Smart Door Lock System Using Facial Recognition

In the field of sustainable mobility, this study highlights the importance of using machine learning for predictive modeling based on real traffic data collected from instrumented bicycles. The advent of advanced technologies like sustainable mobility apps, sensors, and advanced data analysis methods led to the ability to collect data from various sources, which enabled researchers to estimate battery state of charge (SOC) accurately. Most current research uses them in the lab experiments for data collection. In this work, we use real-time sensors data to construct data-driven models for lithium-ion battery SOC estimation. This research integrates both electric bicycle battery, environmental and route variables to achieve the following goals: (1) Collect a multimodal data set including operational, topography, vehicle, and external variables, (2) Preprocess data obtained from sensors installed on the electric bicycle battery, (3) Create models of lithium-ion battery SOC based on electric bicycle battery and environmental variables, and (4) Assess data-driven models and compare their performance for lithium-ion battery SOC with high accuracy. To achieve that, we conducted a real study to predict the Remaining Useful Life (RUL), as a measure of state of charge, of electric bicycle battery. The study was carried out on a 15 km cycle route in Medellín, Colombia, for 28 days. To estimate the RUL, we used four different machine learning algorithms: Long Short-Term Memory (LSTM), Support Vector Regression (SVR), AdaBoost, and Gradient Boost. Notably, data preprocessing techniques played a pivotal role, with a particular focus on smoothing sensor data using Convolutional Neural Networks (CNN). The results showed a significant improvement in prediction accuracy when using data preprocessing, confirming its importance in improving model performance. Furthermore, the comparison of network performance facilitated the selection of the most effective model for the test data. This study underscores the value of using real-world data to develop and validate predictive models in the pursuit of sustainable mobility solutions, and highlights the critical role of data-driven methodologies in addressing today's urban transportation challenges.

Hybrid renewable energy resource (RER) microgrids offer a sustainable solution to integrating multiple energy sources like solar and wind into modern power systems. However, managing the intermittent nature of these resources presents significant challenges in maintaining stable and efficient power distribution. This study explores innovative control approaches for optimizing power management in hybrid RER microgrids. By utilizing advanced algorithms, including real-time optimization and machine learning techniques, the proposed framework ensures efficient energy distribution, reduces dependency on the grid, and enhances system stability. The research focuses on integrating these control strategies to balance energy supply and demand while maximizing the utilization of renewable energy sources. Simulation results demonstrate the effectiveness of the proposed methods in improving system efficiency and reliability under varying environmental conditions, contributing to the broader adoption of hybrid RER microgrids.

The precise identification of hydrocarbon-rich zones is crucial for optimizing exploration and production processes in the oil industry. Magnetotelluric (MT) surveys play a fundamental role in mapping subsurface geological structures. This study presents a novel methodology for automatically delineating resistivity contrasts in MT models by employing advanced machine learning and computer vision techniques. This approach commences with data augmentation to enhance the diversity and volume of resistivity data. Subsequently, a bilateral filter was applied to reduce noise while preserving edge details within the resistivity images. To further improve image contrast and highlight significant resistivity variations, contrast-limited adaptive histogram equalization (CLAHE) was employed. Finally, k-means clustering was utilized to segment the resistivity data into distinct groups based on resistivity values, enabling the identification of color features in different centroids. This facilitated the detection of regions with significant resistivity contrasts in the reservoir. From the clustered images, color masks were generated to visually differentiate the groups and calculate the area and proportion of each group within the pictures. Key features extracted from resistivity profiles were used to train unsupervised learning models capable of generalizing across different geological settings. The proposed methodology improves the accuracy of detecting zones with oil potential and offers scalable applicability to different datasets with minimal retraining, applicable to different subsurface environments. Ultimately, this study seeks to improve the efficiency of petroleum exploration by providing a high-precision automated framework with segmentation and contrast delineation for resistivity analysis, integrating advanced image processing and machine learning techniques. During initial analyses using only k-means, the resulting optimal value of the silhouette coefficient K was 2. After using bilateral filtering together with contrast-limited adaptive histogram equalization (CLAHE) and validation by an expert, the results were more representative, and six clusters were identified. Ultimately, this study seeks to improve the efficiency of petroleum exploration by providing a high-precision automated framework with segmentation and contrast delineation for resistivity analysis, integrating advanced image processing and machine learning techniques.

Gait monitoring is considered as a significant marker of disability, injury, and gait symmetry. The goal of this study is to develop a real-time consumer health monitoring system based on IoT sensors and Machine learning technique in order to detect health abnormalities such as, stroke onset. The proposed consumer stroke prediction system consists of IoT based gait monitoring sensors, real-time vital sign monitoring and machine learning based disease prediction model to predict the disordered gait and the healthy gait. This study will be useful for post-stroke gait coordination for rehabilitation and consumer health monitoring service.

Real-time image processing and object detection techniques have a great potential to be applied in digital assistive tools for the blind and visually impaired persons. In this paper, a ...

Real-time signal and image processing techniques have recently gained popularity, due to its importance in many potential applications such as communications, robotic vision, military target detection, and part inspection. This paper presents a simple scheme for thresholding signals and grayscale images dynamically and in real-time. The scheme utilizes the transfer function of a simple low-pass filter and an advanced zero-order hold device where the input is the raw signal or the image' s grayscale values, and the output is the corresponding dynamic thresholds. The scheme features high reliability of the results and is effectively applicable for signals and images with different background illumination and different shape, such as noisy step signals, ramp signals, or their combination. Software and hardware experiments have confirmed the method' s viability.

The main aim of this work is to classify the plants according to its medicinal usage using real time processor Raspberry pi with the image of the leaf, as leaves play a major role for the classification of plants. Image of leaf is captured using digital camera of high resolution and is stored in memory card. Feature and pattern of the leaf are extracted using image processing technique which is then interfaced with Raspberry pi. Raspberry pi is stored with a program to classify the input image based on texture parameters of leaves. The quality and authenticity of the leaves are to be ensured for the preparation of herbal medicines. The image of medicinal plant leaves are thoroughly screened analyzed and compared with the database stored in Raspberry pi to give the correct measures of the texture of leaf with which we can find category to which the leaf belongs to. By applying advanced technique of image processing interfaced in Raspberry pi and utilizing the capabilities of the recent advanced computing and data/image storage facilities and the use of computer techniques, the above mentioned parameters are analyzed and compared with the database and classification of plant species is identified.

Feasibility of a real-time hand hygiene notification machine learning system in outpatient clinics

Real-time big data processing for anomaly detection: A Survey

This study presents a novel foundational Regional Dwell Time (RDT) Model tailored for regional railway networks, addressing the gap of existing dwell time models specifically designed for this type of railway. Using video-based observations from Victoria, Australia, empirical data were collected and analysed to develop a statistical regression model integrating passenger flow dynamics and operational constants such as door operation and train dispatch procedures. The proposed RDT Model was calibrated and validated against 398 regional train services at two stations. The model was compared with established statistical dwell time models in the literature. The RDT Model demonstrated superior predictive accuracy, with significantly lower error metrics (RMSE, MAPE, MAE) compared to alternative models. The study emphasizes the significance of operational time in dwell time estimations and highlights opportunities for reducing dwell time through improved operational procedures. Findings suggest the RDT Model's adaptability to various other regional railways, provided operational constants are recalibrated. This model serves as a proof-of-concept and a novel framework for dwell time modelling in peri‑urban regional railways, with broader applicability contingent on future multi-corridor validation. Future research on regional railways should aim to build on this type of model and explore real-time data integration and machine learning techniques to enhance predictive capabilities and network efficiency.

Body slip angle is one of the most important information for vehicle motion control; as specific sensors for body slip angle measurement are expensive, it is necessary to investigate estimation methods using existing popular sensors such as gyro sensor, encoder, camera, etc. For EV (electric vehicle), in particular, the motor response is several milliseconds which enables high performance control with short control period; fast signal feedback is consequently desired. Nevertheless, the sampling rate of a normal camera is much slower compared with other kinds of onboard sensors and the time delay caused by image processing cannot be neglected. In this paper, the two problems are solved using a multi-rate Kalman filter with measurement delay included; the estimated body slip angle can be updated every 1 ms. First of all, vehicle model and visual model are explained followed with experimental setup introduction; then, real-time image processing techniques are briefly introduced; and then, single-rate and multi-rate Kalman filters considering time delay are designed to estimate body slip angle; finally, conclusion and further works are presented.

Diffraction limit is usually a thorny problem in an optical inspection system. In this investigation, a model-based deconvolution technique was developed to recover diffraction-limited images, where images with sizes smaller than the diffraction limit could be recognized. Experiments were carried out with a traditional microscope at 200× magnification coupled with a halogen light source for a series of line width samples. The point spread function of the imaging optics was first obtained from an estimated model and then combined with a nonlinear deconvolution algorithm to calculate the full width at half maximum and reconstruct the line widths. Experimental results indicate that a measurement error below one pixel size of the measurement system is achievable. Accordingly, the target of nanoscale line width inspection based on a low cost and real-time image processing technique can be fulfilled, which greatly increases the ability of nanoscaling on optical microscopes.

We show that negative dielectrophoresis (DEP) spectroscopy is an effective transduction mechanism of a biosensor for the diagnosis and prognosis of pancreatic cancer using the biomarker CA 19-9. A substantial change in the negative DEP force applied to functionalized polystyrene microspheres (PM) was observed with respect to both the concentration level of the pancreatic cancer biomarker CA 19-9 and the frequency of the electric field produced by a pearl shaped interdigitated gold micro-electrode. The velocity of repulsion of a set of PM functionalized to a monoclonal antibody to CA 19-9 was calculated for several concentration cutoff levels of CA 19-9, including 0 U/mL and 37 U/mL, at the frequency range from 0.5 to 2 MHz. The velocity of repulsion of the PM from the electrode was determined using a side illumination and an automated software using a real-time image processing technique that captures the Mie scattering from the PM. Since negative DEP spectroscopy is an effective transduction mechanism for the detection of the cutoff levels of CA 19-9, it has the potential to be used in the early stage diagnosis and in the prognosis of pancreatic cancer.

Although software implementations of different image processing techniques are suitable for general-purpose use, in order to meet the real time requirements, an image processing technique needs to be realized in hardware. Field Programmable Gate Arrays (FPGAs) have many benefits in applications that include digital signal acquisition, but also processing of large data, especially in real time. Mainly due to the ever-decreasing cost and re-configurability, FPGAs have also found its place in digital signal processing (DSP). Xilinx System Generator is a tool from Xilinx that enables the Mathworks Simulink models to be adapted for FPGA design. For comparative study on several levels in edge detection, CT image of a brain with a tumor is used. Performances of gradient based edge detectors - Robert, Prewitt and Sobel were compared. Even from just visual analysis of results, it can be seen that Prewitt and Sobel methods give better results than Robert method. In contrast, the calculation of Robert operator is simpler in comparison to the other operators and occupies less resources, since only adder-subtractor logic is sufficient to detect the edges. As the implemented algorithms could be part of more complex systems for tumor detection, the design architecture used in this paper can be extended to be used in very complex real time image processing techniques.

Overview of the use of machine learning algorithms for real-time occupational risk assessment and monitoring Machine learning (ML) is an area of artificial intelligence dealing with algorithms that can modify their parameters based on processed data and perform tasks without explicit instructions. The development of artificial intelligence systems and Internet of Things technologies is resulting in ML algorithms being increasingly used in various sectors of the economy, including occupational risk assessment and monitoring. This article provides a review of the scientific literature in this field, which has identified the sectors and types of work environments in which these algorithms are used, defined the data used to learn ML algorithms and identified the most commonly used methods. The potential of UM algorithms to automate hazard identification and occupational risk assessment processes in real time is also presented. Keywords: machine learning, occupational risk assessment, real time, artificial intelligence