[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
You seem to have javascript disabled. Please note that many of the page functionalities won't work as expected without javascript enabled.
 
 
Sign in to use this feature.

Years

Between: -

Subjects

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Journals

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Article Types

Countries / Regions

remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline
remove_circle_outline

Search Results (3,217)

Search Parameters:
Keywords = robust statistics

Order results
Result details
Results per page
Select all
Export citation of selected articles as:
15 pages, 3817 KiB  
Article
Flow Cytometric Detection of Waterborne Bacteria Metabolic Response to Anthropogenic Chemical Inputs to Aquatic Ecosystems
by Jill A. Jenkins, Scott V. Mize, Darren Johnson and Bonnie L. Brown
Cells 2025, 14(5), 352; https://doi.org/10.3390/cells14050352 - 28 Feb 2025
Abstract
Typical investigations into the biological consequences of suspected xenobiotics or nutrients introduced in watersheds include analytical chemistry screens of environmental samples—such as periphyton responses or studies of fish condition—which are all costly in terms of equipment, reagents, time, and human resources. An alternative [...] Read more.
Typical investigations into the biological consequences of suspected xenobiotics or nutrients introduced in watersheds include analytical chemistry screens of environmental samples—such as periphyton responses or studies of fish condition—which are all costly in terms of equipment, reagents, time, and human resources. An alternative is to assess pollutant effects on waterborne bacteria. A flow cytometric method was developed to yield rapid, same-day results that could be used to proactively screen for suspected chemical inputs into watersheds using water sampling methods that are identical to those in standard use. The analytical methods are microbe cultivation-independent, for use with waterborne bacteria that are typically viable but not culturable. The procedure is quick and inexpensive, generating measures of bacterial esterase that reflect metabolic activity and are sensitive and statistically robust. After phosphate-EDTA incubation to increase cell wall permeability, staining was performed with 5(6) carboxyfluorescein diacetate (enzyme activity) and propidium iodide (cell viability) with three bacterial species in exponential phase growth having been incubated with organic wastewater compounds (atrazine, pharmaceuticals [17α-ethynylestradiol and trenbolone], and antimicrobials [tylosin and butylparaben]). This method successfully detected metabolic changes in all bacterial species, with atrazine inducing the greatest change. Additional fluorescent stains can target specific microbial structures or functions of interest in a particular watershed. This biotechnology can inform analytical chemistry and study of biota at sites of interest and has the potential to be automated. Full article
(This article belongs to the Special Issue The Applications of Flow Cytometry: Advances, Challenges, and Trends)
Show Figures

Figure 1

Figure 1
<p>Cytograms from <span class="html-italic">Escherichia coli</span> (<span class="html-italic">E. coli</span>) staining controls for metabolic activity of live (left column) or heat-killed (right column) cells. Unstained bacteria and auto-fluorescent bacteria, concentrated near the origin, were excluded (out-gated; round gates above) for statistical computations. Bacteria stained with CDFA (5(6)-carboxyfluorescein diacetate) yield a fluorescent product because of hydrolysis by esterases (FL1-H axis; green fluorescence) shown here as populations focused at the bottom portion of the cytograms. Bacteria stained with propidium iodide (that fluoresces when intercalated into double-stranded nucleic acids of membrane-damaged cells) (FL3-H axis; red fluorescence) shown here as populations focused on the left portions of the cytograms.</p>
Full article ">Figure 2
<p>Representative flow cytometric zebra plots display representative analyses of metabolic activity in strains of bacteria that were laboratory-exposed to organic wastewater compounds. Percentages of stained cells are presented, whereby the upper gate is that of the metabolically inactive cells and the right-hand gate shows the metabolically active percentage. In panels (<b>A</b>,<b>B</b>): <span class="html-italic">Escherichia coli</span> was incubated with atrazine and tylosin, respectively. (<b>C</b>,<b>D</b>): <span class="html-italic">Streptococcus suis</span> was incubated with trenbolone and butylparaben, respectively. (<b>E</b>,<b>F</b>): <span class="html-italic">Streptococcus</span> dysgalactiae was incubated with atrazine and tylosin, respectively. Staining with 5(6)-carboxyfluorescein diacetate yields a fluorescent product upon hydrolysis by esterases (horizontal axis; FL1-H; green fluorescence), propidium iodide counter-staining, whereby nucleic acids are stained in membrane-damaged cells (vertical axis; FL3-H; red fluorescence). Unstained or auto-fluorescent particles were out-gated for the analyses (e.g., ungated population near the origin) (<a href="#cells-14-00352-f001" class="html-fig">Figure 1</a>).</p>
Full article ">
15 pages, 2723 KiB  
Article
Identification of Genetic Relationships and Group Structure Analysis of Yanqi Horses
by Yaru Wang, Chi Tang, Pengfei Xue, Na Yang, Xiaoyuan Sun, Khizat Serik, Tolegen Assanbayer, Malika Shamekova, Zhassulan Kozhanov, Zagipa Sapakhova, Jurakulov Kobil Khurramovich, Xiaoling Zhou, Iskhan Kairat and Gemingguli Muhatai
Genes 2025, 16(3), 294; https://doi.org/10.3390/genes16030294 - 27 Feb 2025
Abstract
Background/Objectives: The Yanqi horse is a distinguished local breed in China, known for its robust physique and strong adaptability. However, due to insufficient breeding populations and a loosely structured breeding system, the number of Yanqi horses has been declining annually. To protect its [...] Read more.
Background/Objectives: The Yanqi horse is a distinguished local breed in China, known for its robust physique and strong adaptability. However, due to insufficient breeding populations and a loosely structured breeding system, the number of Yanqi horses has been declining annually. To protect its genetic resources and develop scientific breeding strategies, this study aimed to analyze the genetic diversity, parentage relationships, and genetic structure of the Yanqi horse conservation population using microsatellite markers. Materials and Methods: A total of 117 Yanqi horses were selected for genotyping analysis using 16 microsatellite markers. Genetic diversity parameters (e.g., allele number, heterozygosity, F-statistics) were calculated using GeneAIEX (v.6.503) and Fstat software (v.2.9.4). Parentage analysis was conducted using Cervus software. Bayesian clustering analysis was performed using STRUCTURE software (v.2.3.4), and a phylogenetic tree was constructed based on Nei’s genetic distance to reveal the population genetic structure. Results: A total of 191 alleles were detected, with an average allele number of 11.969, observed heterozygosity of 0.481, and expected heterozygosity of 0.787. Parentage testing showed a cumulative exclusion probability (CEP) of 0.9652999 when one parent’s genotype was known and 0.9996999 when both parents’ genotypes were known, achieving an accuracy of 99%. Genetic differentiation analysis revealed moderate genetic divergence among populations (FST = 0.128) and moderate inbreeding levels (FIS = 0.396). Bayesian clustering analysis (K = 4) indicated that the Yanqi horse population could be divided into four genetic clusters, reflecting the impact of geographical isolation on genetic structure. Conclusions: The Yanqi horse conservation population exhibits moderate genetic diversity, high accuracy in parentage identification, and moderate genetic differentiation and inbreeding. The findings provide a scientific basis for the conservation and sustainable utilization of Yanqi horse genetic resources. Future efforts should focus on strengthening conservation measures, optimizing breeding strategies, and further investigating the genetic background using genomic technologies to ensure the sustainable development of the Yanqi horse population. Full article
(This article belongs to the Section Animal Genetics and Genomics)
Show Figures

Figure 1

Figure 1
<p>(<b>a</b>) Male Yanqi horse; (<b>b</b>) mare Yanqi; and (<b>c</b>) a group photo of Yanqi horses.</p>
Full article ">Figure 2
<p>Population structure of the Yanqi horse. The clustering results illustrate the structural analysis of 117 Yanqi horses based on 16 microsatellite markers. Each horse’s genotype is represented by a vertical line, which is divided into K colors, where K denotes the number of clusters hypothesized in each structural analysis. Each bar corresponds to an individual horse, and the color on each vertical bar indicates the probability of that individual belonging to each cluster.</p>
Full article ">Figure 3
<p>Phylogenetic tree based on 117 Yanqi horses in NJ. The number on each branch corresponds to each individual Yanqi horse.</p>
Full article ">
19 pages, 11167 KiB  
Article
Robust Sandstorm Image Restoration via Adaptive Color Correction and Saturation Line Prior-Based Dust Removal
by Shan Zhou, Fei Shi, Zhenhong Jia, Guoqiang Wang and Jian Huang
Appl. Sci. 2025, 15(5), 2594; https://doi.org/10.3390/app15052594 - 27 Feb 2025
Abstract
Enhancing the visibility of outdoor images under sandstorm conditions remains a significant challenge in computer vision due to the complex atmospheric interference caused by dust particles. While existing image enhancement algorithms perform well in mild sandstorm scenarios, they often struggle to produce satisfactory [...] Read more.
Enhancing the visibility of outdoor images under sandstorm conditions remains a significant challenge in computer vision due to the complex atmospheric interference caused by dust particles. While existing image enhancement algorithms perform well in mild sandstorm scenarios, they often struggle to produce satisfactory results in more severe conditions, where residual color casts and chromatic artifacts become pronounced. These limitations highlight the need for a more robust and adaptable restoration method. In this study, we propose an advanced algorithm designed to restore sand-dust images under varying sandstorm intensities, effectively addressing the aforementioned challenges. The approach begins with a color correction step, achieved through channel compensation and color transfer techniques, which leverage the unique statistical properties of sand-dust images. To further refine the restoration, we improve the boundary constraints of the saturation line prior (SLP) by adjusting the local illumination in the atmospheric light map, enhancing the model’s robustness to environmental variations. Finally, the atmospheric scattering model is employed for comprehensive image restoration, ensuring that color correction and dust removal are optimized. Extensive experiments on real-world sandstorm images demonstrate that the proposed method performs on par with state-of-the-art (SOTA) techniques in weaker sandstorm scenarios, showing marked improvements in more severe conditions. These results highlight the potential of our approach for practical applications in outdoor image enhancement under challenging environmental conditions. Full article
Show Figures

Figure 1

Figure 1
<p>Sandstorm images from different scenarios with corresponding color histograms: (<b>a</b>–<b>c</b>) are sandstorm images. (<b>d</b>–<b>f</b>) are their corresponding color histograms.</p>
Full article ">Figure 2
<p>Overview of the proposed sand dust image restoration algorithm.</p>
Full article ">Figure 3
<p>Comparison of recovery results from different atmospheric light calculation methods.</p>
Full article ">Figure 4
<p>Comparison of color correction results.</p>
Full article ">Figure 5
<p>Comparison of recovery results of weak sandstorm images.</p>
Full article ">Figure 6
<p>Comparison of recovery results of yellow sandstorm images.</p>
Full article ">Figure 7
<p>Comparison of recovery results of severe sandstorm images.</p>
Full article ">Figure 8
<p>More experimental results for the proposed method: (<b>a</b>) Restoration results of yellow sandstorm images; (<b>b</b>) Restoration results of red sandstorm images.</p>
Full article ">Figure 9
<p>Visual object detection results for YOLOv5: (<b>a</b>) The detection results of sandstorm images; (<b>b</b>) The detection results of de-dusted images obtained by the proposed method.</p>
Full article ">Figure 10
<p>Failure cases of dust removal.</p>
Full article ">Figure 11
<p>Failure cases of color correction.</p>
Full article ">
25 pages, 850 KiB  
Article
Similarities: The Key Factors Influencing Cross-Site Password Guessing Performance
by Zhijie Xie, Fan Shi, Min Zhang, Zhihong Rao, Yuxuan Zhou and Xiaoyu Ji
Electronics 2025, 14(5), 945; https://doi.org/10.3390/electronics14050945 (registering DOI) - 27 Feb 2025
Viewed by 8
Abstract
Password guessing is a crucial research direction in password security, considering vulnerabilities like password reuse and data breaches. While research has extensively explored intra-site password guessing, the complexities of cross-site attacks, where attackers use leaked data from one site to target another, remain [...] Read more.
Password guessing is a crucial research direction in password security, considering vulnerabilities like password reuse and data breaches. While research has extensively explored intra-site password guessing, the complexities of cross-site attacks, where attackers use leaked data from one site to target another, remain less understood. This study investigates the impact of dataset feature similarity on cross-site password guessing performance, revealing that dataset differences significantly influence guessing success more than model variations. By analyzing eight password datasets and four guessing methods, we identified eight key features affecting guessing success, including general data features like length distribution and specific semantic features like PCFG grammar. Our research reveals that syntactic and statistical patterns in passwords, particularly PCFG features, are most effective for cross-site password guessing due to their strong generalization across datasets. The Spearman correlation coefficient of 0.754 between PCFG feature similarity and guessing success rate indicates a significant positive correlation, unlike the minimal impact of length distribution features (0.284). These findings highlight the importance of focusing on robust semantic features like PCFG for improving password guessing techniques and security strategies. Additionally, the study underscores the importance of dataset selection for attackers and suggests that defenders can enhance security by mitigating feature similarity with commonly leaked data. Full article
(This article belongs to the Section Computer Science & Engineering)
Show Figures

Figure 1

Figure 1
<p>A systematic analysis framework to study the impact of dataset similarity on the performance of cross-site password guessing.</p>
Full article ">Figure 2
<p>Comparison of the performance of four password guessing methods under different training sample sizes in the cross-site password guessing scenario with <tt>Dodonew</tt> as the training set and <tt>Tianya</tt> as the testing set.</p>
Full article ">Figure 3
<p>Comparison of guessing performance across different training and testing data combination scenarios. The <math display="inline"><semantics> <msub> <mi>Min</mi> <mi>auto</mi> </msub> </semantics></math> metric for four password guessing methods at <math display="inline"><semantics> <msup> <mn>10</mn> <mn>5</mn> </msup> </semantics></math> guess number is used as the guessing performance evaluation metric, with a training set size of <math display="inline"><semantics> <msup> <mn>10</mn> <mn>6</mn> </msup> </semantics></math>.</p>
Full article ">Figure 4
<p>Length distribution of datasets.</p>
Full article ">Figure 5
<p>Character composition type distribution of datasets.</p>
Full article ">Figure 6
<p>Feature similarity distribution across datasets.</p>
Full article ">Figure 7
<p>Analysis of the relationship between feature similarity and guessing performance in cross-site password guessing scenarios.</p>
Full article ">Figure 8
<p>Training data evaluation values for cross-site password guessing. Each horizontal column representing the evaluation values of training data for the same target data.</p>
Full article ">
20 pages, 2789 KiB  
Article
Evaluating Interlaboratory Variability in Wastewater-Based COVID-19 Surveillance
by Arianna Azzellino, Laura Pellegrinelli, Ramon Pedrini, Andrea Turolla, Barbara Bertasi, Sandro Binda, Sara Castiglioni, Clementina E. Cocuzza, Fabio Ferrari, Andrea Franzetti, Maria Giovanna Guiso, Marina Nadia Losio, Marianna Martinelli, Antonino Martines, Rosario Musumeci, Desdemona Oliva, Laura Sandri, Valeria Primache, Francesco Righi, Annalisa Scarazzato, Silvia Schiarea, Elena Pariani, Emanuela Ammoni, Danilo Cereda and Francesca Malpeiadd Show full author list remove Hide full author list
Microorganisms 2025, 13(3), 526; https://doi.org/10.3390/microorganisms13030526 - 27 Feb 2025
Viewed by 125
Abstract
Wastewater-based environmental surveillance enables the monitoring of SARS-CoV-2 dynamics within populations, offering critical epidemiological insights. Numerous workflows for tracking SARS-CoV-2 have been developed globally, underscoring the need for interlaboratory comparisons to ensure data consistency and comparability. An inter-calibration test was conducted among laboratories [...] Read more.
Wastewater-based environmental surveillance enables the monitoring of SARS-CoV-2 dynamics within populations, offering critical epidemiological insights. Numerous workflows for tracking SARS-CoV-2 have been developed globally, underscoring the need for interlaboratory comparisons to ensure data consistency and comparability. An inter-calibration test was conducted among laboratories within the network monitoring SARS-CoV-2 in wastewater samples across the Lombardy region (Italy). The test aimed to evaluate data reliability and identify potential sources of variability using robust statistical approaches. Three wastewater samples were analyzed in parallel by four laboratories using identical pre-analytical (PEG-8000-based centrifugation) and analytical processes (qPCR targeting N1/N3 and Orf-1ab). A two-way ANOVA framework within Generalized Linear Models was applied, and multiple pairwise comparisons among laboratories were performed using the Bonferroni post hoc test. The statistical analysis revealed that the primary source of variability in the results was associated with the analytical phase. This variability was likely influenced by differences in the standard curves used by the laboratories to quantify SARS-CoV-2 concentrations, as well as the size of the wastewater treatment plants. The findings of this study highlight the importance of interlaboratory testing in verifying the consistency of analytical determinations and in identifying the key sources of variation. Full article
(This article belongs to the Special Issue Surveillance of SARS-CoV-2 Employing Wastewater)
Show Figures

Figure 1

Figure 1
<p>Workflow followed by the laboratories during the interlaboratory ring test (created in BioRender. <a href="https://BioRender.com/b09r836" target="_blank">https://BioRender.com/b09r836</a>, accessed 14 February 2025).</p>
Full article ">Figure 2
<p>Linearized relationships of the log-transformed concentrations (g.c./μL) across the laboratories responsible for the analytical phase. Full regression statistics for the linear relationship depicted in the chart are provided in <a href="#app1-microorganisms-13-00526" class="html-app">Table S1</a>.</p>
Full article ">Figure 3
<p>Log-transformed detections of N1 gene fragment copies/µL concentration: differences among laboratories concerning the analytical and pre-analytical phases. Full regression statistics for the linear relationship depicted in the chart are provided in <a href="#app1-microorganisms-13-00526" class="html-app">Table S2</a>.</p>
Full article ">Figure 4
<p>Log-transformed detections of N3 gene fragment copies/µL concentration: differences among laboratories concerning the analytical and pre-analytical phases. Full regression statistics for the linear relationship depicted in the chart are provided in <a href="#app1-microorganisms-13-00526" class="html-app">Table S3</a>.</p>
Full article ">Figure 5
<p>Log-transformed detections of ORF1ab gene fragment copies/µL concentration: differences among laboratories concerning the analytical and pre-analytical phases. Full regression statistics for the linear relationship depicted in the chart are provided in <a href="#app1-microorganisms-13-00526" class="html-app">Table S4</a>.</p>
Full article ">Figure 6
<p>Model estimated marginal means of the log-transformed detections of ORF gene fragments: it can be observed that Lab2 shows the most significant variability with respect to the other laboratories and with respect to the WWTP.</p>
Full article ">Figure 7
<p>Linear relationships of the log-transformed gene copy values and Cq for the three gene fragments (N1, N3, and ORFab) of the two different RT-PCR systems (e.g., AgPath and QuantaBio).</p>
Full article ">Figure 8
<p>Interlaboratory comparison of standard curves before (<b>upper chart</b>) and after (<b>lower chart</b>) the harmonization process.</p>
Full article ">
20 pages, 3908 KiB  
Article
O/C Isotopic and EPR Signature of Marble from the Apuan Alps (Italy): A Critical Review
by Massimo Coli, Francesco Di Benedetto and Antonella Buccianti
Appl. Sci. 2025, 15(5), 2533; https://doi.org/10.3390/app15052533 - 26 Feb 2025
Viewed by 200
Abstract
A critical review of studies concerning the attribution of the provenance of marble from the Apuan Alps (Italy) (AAM) used for historical–monumental buildings and artefacts is proposed based on its O/C isotopic and EPR signature. First, a summary of the geological origin of [...] Read more.
A critical review of studies concerning the attribution of the provenance of marble from the Apuan Alps (Italy) (AAM) used for historical–monumental buildings and artefacts is proposed based on its O/C isotopic and EPR signature. First, a summary of the geological origin of AAM and its geo-structural evolution and setting is presented. A review of the exploitation history of AAM is then discussed. This geological and historical information is used as categorical information to better constrain the literature multimethodic database, containing numerous data, including O/C isotopic and EPR spectroscopic parameters. A robust multivariate statistical analysis of the combination of all these data is performed. The results point to the fact that the O/C isotopic and EPR signature can help in attributing an analysed AAM sample to a marble extraction district, and to a certain extent also to a site, whereas the discrimination of the individual quarry appears to not yet be achievable. Full article
(This article belongs to the Section Earth Sciences)
Show Figures

Figure 1

Figure 1
<p>Tectonic sketch map of the Apuan Alps metamorphic core complex. ND = Neogene deposits; TN = Tuscan Nappe; MU = Massa unit; PB = Palaeozoic basement; CP = Carbonate Platform: Late Triassic dolomite (Grezzoni) and Early Jurassic marble; PU = Middle Jurassic to Oligocene pelagic units. CS = Carrara Syncline; VS = Vallini Syncline; VA = Vinca Anticline; OS = Orto di Donna Syncline; TA = Tambura Anticline; AS = Arnetola Syncline; BD = Boana dome; blue line = top of the uplifting dome (elaboration, M. Coli).</p>
Full article ">Figure 2
<p>Cumulative structural setting of the Apuan Alps metamorphic complex through serial cross-sections from south to north obtained by projecting the main structures that gently dip towards the NW. PB = Palaeozoic basement; CP = carbonate platform: Late Triassic dolomite (Grezzoni) and Early Jurassic marble; PU = Middle Jurassic to Oligocene pelagic units. In red quarry sites, B = Boccanaglia; T = Torano; M = Miseglia; C = Colonnata and Caglieglie; S = Seravezza; F = Frigido; A-C = Altissimo Cervaiole. In black main folds, CS = Carrara Syncline; VS = Vallini Syncline; VA = Vinca Anticline; OS = Orto di Donna Syncline; TA = Tambura Anticline; AS = Arnetola Syncline; BD = Boana dome. Numbers refer to the structural level from the lowermost (#1) to the topmost (#9), with a total thickness of the folds stack-pile valuables at about 10 km; in black, not sampled levels; in blue, sampled levels (elaboration, M. Coli).</p>
Full article ">Figure 3
<p>O/C isotope data for the AAMs, categorised by sampled quarries, and the reference trends for the Hettangian and the Sinemurian in the Global Time Scale (elaboration, M. Coli).</p>
Full article ">Figure 4
<p>O/C isotope data for the AAMs, categorised by marble type (i.e., carbonate-platform environments), and the reference trends for the Hettangian and the Sinemurian in the Global Time Scale (elaboration, M. Coli).</p>
Full article ">Figure 5
<p>O/C isotope data for the AAMs, categorised by structural level, and the reference trends for the Hettangian and the Sinemurian in the Global Time Scale (elaboration, M. Coli).</p>
Full article ">Figure 6
<p>Ternary plot of the SPREAD, SPLI and W parameters. The sum of the three parameters has been normalised. A rough linear correlation involving the three parameters is apparent (elaboration, F. Di Benedetto).</p>
Full article ">Figure 7
<p>Correlation between the irl1 and irl2 parameters considered. Cases are discriminated by stratigraphic level (LS) (elaboration, F. Di Benedetto).</p>
Full article ">Figure 8
<p>Dendrogram of the 169 cases evaluated using variable standardisation and Euclidean distance evaluation. The branches highlighted in colour are discussed in the text. Labels refer to the district of the sample: red stars indicate Seravezza samples, blue ellipses indicate Massa samples and no labelling indicates Carrara samples (elaboration, F. Di Benedetto).</p>
Full article ">Figure 9
<p>PC2 versus PC1 diagrams. Cases are discriminated by colour as discussed in the text: (<b>a</b>) district, (<b>b</b>) site, (<b>c</b>) TM, (<b>d</b>) LS and (<b>e</b>) HISTORY (elaboration, F. Di Benedetto).</p>
Full article ">
32 pages, 14603 KiB  
Review
A Review of Machine Learning and Deep Learning Methods for Person Detection, Tracking and Identification, and Face Recognition with Applications
by Beibut Amirgaliyev, Miras Mussabek, Tomiris Rakhimzhanova and Ainur Zhumadillayeva
Sensors 2025, 25(5), 1410; https://doi.org/10.3390/s25051410 - 26 Feb 2025
Viewed by 171
Abstract
This paper provides a comprehensive analysis of recent developments in face recognition, tracking, identification, and person detection technologies, highlighting the benefits and drawbacks of the available techniques. To assess the state-of-art in these domains, we reviewed more than one hundred eminent journal articles [...] Read more.
This paper provides a comprehensive analysis of recent developments in face recognition, tracking, identification, and person detection technologies, highlighting the benefits and drawbacks of the available techniques. To assess the state-of-art in these domains, we reviewed more than one hundred eminent journal articles focusing on current trends and research gaps in machine learning and deep learning methods. A systematic review using the PRISMA method helped us to generalize the search for the most relevant articles in this area. Based on our screening and evaluation procedures, we found and examined 142 relevant papers, evaluating their reporting compliance, sufficiency, and methodological quality. Our findings highlight essential methods of person detection, tracking and identification, and face recognition tasks, emphasizing current trends and illustrating a clear transition from classical to deep learning methods with existing datasets, divided by task and including statistics for each of them. As a result of this comprehensive review, we agree that the results demonstrate notable improvements. Still, there remain several key challenges like refining model robustness under varying environmental conditions, including diverse lighting and occlusion; adaptation to different camera angles; and ethical and legal issues related to privacy rights. Full article
(This article belongs to the Section Sensing and Imaging)
Show Figures

Figure 1

Figure 1
<p>Table of used PRISMA model in this review.</p>
Full article ">Figure 2
<p>(<b>a</b>) Example of object detection, where the model identifies and locates objects in an image using bounding boxes. (<b>b</b>) Example of segmentation, where the model assigns pixel-level labels to different regions of the image. (<b>c</b>) Example of pose estimation, where the model forecasts the locations and orientations of a person’s major body joints [<a href="#B22-sensors-25-01410" class="html-bibr">22</a>].</p>
Full article ">Figure 3
<p>Face recognition challenges due to variations in pose, lighting, and facial expression (image of one of our team members).</p>
Full article ">Figure 4
<p>The architecture of the FSOD-KT network [<a href="#B117-sensors-25-01410" class="html-bibr">117</a>].</p>
Full article ">Figure 5
<p>The architecture of the proposed people counting and tracking system [<a href="#B79-sensors-25-01410" class="html-bibr">79</a>].</p>
Full article ">Figure 6
<p>Flow chart of the intruder detection system [<a href="#B134-sensors-25-01410" class="html-bibr">134</a>].</p>
Full article ">Figure 7
<p>The proposed model for the loitering detection system [<a href="#B135-sensors-25-01410" class="html-bibr">135</a>].</p>
Full article ">
14 pages, 5120 KiB  
Article
An Enhanced Neural Network Forecasting System for July Precipitation over the Middle-Lower Reaches of the Yangtze River
by Wenyan Liu and Xiangjun Shi
Atmosphere 2025, 16(3), 272; https://doi.org/10.3390/atmos16030272 - 26 Feb 2025
Viewed by 81
Abstract
Forecasting July precipitation using prophase winter sea surface temperature through a nonlinear machine learning model remains challenging. Given the scarcity of observed samples and more attention should be paid to anomalous precipitation events, the shallow neural network (NN) and several improving techniques are [...] Read more.
Forecasting July precipitation using prophase winter sea surface temperature through a nonlinear machine learning model remains challenging. Given the scarcity of observed samples and more attention should be paid to anomalous precipitation events, the shallow neural network (NN) and several improving techniques are employed to establish the statistical forecasting system. To enhance the stability of predicted precipitation, the final output precipitation is an ensemble of multiple NN models with optimal initial seeds. The precipitation data from anomalous years are amplified to focus on anomalous events rather than normal events. Some artificial samples are created based on the relevant background theory to mitigate the problem of insufficient sample size for model training. Sensitivity experiments indicate that the above techniques could improve the stability and interpretability of the forecasting system. Rolling forecasts further indicate that the forecasting system is robust and half of the anomalous events can be successfully predicted. These improving techniques used in this study can be applied not only to the precipitation over the middle-lower reaches of the Yangtze River but also to other climate events. Full article
Show Figures

Figure 1

Figure 1
<p>The original precipitation anomaly percentage (black line) alongside the scaled precipitation anomaly percentage (red line). Dashed lines indicate the threshold values for anomalous events.</p>
Full article ">Figure 2
<p>The linear correlation coefficients between SST and Rain (<b>upper panel</b>, stippling indicates the <span class="html-italic">t</span>-test confidence level greater than 95%) and the composite analyses (<b>lower panel</b>) in high (<b>left</b>) and low (<b>right</b>) precipitation anomalous years. The values in the top right corner correspond to Rain. The red box indicates the SST area used for the forecasting system.</p>
Full article ">Figure 3
<p>The SST (colored, in units of °C) and corresponding Rain (top right corner) from the six synthetic samples created by the high (upper panel) or low (lower panel) composite analysis data. These synthetic samples have been sorted by strength (order number in the top left corner).</p>
Full article ">Figure 4
<p>Flow–process diagram for calculating the score (a correlation value) of a given seed.</p>
Full article ">Figure 5
<p>Rolling forecasts from the OPT (red line) and FMK (blue line) experiments. The observed Precipitation is denoted by the black line. Gray dashed lines indicate the threshold values for anomalous events. Forecast skills (Corr, Succ, and Wrong) are presented adjacent to the experiment names.</p>
Full article ">Figure 6
<p>Forecasting stability from the OPT, NoSD, and NoEM experiments. The observed and predicted Precipitation are denoted by black and blue solid lines, respectively. Gray dashed lines indicate the threshold values for anomalous events. Forecast skills (Corr, Succ, and Wrong) are presented adjacent to the experiment names. Note that, the Round1 of OPT experiment is the OPT experiment shown in <a href="#atmosphere-16-00272-f005" class="html-fig">Figure 5</a>.</p>
Full article ">Figure 7
<p>Rolling forecasts from the OPT, NoSS, and NoAF experiments (upper panel) and interpretability analyses in 1956, 1998 and 2001 (lower panel). The SST fields and the corresponding sensitivity maps are presented in the 1st column and the 2nd–4th columns, respectively.</p>
Full article ">
41 pages, 3067 KiB  
Article
Agricultural Land, Sustainable Food and Crop Productivity: An Empirical Analysis on Environmental Sustainability as a Moderator from the Economy of China
by Fahmida Laghari, Farhan Ahmed, Babar Ansari and Paulo Jorge Silveira Ferreira
Sustainability 2025, 17(5), 1980; https://doi.org/10.3390/su17051980 - 25 Feb 2025
Viewed by 230
Abstract
The availability of agricultural land is central to stimulating reserves in sustainable food and crop production amidst accelerating economic sustainability and growth. Therefore, this article aims to investigate the influence of agricultural land (AGL) on food production (FP) and crop production (CP) with [...] Read more.
The availability of agricultural land is central to stimulating reserves in sustainable food and crop production amidst accelerating economic sustainability and growth. Therefore, this article aims to investigate the influence of agricultural land (AGL) on food production (FP) and crop production (CP) with the linkage of environmental sustainability (ES) as a moderator from 1990 to 2021 for the economy of China with the autoregressive distributed lag (ARDL) bounds testing estimation model. Our findings showed that the ARDL model estimates the long-term and short-term joint matching relationships between agricultural land and the independent variables in the model, which is a statistically significant outcome. Therefore, in the long term, the food and crop production adjustment for speed to steadiness was huge as it was projected at 1.337%, 53.6%, 133.5%, and 37.4%, respectively, in all the models, which shows that the adjustment for speed of models is a good post-shock association process. We found evidence for a significant and positive relationship between agricultural land and food and crop production in ordinary least square (OLS) estimation, which also ensured the outcomes of the primary model. Furthermore, Toda–Yamamoto Granger causality test estimation found reverse causality between food production (FP) and crop production (CP) and showed evidence of the conservation hypothesis. We found bidirectional causality between food production and agricultural land and between crop production and agricultural land, which shows evidence of the feedback hypothesis. Additionally, the empirical findings of a robustness check with fully modified ordinary least square (FMOLS) and dynamic ordinary least square (DOLS) techniques showed consistency with the investigations of ARDL estimation in the long run, ensuring the validity and strength of the primary outcomes. Overall, the present paper brings fresh knowledge about agricultural land use, and food and crop production to promote environmental sustainability. Full article
(This article belongs to the Special Issue Sustainable Development of Agricultural Systems)
Show Figures

Figure 1

Figure 1
<p>Conceptual framework of the study.</p>
Full article ">Figure 2
<p>Food production and agricultural land.</p>
Full article ">Figure 3
<p>Crop production and agricultural land.</p>
Full article ">Figure 4
<p>Carbon dioxide emission (CO<sub>2</sub>) and agricultural land.</p>
Full article ">Figure 5
<p>GDP growth and agricultural land.</p>
Full article ">Figure 6
<p>Urban population and agricultural land.</p>
Full article ">Figure 7
<p>Inflation and agricultural land.</p>
Full article ">Figure 8
<p>Plots of cumulative sum (CUSUM) and cumulative sum of square (CUSUMQ) for the model FP = f (AGL, CO<sub>2</sub>, GDPG, UP, INF).</p>
Full article ">Figure 9
<p>Plots of cumulative sum (CUSUM) and cumulative sum of square (CUSUMQ) for the model (CP = f (AGL, CO<sub>2</sub>, GDPG, UP, INF).</p>
Full article ">Figure 10
<p>Plots of cumulative sum (CUSUM) and cumulative sum of square (CUSUMQ) for the model (FP= f (AGL, AGL*ES, CO<sub>2</sub>, GDPG, UP, INF).</p>
Full article ">Figure 11
<p>Plots of cumulative sum (CUSUM) and cumulative sum of square (CUSUMQ) for the model (CP= f (AGL, AGL*ES, CO<sub>2</sub>, GDPG, UP, INF).</p>
Full article ">Figure 12
<p>Chart of actual and estimated value for the model FP = f (AGL, CO<sub>2</sub>, GDPG, UP, INF).</p>
Full article ">Figure 13
<p>Chart of actual and estimated value for the model CP = f (AGL, CO<sub>2</sub>, GDPG, UP, INF).</p>
Full article ">Figure 14
<p>Chart of actual and estimated value for the model FP= f (AGL, AGL*ES, CO<sub>2</sub>, GDPG, UP, INF).</p>
Full article ">Figure 15
<p>Chart of the actual and estimated value for the model (CP= (f AGL, AGL*ES, CO<sub>2</sub>, GDPG, UP, INF).</p>
Full article ">
18 pages, 1157 KiB  
Article
From Development to Validation: Exploring the Efficiency of Numetrive, a Computerized Adaptive Assessment of Numerical Reasoning
by Marianna Karagianni and Ioannis Tsaousis
Behav. Sci. 2025, 15(3), 268; https://doi.org/10.3390/bs15030268 - 25 Feb 2025
Viewed by 169
Abstract
The goal of the present study is to describe the methods used to assess the effectiveness and psychometric properties of Numetrive, a newly developed computerized adaptive testing system that measures numerical reasoning. For this purpose, an item bank was developed consisting of 174 [...] Read more.
The goal of the present study is to describe the methods used to assess the effectiveness and psychometric properties of Numetrive, a newly developed computerized adaptive testing system that measures numerical reasoning. For this purpose, an item bank was developed consisting of 174 items concurrently equated and calibrated using the two-parameter logistic model (2PLM), with item difficulties ranging between −3.4 and 2.7 and discriminations spanning from 0.51 up to 1.6. Numetrive constitutes an algorithmic combination that includes maximum likelihood estimation with fences (MLEF) for θ estimation, progressive restricted standard error (PRSE) for item selection and exposure control, and standard error of estimation as the termination rule. The newly developed CAT was evaluated in a Monte Carlo simulation study and was found to perform highly efficiently. The study demonstrated that on average 13.6 items were administered to 5000 simulees while the exposure rates remained significantly low. Additionally, the accuracy in determining the ability scores of the participants was exceptionally high as indicated by various statistical indices, including the bias statistic, mean absolute error (MAE), and root mean square error (RMSE). Finally, a validity study was performed, aimed at evaluating concurrent, convergent, and divergent validity of the newly developed CAT system. Findings verified Numertive’s robustness and applicability in the evaluation of numerical reasoning. Full article
Show Figures

Figure 1

Figure 1
<p>Conditional BIAS (average BIAS in each theta area).</p>
Full article ">Figure 2
<p>Conditional MAE (average MAE in each theta area). Note: CMAE = conditional mean absolute error.</p>
Full article ">Figure 3
<p>Conditional CRMSE (average RMSE in each theta area). Note: CRMSE = conditional root mean square error.</p>
Full article ">Figure 4
<p>Conditional SEE (average SEE in each theta area). Note: SEE: standard error estimate.</p>
Full article ">
17 pages, 1337 KiB  
Article
Chaotic Real Number Generator with Quantum Wave Equation
by Muharrem Tuncay Gençoğlu, Özgür Karaduman and Fatih Özkaynak
Symmetry 2025, 17(3), 349; https://doi.org/10.3390/sym17030349 - 25 Feb 2025
Viewed by 115
Abstract
Randomness plays a crucial role in numerous applications, with cryptography being one of the most significant areas where its importance is evident. A major challenge in cryptographic applications is designing a reliable key generator that meets stringent security requirements. Existing methods often suffer [...] Read more.
Randomness plays a crucial role in numerous applications, with cryptography being one of the most significant areas where its importance is evident. A major challenge in cryptographic applications is designing a reliable key generator that meets stringent security requirements. Existing methods often suffer from predictability and fail to provide robust randomness, necessitating novel mathematical approaches. In this study, we propose an innovative mathematical framework that integrates quantum wave functions with chaotic systems to enhance the unpredictability and security of random number generation. The proposed approach leverages the inherent uncertainty of quantum mechanics and the dynamic behavior of chaos to generate statistically strong random sequences. The analysis results confirm that the proposed generator successfully passes all standard statistical randomness tests, demonstrating its effectiveness in cryptographic applications. Additionally, we present a practical implementation of the proposed method as an image encryption algorithm, showcasing its potential for real-world information security solutions. The findings suggest that this approach can contribute significantly to secure communication systems, financial transactions, and other domains requiring high-level cryptographic security. Full article
(This article belongs to the Special Issue Symmetries and Symmetry-Breaking in Data Security)
Show Figures

Figure 1

Figure 1
<p>The particle situation in a box.</p>
Full article ">Figure 2
<p>Overview of the proposed random number generator.</p>
Full article ">Figure 3
<p>Proposed post-processing approach.</p>
Full article ">Figure 4
<p>Proposed approach to solve the correlation problem.</p>
Full article ">
22 pages, 256 KiB  
Article
Navigating Environmental Uncertainty: The Role of ESG Performance in Driving Firm-Level High-Quality Development
by Yatao Zhang, Qi Ban and Jialing Li
Sustainability 2025, 17(5), 1947; https://doi.org/10.3390/su17051947 - 25 Feb 2025
Viewed by 127
Abstract
Total factor productivity serves as a critical indicator of high-quality corporate development. This study systematically examines the impact of ESG performance on TFP using panel data from Shanghai and Shenzhen A-share listed firms spanning 2009 to 2023. The findings reveal three key insights: [...] Read more.
Total factor productivity serves as a critical indicator of high-quality corporate development. This study systematically examines the impact of ESG performance on TFP using panel data from Shanghai and Shenzhen A-share listed firms spanning 2009 to 2023. The findings reveal three key insights: first, corporate ESG performance significantly enhances TFP, with regression analysis demonstrating a statistically robust positive correlation (1% significance level) and high explanatory power (R2 > 0.8). Second, under environmental uncertainty, ESG-driven total factor productivity improvements operate through dual mechanisms: energy conservation and resource allocation optimisation. Third, heterogeneity analysis highlights that non-state-owned enterprises exhibit a more pronounced relationship compared to state-owned counterparts, particularly in high-environmental-uncertainty scenarios. Beyond enriching academic discourse on ESG metrics, this research elucidates the intrinsic linkage between ESG practices and TFP under dynamic environmental conditions, offering actionable strategies for firms to align sustainability goals with productivity growth. For international stakeholders, this study provides empirical evidence from China—the world’s second-largest economy—to inform global ESG policy design and cross-border investment decisions, emphasising the role of institutional contexts in sustainability-driven value creation. The insights are pivotal for investors, policymakers, and multinational corporations seeking to navigate ESG complexities while advancing sustainable development goals in emerging markets. Full article
25 pages, 4930 KiB  
Article
Implementation of a Data-Parallel Approach on a Lightweight Hash Function for IoT Devices
by Abdullah Sevin
Mathematics 2025, 13(5), 734; https://doi.org/10.3390/math13050734 - 24 Feb 2025
Viewed by 126
Abstract
The Internet of Things is used in many application areas in our daily lives. Ensuring the security of valuable data transmitted over the Internet is a crucial challenge. Hash functions are used in cryptographic applications such as integrity, authentication and digital signatures. Existing [...] Read more.
The Internet of Things is used in many application areas in our daily lives. Ensuring the security of valuable data transmitted over the Internet is a crucial challenge. Hash functions are used in cryptographic applications such as integrity, authentication and digital signatures. Existing lightweight hash functions leverage task parallelism but provide limited scalability. There is a need for lightweight algorithms that can efficiently utilize multi-core platforms or distributed computing environments with high degrees of parallelization. For this purpose, a data-parallel approach is applied to a lightweight hash function to achieve massively parallel software. A novel structure suitable for data-parallel architectures, inspired by basic tree construction, is designed. Furthermore, the proposed hash function is based on a lightweight block cipher and seamlessly integrated into the designed framework. The proposed hash function satisfies security requirements, exhibits high efficiency and achieves significant parallelism. Experimental results indicate that the proposed hash function performs comparably to the BLAKE implementation, with slightly slower execution for large message sizes but marginally better performance for smaller ones. Notably, it surpasses all other evaluated algorithms by at least 20%, maintaining a consistent 20% advantage over Grostl across all data sizes. Regarding parallelism, the proposed PLWHF achieves a speedup of approximately 40% when scaling from one to two threads and 55% when increasing to three threads. Raspberry Pi 4-based tests for IoT applications have also been conducted, demonstrating the hash function’s effectiveness in memory-constrained IoT environments. Statistical tests demonstrate a precision of ±0.004, validate the hypothesis in distribution tests and indicate a deviation of ±0.05 in collision tests, confirming the robustness of the proposed design. Full article
(This article belongs to the Section E1: Mathematics and Computer Science)
Show Figures

Figure 1

Figure 1
<p>Hash function in IoT application.</p>
Full article ">Figure 2
<p>Basic hash properties.</p>
Full article ">Figure 3
<p>The general structure of the proposed hash function.</p>
Full article ">Figure 4
<p>The SIMON block cipher with CTR mode.</p>
Full article ">Figure 5
<p>The sequences of the bit binary presentations under six conditions.</p>
Full article ">Figure 6
<p>Distribution of changed bit numbers.</p>
Full article ">Figure 7
<p>Statistical histogram of changed bit numbers.</p>
Full article ">Figure 8
<p>Frequency distribution of hash value.</p>
Full article ">Figure 9
<p>The variance in bit location index.</p>
Full article ">Figure 10
<p>Comparison of running times.</p>
Full article ">Figure 11
<p>Comparison of running times for various threads.</p>
Full article ">Figure 12
<p>Multi-threading performance on Raspberry Pi 4.</p>
Full article ">
18 pages, 5736 KiB  
Article
Surveillance of Pediatric Invasive Bacterial Diseases in the Veneto Region: Epidemiological Trends and Outcomes over 17 Years (2007–2023)
by Silvia Cocchio, Claudia Cozzolino, Andrea Cozza, Patrizia Furlan, Enrica Frasson, Sara Tarantino, Elisabetta Conte, Lorenzo Chiusaroli, Irene Amoruso, Francesca Zanella, Davide Gentili, Michele Tonon, Francesca Russo, Tatjana Baldovin and Vincenzo Baldo
Vaccines 2025, 13(3), 230; https://doi.org/10.3390/vaccines13030230 - 24 Feb 2025
Viewed by 171
Abstract
Introduction: Invasive bacterial diseases (IBDs) such as meningitis and sepsis are significant public health concerns, particularly in pediatric populations. This study analyzes the incidence, outcomes, and bacterial serotype distribution of pediatric IBDs in the Veneto Region over 17 years. Methods: An observational study [...] Read more.
Introduction: Invasive bacterial diseases (IBDs) such as meningitis and sepsis are significant public health concerns, particularly in pediatric populations. This study analyzes the incidence, outcomes, and bacterial serotype distribution of pediatric IBDs in the Veneto Region over 17 years. Methods: An observational study was conducted using data (2007–2023) from the surveillance system of the Veneto Region, including microbiologically confirmed cases in individuals < 18 years. Differences by age groups and trends were statistically assessed. Results: A total of 535 pediatric IBD cases were reported, with Streptococcus pneumoniae (54.6%), Neisseria meningitidis (19.6%), and Streptococcus agalactiae (13.5%) being the most common pathogens. Haemophilus influenzae infections were more commonly represented in infants under 1 year (41.5%), whereas S. pneumoniae and N. meningitidis were more frequent in the 1–4-year age group (40.8% and 37.1%, respectively). Sepsis was the most common clinical presentation (57.2%), followed by meningitis (36.3%). Serotype analysis revealed that S. pneumoniae serotype 3 was the most prevalent, while serogroup B dominated N. meningitidis cases. Temporal trends generally showed a decline in cases until 2019, a drop during the COVID-19 pandemic, and a subsequent resurgence in 2022–2023. Conclusions: Our research underscores the value of evidence-based epidemiology through robust surveillance systems in tracking IBD trends and serotype shifts, essential for guiding vaccination strategies and public health interventions. These insights highlight the effectiveness of vaccination programs and the necessity of ongoing monitoring to inform public health policies. Improved data integration and completeness are recommended to enhance surveillance accuracy. Full article
(This article belongs to the Special Issue Vaccination and Public Health in the 21st Century)
Show Figures

Figure 1

Figure 1
<p>Age-specific trend of pediatric invasive bacterial disease notification rates per 100,000 in the Veneto Region from 2007 to 2023.</p>
Full article ">Figure 2
<p>Trend of pediatric invasive bacterial disease notification rate per 100,000 in the Veneto Region from 2007 to 2023, stratified by etiological agent and age group.</p>
Full article ">Figure 3
<p>Trend of serotype distribution for pediatric (0–17 years) invasive bacterial disease caused by <span class="html-italic">Streptococcus pneumoniae</span> (<b>A</b>), <span class="html-italic">Neisseria meningitidis</span> (<b>B</b>), and <span class="html-italic">Haemophilus influenzea</span> (<b>C</b>) in the Veneto Region from 2007 to 2023.</p>
Full article ">
25 pages, 3894 KiB  
Article
Data-Driven Analysis of Causes and Risk Assessment of Marine Container Losses: Development of a Predictive Model Using Machine Learning and Statistical Approaches
by Myung-Su Yi, Byung-Keun Lee and Joo-Shin Park
J. Mar. Sci. Eng. 2025, 13(3), 420; https://doi.org/10.3390/jmse13030420 - 24 Feb 2025
Viewed by 100
Abstract
This study presents a comprehensive, data-driven analysis of the causes and risks associated with container loss during maritime transport, utilizing incident data from 2011 to 2023. By employing advanced statistical analysis, machine-learning techniques, and data preprocessing, the study identifies key factors influencing container [...] Read more.
This study presents a comprehensive, data-driven analysis of the causes and risks associated with container loss during maritime transport, utilizing incident data from 2011 to 2023. By employing advanced statistical analysis, machine-learning techniques, and data preprocessing, the study identifies key factors influencing container loss, including vessel size, incident locations, and primary causes. A predictive model based on decision trees was developed to assess the severity of container loss incidents, while K-means clustering was used to classify incident zones. Adverse weather conditions were found to be the predominant cause, accounting for 57.14% of incidents. The study reveals that larger vessels, despite experiencing fewer incidents, face more severe losses, whereas smaller vessels are more prone to frequent but less severe losses. The decision-tree model demonstrated high accuracy in predicting low-risk incidents but showed limitations in moderate- and high-risk scenarios. The findings underscore the importance of understanding the correlation between vessel parameters and incident outcomes to enhance risk management strategies. The study also highlights the potential for improving predictive capabilities by incorporating environmental data. These insights provide a robust framework for ship owners and maritime authorities to anticipate and mitigate risks, emphasizing the need for continuous monitoring and enhanced safety measures in maritime operations. Full article
(This article belongs to the Section Ocean Engineering)
Show Figures

Figure 1

Figure 1
<p>MOL COMFORT incident (<a href="https://gcaptain.com/mol-comfort-incident-photos/" target="_blank">https://gcaptain.com/mol-comfort-incident-photos/</a> accessed on 10 October 2024).</p>
Full article ">Figure 2
<p>Flowchart of the study.</p>
Full article ">Figure 3
<p>Total number and detailed information about container loss at sea [<a href="#B3-jmse-13-00420" class="html-bibr">3</a>].</p>
Full article ">Figure 4
<p>Histogram of each parameter: (<b>a</b>) LOA, (<b>b</b>) LBP, (<b>c</b>) B, (<b>d</b>) D, (<b>e</b>) Built Year, and (<b>f</b>) Capacity.</p>
Full article ">Figure 4 Cont.
<p>Histogram of each parameter: (<b>a</b>) LOA, (<b>b</b>) LBP, (<b>c</b>) B, (<b>d</b>) D, (<b>e</b>) Built Year, and (<b>f</b>) Capacity.</p>
Full article ">Figure 5
<p>Distribution of container loss causes [<a href="#B4-jmse-13-00420" class="html-bibr">4</a>].</p>
Full article ">Figure 6
<p>Scatter plot of LOA, LBP, B, D, and draught max; (<b>a</b>) LOA-LBP, (<b>b</b>) LOA-B, (<b>c</b>) LOA-D, and (<b>d</b>) LOA-draught Max.</p>
Full article ">Figure 7
<p>Results of the deep-learning model in predicting capacity.</p>
Full article ">Figure 8
<p>Results of the K-means algorithm.</p>
Full article ">Figure 9
<p>Distribution of the loss ratio.</p>
Full article ">Figure 10
<p>Result of the decision tree.</p>
Full article ">
Back to TopTop