of the individual beams were recognized and differentiated. The scheme was successfully verified on an 18-inch screen, where two-point optical touch with a position accuracy error of under 3 mm and a response time of less than 7 ms under a modulation frequency of 10 kHz on both incident beams was achieved. This work demonstrates a practical method to achieve remote multi-point touch, which can make digital mice more accurately represent the users' pointing directions by obeying the natural three-point one-line aiming rule instantaneously.

Assessing the compressive strength of eco-friendly concrete made with rice husk ash: A hybrid artificial intelligence-aided technique

Providing an eco-friendly replacement for cement in cementitious materials mitigates not only its consumption but also the environmental effects such as extraction of natural resources and greenhouse gas emissions resulting from its production. In light of this, the utilization of rice husk as a by-product of rice fields in the form of rice husk ash (RHA) as a partial substitute for cement, in addition to solving the problem of its disposal due to its silica content and high pozzolanic reaction, has been considered in the construction of sustainable concretes. This study seeks to move from the traditional laboratory-based methods for the examination of compressive strength (fc′) of concretes containing RHA towards artificial intelligence-based techniques by developing a novel hybrid model by integrating biogeography-based optimization (BBO) technique with artificial neural network (ANN). Besides, the performance of the hybrid ANN-BBO model was assessed with the single ANN model. To this end, the models were developed on an extensive dataset including 1276 data records collected from the literature. The efficiency of the models was evaluated by performance criteria and error histogram, and also cross-validation was utilized to avoid overfitting problems and generalize the models. Results demonstrated that the hybrid ANN-BBO model outperformed the single ANN model with high accuracy and minimum error values. Error histograms revealed that over 90 % of errors in the ANN-BBO model occurred in the range (−4 MPa, 4 MPa], while this range for the ANN model fell within (−6 MPa, 6 MPa]. Analyzing the influence of input parameters showed that more than 70 % of the contribution rate in the fc′ of RHA concrete belonged to four influential parameters of specimen age, cement, rice husk ash, and water, respectively. Eventually, to validate the proposed ANN-BBO model, a comparison was made with recent studies, in which the performance criteria revealed that the proposed model based on a more extensive dataset outperformed the models in previous studies.