Automatic Monitoring Research Articles

First high-resolution vertical-looking radar for long-term automatic observation of high-flying insects in Asia.

The increasing occurrence of migrant insect pests poses a serious threat to the sustainability of agriculture and to food security. Continuous monitoring of high-flying insects plays a crucial role in developing effective pest management strategies and implementing successful control measures. The present study established vertical-looking radar (VLR) monitoring of insects in Henan Province, China, with a unit incorporating up-to-date high frequency digitization technology and rapid target-finding procedures. This radar produced detailed information on target identification (size, shape, wingbeat frequency) and flight behavior (flight time, height, track speed, track direction, body alignment, and climb rate) for insects flying at altitudes of from 70 m to 1810 m above the ground. The lowest detection range for insects is lower than that (150 m) normal in previous VLR systems. The VLR-inferred tracking of small insects could also provide accurate estimates of wind velocity. The VLR achieved long-term automatic observation of high-flying insects for the first time in Asia. This provided a unique tool for automatic long-term monitoring of high-flying insects to help to answer some basic scientific questions, such as the impacts of climate change on insect populations, and provide surveillance information for insect pest control in this region. The three-step target identification method and the performance calibration protocol for the VLR established in this study are both straightforward and reliable. These methods can be easily implemented and adapted for use in other settings, making them valuable tools for enhancing radar-based entomological research and monitoring. © 2025 Society of Chemical Industry.

Face Recognition-Based Attendance System Using Machine Learning

The attendance marking process is an important aspect in academic and corporate settings. Inefficient, inexact, and forged attendance is the risk associated with traditional processes. This article suggests an automated system based on machine learning and face recognition technology for increased accuracy, security, and efficiency. The system uses OpenCV, Deep Face, and Flask to acquire, process, and verify user identities. The system follows the client- server model for implementation, allowing for real-time processing and safe data storage. The paper discusses the approach, experimental findings, and system performance measurement, making it evident that the system is effective in automatic attendance monitoring. Keywords Face Recognition, Deep Learning, Machine Learning, Attendance System, OpenCV, Flask, Deep Face.

AI-Powered Real-Time Runway Safety: UAV-Based Video Analysis with ICSO-Enhanced Deep Learning

In the aviation sector, ensuring safe landings while prioritizing the safety of runways is crucial to prevent accidents and incidents during the landing phase of flights. However, many studies analyzing unsafe events, such as runway cracks or inadequate friction, often fail to quantify their impacts on flight safety during landing. In airport pavement management systems (APMS), the condition of the runway surface is a critical factor in ensuring the operational safety of aircraft during take-off and landing. Therefore, it is essential to provide pilots with reports on runway conditions, including measurements of surface performance, to support informed decision-making. To tackle these challenges, we propose a real-time automatic monitoring system for runway safety utilizing video analysis. Specifically, we employ a time-series analysis approach using the improved chameleon swarm optimization (ICSO) algorithm to mine runway surface characteristics from real-time video data captured by unmanned aerial vehicles (UAVs). Subsequently, we introduce the fuzzy reinforced polynomial neural network (FR-PNN) to detect risks in runway surface characteristics, enabling automatic monitoring to enhance the safety of aircraft landings. Finally, the effectiveness of the proposed system is validated using real-time videos obtained from Bechyne military airport, located in Bohemia. This system aims to improve runway safety by providing timely and accurate assessments of runway conditions, thereby facilitating safer landings for aircraft.

Automatic monitoring of activity intensity in a chicken flock using a computer vision-based background image subtraction technique: An experimental infection study with fowl adenovirus

Automatic monitoring of activity intensity in a chicken flock using a computer vision-based background image subtraction technique: An experimental infection study with fowl adenovirus

Computational polarized holography for automatic monitoring of microplastics in scattering aquatic environments

Computational polarized holography for automatic monitoring of microplastics in scattering aquatic environments

Optimizing black cattle tracking in complex open ranch environments using YOLOv8 embedded multi-camera system

Monitoring the daily activity levels of black cattle is a crucial aspect of their well-being. The rapid advancements in artificial intelligence have transformed computer vision applications, including object detection, segmentation, and tracking. This has led to more effective and precise monitoring techniques for livestock. In modern cattle farms, video monitoring is essential for analyzing behavior, evaluating health, and predicting estrus events in precision farming. This paper introduces the novel Customized Multi-Camera Multi-Cattle Tracking (MCMCT) system. This unique approach uses four cameras to overcome the challenges of detecting and tracking black cattle in complex open ranch environments. The MCMCT system enhances a tracking-by-detection model with the YOLO v8 segmentation model as the detection backbone network to develop a precision black cattle monitoring system. Single-camera setups in real-world datasets of our open ranches, covering 23.3 m x 20 m with 55 cattle, have limitations in capturing all necessary details. Therefore, a multi-camera solution provides better coverage and more accurate behavior detection of cattle. The effectiveness of the MCMCT system is demonstrated through experimental results, with the YOLOv8-MCMCT system achieving an average Multi-Object Tracking Accuracy (MOTA) of 95.61% across 10 cases of 4 cameras at a processing speed of 30 frames per second. This high accuracy is a testament to the performance of the proposed MCMCT system. Additionally, integrating the Segment Anything Model (SAM) with YOLOv8 enhances the system’s capability by automating cattle mask region extraction, reducing the need for manual labeling. Comparative analysis with state-of-the-art deep learning-based tracking methods, including Bot-sort, Byte-track, and OC-sort, further highlights the MCMCT’s performance in multi-cattle tracking within complex natural scenes. The advanced algorithms and capabilities of the MCMCT system make it a valuable tool for non-contact automatic livestock monitoring in precision cattle farming. Its adaptability ensures effective performance across varied ranch environments without extensive retraining. This research significantly contributes to livestock monitoring, offering a robust solution for tracking black cattle and enhancing overall agricultural efficiency and management.

Aircraft Detection and Registration Number Recognition System with YOLO and OCR

Recently, large airports are promoting a smart airport business that introduces robots and IoT using artificial intelligence and big data, which are major technologies in the fourth industrial era. Various artificial intelligence technologies are applied not only to customer convenience but also to airport security and control, and in particular, video monitoring and analysis technologies such as missing children are introduced through intelligent CCTV. In this paper, we propose a system for recognizing aircraft detection and registration number (tail number) taking off and landing on the runway using YOLO, a deep learning object detection model, and character recognition technology (OCR). It acquires aircraft data through cameras installed on the ground and learns it with the fastest YOLO model among deep learning object detection models to automatically detect aircraft in the airport and its registration number area. And the registration number recognized the result detected by YOLO using the OCR algorithm through the image preprocessing process. This study conducted data acquisition and real-time detection tests at Taean Airfield (RKTA) at Hanseo University in Korea, and real-time aircraft detection was more than 90% and registration number recognition was more than 80%. Through this system, information on the direction, location, model, and registration number of the aircraft can be acquired, confirming its utility as an automatic ground monitoring system for small airports in the future.

Automatic measurement of surface runoff at the plot scale

ABSTRACT Although the hydrological importance of water infiltration into the soil and surface runoff is well known, manual measurement of these processes, especially at the plot and hillslope scales, is laborious and requires a trained team. This study aimed to develop prototypes with the tipping bucket principle for the automatic monitoring of surface runoff at a plot scale (14.5 × 3 m). In total, 18 prototypes were built and installed in an experiment with different soil covers. Aspects of the construction, calibration, operation and maintenance of the prototypes were reported in this study. Prototypes showed a satisfactory ability to monitor surface runoff with flow rates up to ~80 L min -1 . The devices have an adjustable resolution of up to 5 L pulse -1 , and they are inexpensive, easy to construct and simple to maintain. Although designed for a specific plot size (43.5 m 2 ), the equipment can be adapted to smaller or larger plots by adjusting the volume of the tipping chambers. The proposed systems allow greater practicality in conducting experiments to monitor water loss through surface runoff at the plot scale and provide more detailed recordings of information.

Agreement of Automated Oscillometric Blood Pressure Measurement Device with the Manual Mercury Sphygmomanometer

Abstract The usage of the manual mercury sphygmomanometer to measure blood pressure (BP) is universally regarded as the gold standard, but its use has decreased globally due to concerns about potential mercury toxicity. Automated BP devices have become more popular, though their accuracy in detecting hypertension hasn’t been studied extensively. This cross-sectional study was conducted on 301 adult female students of the University of Delhi with the aim to assess the degree of agreement between standard manual mercury sphygmomanometer and automated BP monitor (Omron JPN1). Three consecutive BP readings were taken on the left arm by a certified medical doctor using both instruments. Bland Altman plots were created to assess the degree of agreement between the two BP devices and to test the accuracy of automated BP monitor in detecting hypertensive adults. The automated BP monitor underestimated the systolic BP measurement by 4 mmHg and diastolic BP measurement by 7 mmHg as determined by Bland Altman plot. Based on international validation protocols, a mean difference of < 5 ± 8 mmHg was established to assess the level of agreement between the two devices. The automatic BP monitor did not show complete agreement with the “gold standard” mercury sphygmomanometer. With growing restrictions on the use of mercury sphygmomanometers, its replacement with an equivalent mercury-free validated device is urgently needed.

Automatic Monitoring in Construction Projects: Scientometric Analysis and Visualization of Research Activities

Abstract This paper presents a comprehensive scientific analysis of research on automatic monitoring in construction projects. Through a methodical examination of 857 bibliographic records from three databases, we find important trends, novel themes, and key research areas in this field. Our findings reveal that machine learning (ML), building information modelling (BIM), and deep learning are the most the most popular techniques for automatic monitoring. Furthermore, we identify construction technologies and artificial intelligence (AI) as the primary research foci. So, the study uncovers collaboration patterns among researchers and institutions, highlighting key players and their contributions, identifies research gaps and challenges, such as the need for integrating AI, big data, and cloud computing into construction project monitoring, and proposes future research directions to address these challenges and enhance the effectiveness of automatic monitoring systems. By providing a systematic review and insightful analysis, this study contributes to the advancement of construction project monitoring. It offers valuable insights for researchers, practitioners, and policymakers to foster innovation, improve project performance, and ensure sustainable construction practices.

Evaluation of an aquatic liverwort and terrestrial moss as biomonitors of heavy metals associated with particulate matter

In this study, the capacity of the aquatic liverwort Ricciocarpus natans L. and the terrestrial moss Entodon serrulatus Mitt. as biomonitors of heavy metals associated with particulate matter from a highly polluted urban area was evaluated, and concentrations in moss tissues were correlated with concentrations of PM10 and PM2.5 present in the atmosphere. The two species were exposed by the moss bag technique to the pollution of the Toluca Valley Metropolitan Area (TVMA) for two periods of 6 months, using the sites of the Automatic Atmospheric Monitoring Network of the Government of the State of Mexico, and were subsequently analyzed using elemental and structural characterization techniques. The results show that mainly the functional groups -OH and -NH, N-H and C-O on the surface of the liverwort and moss participate in the adsorption of heavy elements. The average enrichment factors of Cd and Pb show to be highly enriched (> 10) in the study area while chromium is not enriched (< 2). The statistical results indicate a temporary variation in the concentration of metals and particles in the atmosphere, where there is a lower concentration of these pollutants in the rainy and dry-cold season and a higher concentration in the dry-hot season and a possible association of Cr and Cd with PM10 and PM2.5. In addition, except for Cr, both species accumulate the metals associated with airborne particulate matter at equivalent levels. There is strong association between PM2.5 and PM10 particles and between the metals Cr-Pb-Fe in R. natans and between PM2.5-PM10 and Fe and between Cd-Cr-Pb in E. serrulatus and these pollutants are mainly associated with sampling sites with the highest concentrations of metals in the TVMA. Although terrestrial moss showed slightly better characteristics than aquatic liverwort as a biomonitor of heavy metals associated with atmospheric particles, these differences were not statistically significant for all metals, so both species could be useful for heavy metal biomonitoring in highly polluted urban areas.

Identification of particulate matter (PM10 and PM2.5) sources using bivariate polar plots and k-means clustering in a South American megacity: Metropolitan Area of Lima-Callao, Peru.

The identification of different air pollution sources is essential to effectively control atmospheric pollution, particularly in megacities of emerging countries with rapid economic development, such as the Metropolitan Area of Lima-Callao (MALC). The objective of this research was to identify the main sources of particulate matter pollution by applying bivariate polar plots and the k-means clustering algorithm. These statistical techniques were applied to hourly in situ data of four variables collected over a 5-year period (2015-2019) by the Automatic Air Quality Monitoring Network of the MALC: wind direction, wind speed, PM10, and PM2.5 concentrations. Average PM10 concentrations ranged from 34μgm-3 (CDM station) to 126.7μgm-3 (VMT station), while average PM2.5 concentrations ranged from 16.8μgm-3 (CDM station) to 41.2μgm-3 (ATE station). The diurnal variation of PM presented two peaks, one in the morning (from 0800 to 1000h) and the other at night (from 1900 to 2300h), with the highest concentrations of PM10 recorded at the ATE (0800h: 155.8μgm-3) and VMT (2100h: 154.6μgm-3) stations, and PM2.5 at the ATE station (0800h: 60.3μgm-3 and 2300h: 37.5μgm-3). The results showed that the contributions of PM10 are directly related to emissions from industrial activities, automotive fleet, construction, demolition, wind erosion, and the suspension and resuspension of particulates from unpaved roads. Meanwhile, high concentrations of PM2.5 are mainly attributed to vehicle exhaust emissions, industrial emissions, secondary particulate formation, and drag by the action of the winds. The major source of particulate matter contamination is the vehicle fleet, and within this, automobiles, station wagons, combi vans, and 2 and 3-wheel motorcycles are those that have the greatest contribution. These results were supported by non-parametric statistical tests such as Kruskal-Wallis and Mann-Whitney U and validated by the conditional bivariate probability function. The findings of this work may help to implement pollution prevention and control strategies in the future within this South American megacity.

Automatic monitoring method for seismic response of building structures and equipment based on indoor surveillance cameras

Automatic monitoring method for seismic response of building structures and equipment based on indoor surveillance cameras

Automated process assessment of primary healthcare for hyperlipidemia: preliminary findings and implications form Anhui, China

BackgroundPrimary healthcare (PHC) plays a key role in hyperlipidemia (HL) management yet lacks adequate monitoring and feedback. This study aims at identifying pragmatic measures out from routinely collected electronic records to enable automatic monitoring and inform continuous optimization of HL-management at PHC settings.MethodsThe study used randomly selected electronic records of PHC (from the province-wide data center of Anhui-province, China) as the main data source and generated both procedure-based and encounter-based measures for assessing HL-management. The procedure-based measures were derived from specific quality-facts of 21 stages/procedures (e.g., lipid lowering medication prescription) using self-designed algorithms. While the encounter-based measures included number or rate of visits for HL, currently-noticed hyperlipidemia (CNHL, or HL noticed during the current consultation), and ever-diagnosed hyperlipidemia (EDHL). Analysis of these measures employed mainly simple descriptives and linear regression modeling.ResultsThe study revealed interesting findings including: low and varied rates of visits for HL(from 0.01 to 1.43%) and visits by patients with EDHL/CNHL(from 0.13 to 20.54% or from 0.02 to 2.99%) between regions; large differences (5.14 to 22.20 times) between the mean or cumulative proportions of visits by patients with EDHL versus CNHL among clinician groups; consistent increase in the ratio of visits for HL in all cause visits over the study period (from 0.087 to 1.000%) accompanied with relatively stable proportions of patients with CNHL/EDHL; Relatively low scores in the procedure-based measures (ranged from 0.00 to 36.08% for specific procedures by seasons).ConclusionsThe measures identified are not only feasible from real-world PHC records but also give some useful metrics about how well current HL-management is going and what future actions are needed.

Soil moisture partitioning strategies in blowouts in the Hulunbeier grassland and response to rainfall

IntroductionSoil moisture and soil water retention capacity are key influencing factors for the normal growth and development of vegetation. Understanding the dynamic change characteristics of soil moisture in blowouts and soil water retention capacity is of great significance for the management of blowouts.MethodsThis study employs drying and in situ monitoring methods to select typical blowouts in different regions (sand pits, fringe zones, sand accumulation zones, and sand-grass transition zones) on the Hulunbuir Grassland. A large area of natural grassland surrounding these regions was chosen as the control (CK). Soil moisture at depths of 20, 40, 60, 100 and 200 cm below the surface was measured along the soil profile using the ECH2O-10HS soil moisture automatic monitoring instrument. The HOBO-RG3-M self-recording rain gauge was used to monitor rainfall. Soil water storage, coefficient of variation, and Pearson’s correlation coefficient were calculated to study the differences in soil moisture and the dynamic change regularity in soil moisture under different rainfall conditions. This research provides important theoretical support for the soil moisture distribution and vegetation restoration in the blowouts of the Hulunbuir Grassland.Results and discussionThe volumetric water content of the soil in the blowouts was 15.95%, the volumetric soil water content in different parts of the soil varied from low to high as follows: sand pit-I &amp;lt; sand-grass transition zone-IV &amp;lt; fringe zone-II &amp;lt; CK &amp;lt; sand accumulation zone-III. The soil volumetric water content of the 0–40 cm soil layer of the blowout was higher than 17.47%, and the soil volumetric water content of the 40–200 cm soil layer ranged from 12.13% to 17.47%. The volumetric water content of soil in various parts of the blowouts under different rainfall amounts had significant differences, with rainstorms and heavy rainfall effectively recharging the blowouts to a depth of 200 cm, and the blowouts responded strongly to heavy rainfall (71.5 mm). A gradual recovery of the pre-rainfall volumetric soil moisture content was seen approximately a week after rainstorms. The water retention and storage capacity of blowout soils was significantly higher than that of CK, the soil water storage capacity of different zones ranked in descending order as the sand accumulation zone (1875.38 mm) &amp;gt; edge zone (1373.22 mm) &amp;gt; CK (1188.36 mm) &amp;gt; sand pit (1000.39 mm) &amp;gt; sand–grass transition zone (803.90 mm). The correlation coefficient of sand pits and sand cover was 0.5612, and that of sand accumulation zones and sand cover was 0.5845, which confirmed that sand cover enhanced the water retention capacity of the localized area of blowouts (sand accumulation zones).

Design and Fabrication of an Automatic Waste Segregation and Monitoring System

The quantity and composition of solid waste are increasing rapidly due to continuous economic development in many countries. Variation in quantity and composition of waste has become a serious problem for the waste management system in ensuring effectual and eco-friendly management of solid waste. The world generated approximately 2.01 billion tonnes (BT) of solid waste in 2016, which is anticipated to increase to 2.58 BT by 2030 and 3.40 BT by 2050. Currently, only one-fifth of the waste generated is being processed and the rest is dumped in landfills without any treatment. The optimal economic benefit derived from solid waste is attained when all its components are thoroughly segregated. Waste segregation at source helps significantly in reducing complexity in waste treatment plants. There is no system for segregating dry, wet, organic, plastic, and metallic waste at household, office, college, and industrial levels. So, the goal of this paper is to design, fabricate, and testing of an easy-to-use and cost-effective automatic waste separation system for homes and small local societies so that waste can be directly sent to the waste management plants. This system minimizes human interference and shortens the time and cost of segregation. Powered by Arduino UNO and various sensors, this system easily separates solid waste into three main categories namely metal, dry, and wet. During the performance testing of the system, the accuracy of the system was noted at 98% for 100 samples and also the time taken by the system to separate metal waste, wet waste, and dry waste respectively was 2.5, 3.5, and 5.6 seconds.

Air quality in the Metropolitan Zone of the Valley of Puebla: Comparative evaluation of CAMS and persistence forecasts

Background on air quality in the Metropolitan Zone of the Valley of Puebla shows that suspended particles smaller than 10 micrometers (PM10) and smaller than 2.5 micrometers (PM2.5) represent a health risk. Puebla’s automatic air quality monitoring system measures PM10 and PM2.5 at five stations in the municipalities of Puebla and Coronango. These measurements allow for determining the Air and Health Index according to the NOM-172-SEMARNAT-2019 standard for these pollutants. The advancement of global pollutant modeling techniques represents an opportunity for air quality management in areas with scarce terrestrial measurements. However, it is necessary to validate global forecasts with ground measurements from georeferenced monitoring stations to reduce uncertainties and determine reliability. The Copernicus Atmospheric Monitoring Service (CAMS) forecast allows atmospheric pollution exploration processes in the study region. This study presents an analysis of the CAMS forecast against the Persistence forecast. The results show that the persistence forecast performs better than the CAMS forecast in general, both for PM10 and for PM2.5. However, using the CAMS forecast for a preliminary evaluation of the prediction of PM2.5 is feasible due to its acceptable values in the comparison criteria of the dichotomous statistics ACCURACY, probability of detection (POD), false alarm rate (FAR), probability of false detection (POFD), success ratio (SR), threat score (TS), equitable threat score (ETS), Heidke skill score (HSS), and odds ratio skill score (ORSS). This work provides valuable insights to both the population and decision-makers, aiding in the enhancement of air quality management and public health strategies.

Polynomial Regression and Faster R-CNN Models for University Library Decision Implementation Discovery based on Deep Learning

We deal with the development of a seat occupancy detection algorithm for the University’s library utilizing the Faster R-CNN algorithm. The university’s library is widely used by students, particularly during exam season when it can become difficult to find a seat. Going from one building to another is often time-consuming and useless when there are no available seats. This system uses the features of Faster R-CNN in a way to facilitate an automatic seat occupancy monitoring system. Unlike conventional methods of using manual monitoring or weights and occupancy switches as inanimate indicators, it provides real-time seat availability data which allows no human intervention to be a part of the process. Data collection and model training evaluation are considered using annotated datasets with images of library seating layouts. The Faster R-CNN model is trained such that it can accurately detect vacancy or occupancy at library seats. This work takes a futuristic approach towards smart library management systems, in which user needs are changing, and considers the use of high-end computer vision technologies to be integrated into all such libraries. The proposed system aims to leverage the effectiveness of Faster R-CNN and go a long way in redefining seat occupancy management for university libraries by enabling better efficiency, resource utilization, as well as user satisfaction in prospect.

Automatic monitoring and on-line chiral separation of chiral drug synthesis.

Automatic monitoring and on-line chiral separation of chiral drug synthesis.

Adaptive Transmit Sequencing for Robust Flow Monitoring in Cross-sectional Doppler.

Doppler ultrasound is a non-invasive imaging technique that measures blood flow velocity and is commonly used in cardiac evaluation and vascular assessment. Compared to the conventional longitudinal view, cross-sectional Doppler is more robust to motion, making it more suitable for monitoring applications. In this paper, an adaptive framework is presented to automatically monitor flow in the common carotid artery using cross-sectional Doppler. Based on a vessel segmentation and geometry estimation, transmit parameters such as the focal point, steering angle and aperture width are adaptively adjusted to optimize the Doppler angle and to maximize SNR. The velocity profile is estimated using multiple gates along a single line, resulting in velocity estimates with high temporal resolution. The effect and optimal settings of relevant non-adaptive ultrasound parameters is explored through a design of experiments, making use of simulated and phantom data. These optimal parameters result in accurate estimates of average velocity with a mean error of 0.8% in silico and 1.6% in vitro. In addition, velocity estimates show a reduced variance and improved temporal resolution compared to conventional line-by-line scanning. Feasibility of the method is also demonstrated in vivo, where a diverse range of velocity profiles was observed. These findings suggest that this method could be feasible for automatic flow monitoring or cardiac output estimation through hemodynamic modeling.