Information

Research

Jump to: Review

10 pages, 2129 KiB

Open AccessArticle

Automatic Detection of Camera Rotation Moments in Trans-Nasal Minimally Invasive Surgery Using Machine Learning Algorithm

by Zhong Shi Zhang, Yun Wu and Bin Zheng

Information 2025, 16(4), 303; https://doi.org/10.3390/info16040303 - 11 Apr 2025

Viewed by 213

Abstract

Background: Minimally invasive surgery (MIS) is an advanced surgical technique that relies on a camera to provide the surgeon with a visual field. When the camera rotates along its longitudinal axis, the horizon of the surgical view tilts, increasing the difficulty of the [...] Read more.

Background: Minimally invasive surgery (MIS) is an advanced surgical technique that relies on a camera to provide the surgeon with a visual field. When the camera rotates along its longitudinal axis, the horizon of the surgical view tilts, increasing the difficulty of the procedure and the cognitive load on the surgeon. To address this, we proposed training a convolutional neural network (CNN) to detect camera rotation, laying the groundwork for the automatic correction of this issue during MIS procedures. Methods: We collected trans-nasal MIS procedure videos from YouTube and labeled each frame as either “tilted” or “non-tilted”. The dataset consisted of 2116 video frames, with 497 frames labeled as “tilted” and 1619 frames as “non-tilted”. This dataset was randomly divided into three subsets: training (70%), validation (20%), and testing (10%) Results: The ResNet50 was trained on the dataset for 10 epochs, achieving an accuracy of 96.9% at epoch 6 with a validation loss of 0.0242 before validation accuracy began to decrease. On the test set, the model achieved an accuracy of 96% with an average loss of 0.0256. The final F1 score was 0.94, and the Matthews Correlation Coefficient was 0.9168, with no significant bias toward either class. The trained ResNet50 model demonstrated a high success rate in predicting significant camera rotation without favoring the more frequent class in the dataset. Conclusions: The trained CNN accurately detected camera rotation with high precision, establishing a foundation for developing an automatic correction system for camera rotation in MIS procedures. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

27 pages, 11491 KiB

Open AccessArticle

Detecting Driver Drowsiness Using Hybrid Facial Features and Ensemble Learning

by Changbiao Xu, Wenhao Huang, Jiao Liu and Lang Li

Information 2025, 16(4), 294; https://doi.org/10.3390/info16040294 - 7 Apr 2025

Viewed by 462

Abstract

Drowsiness while driving poses a significant risk in terms of road safety, making effective drowsiness detection systems essential for the prevention of accidents. Facial signal-based detection methods have proven to be an effective approach to drowsiness detection. However, they bring challenges arising from [...] Read more.

Drowsiness while driving poses a significant risk in terms of road safety, making effective drowsiness detection systems essential for the prevention of accidents. Facial signal-based detection methods have proven to be an effective approach to drowsiness detection. However, they bring challenges arising from inter-individual differences among drivers. Variations in facial structure necessitate personalized feature extraction thresholds, yet existing methods apply a uniform threshold, leading to inaccurate feature extraction. Furthermore, many current methods focus on only one or two facial regions, overlooking the possibility that drowsiness may manifest differently across different facial areas among different drivers. To address these issues, we propose a drowsiness detection method that combines an ensemble model with hybrid facial features. This approach enables the accurate extraction of features from four key facial regions—the eye region, mouth contour, head pose, and gaze direction—through adaptive threshold correction to ensure comprehensive coverage. An ensemble model, combining Random Forest, XGBoost, and Multilayer Perceptron with a soft voting criterion, is then employed to classify the drivers’ drowsiness state. Additionally, we use the SHAP method to ensure model explainability and analyze the correlations between features from various facial regions. Trained and tested on the UTA-RLDD dataset, our method achieves a video accuracy (VA) of 86.52%, outperforming similar techniques introduced in recent years. The interpretability analysis demonstrates the value of our approach, offering a valuable reference for future research and contributing significantly to road safety. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

25 pages, 4058 KiB

Open AccessArticle

Kubernetes-Powered Cardiovascular Monitoring: Enhancing Internet of Things Heart Rate Systems for Scalability and Efficiency

by Hans Indrawan Sucipto, Gregorius Natanael Elwirehardja, Nicholas Dominic and Nico Surantha

Information 2025, 16(3), 213; https://doi.org/10.3390/info16030213 - 10 Mar 2025

Viewed by 527

Abstract

Reliable system design is an important component to ensure data processing speed, service availability, and an improved user experience. Several studies have been conducted to provide data processing speeds for health monitors using clouds or edge devices. However, if the system design used [...] Read more.

Reliable system design is an important component to ensure data processing speed, service availability, and an improved user experience. Several studies have been conducted to provide data processing speeds for health monitors using clouds or edge devices. However, if the system design used cannot handle many requests, the reliability of the monitoring itself will be reduced. This study used the Kubernetes approach for system design, leveraging its scalability and efficient resource management. The system was deployed in a local Kubernetes environment using an Intel Xeon CPU E5-1620 with 8 GB RAM. This study compared two architectures: MQTT (traditional method) and MQTT-Kafka (proposed method). The proposed method shows a significant improvement, such as throughput results on the proposed method of 1587 packets/s rather than the traditional methods at 484 packets/s. The response time and latency are 95% more stable than the traditional method, and the performance of the proposed method also requires a larger resource of approximately 30% more than the traditional method. The performance of the proposed method requires the use of a large amount of RAM for a resource-limited environment, with the highest RAM usage at 5.63 Gb, while the traditional method requires 4.5 Gb for the highest RAM requirement. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Graphical abstract

19 pages, 1210 KiB

Open AccessArticle

Applied Machine Learning to Anomaly Detection in Enterprise Purchase Processes: A Hybrid Approach Using Clustering and Isolation Forest

by Antonio Herreros-Martínez, Rafael Magdalena-Benedicto, Joan Vila-Francés, Antonio José Serrano-López, Sonia Pérez-Díaz and José Javier Martínez-Herráiz

Information 2025, 16(3), 177; https://doi.org/10.3390/info16030177 - 26 Feb 2025

Viewed by 759

Abstract

In the era of increasing digitalisation, organisations face the critical challenge of detecting anomalies in large volumes of data, which may indicate suspicious activities. To address this challenge, audit engagements are conducted regularly, and internal auditors and purchasing specialists seek innovative methods to [...] Read more.

In the era of increasing digitalisation, organisations face the critical challenge of detecting anomalies in large volumes of data, which may indicate suspicious activities. To address this challenge, audit engagements are conducted regularly, and internal auditors and purchasing specialists seek innovative methods to streamline these processes. This study introduces a methodology to prioritise the investigation of anomalies identified in two large real-world purchase datasets. The primary objective is to enhance the effectiveness of companies’ control efforts and improve the efficiency of anomaly detection tasks. The approach begins with a comprehensive exploratory data analysis, followed by the application of unsupervised machine learning techniques to identify anomalies. A univariate analysis is performed using the z-Score index and the DBSCAN algorithm, while multivariate analysis employs k-Means clustering and Isolation Forest algorithms. Additionally, the Silhouette index is used to evaluate the quality of the clustering, ensuring each method produces a prioritised list of candidate transactions for further review. To refine this process, an ensemble prioritisation framework is developed, integrating multiple methods. Furthermore, explainability tools such as SHAP are utilised to provide actionable insights and support specialists in interpreting the results. This methodology aims to empower organisations to detect anomalies more effectively and streamline the audit process. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Graphical abstract

20 pages, 8214 KiB

Open AccessArticle

Convolutional Neural Network Applications in Current Sensor Fault Classification Mechanisms in Permanent Magnet Synchronous Motor Drive Systems

by Kamila Jankowska and Mateusz Dybkowski

Information 2025, 16(2), 142; https://doi.org/10.3390/info16020142 - 14 Feb 2025

Viewed by 453

Abstract

In this article, the possibilities of Convolutional Neural Network applications to classify stator current sensor faults in a vector-controlled drive with a Permanent Magnet Synchronous Motor are described. It was assumed that three basic faults, consisting of signal loss from the current sensor, [...] Read more.

In this article, the possibilities of Convolutional Neural Network applications to classify stator current sensor faults in a vector-controlled drive with a Permanent Magnet Synchronous Motor are described. It was assumed that three basic faults, consisting of signal loss from the current sensor, measurement noise, and gain error, can be effectively classified by the Convolutional Neural Networks. This work presents the results obtained in experimental research on a 0.894-kilowatt Permanent Magnet Synchronous Motor. Fault classification is based on raw phase current signals transformed into matrices. Classification is carried out using two neural structures operating in parallel for phases A and B. This article includes a description of the process of selecting input matrices, developing classifiers, and the experimental results in offline classification obtained at the efficiency level of 99.2% and 98.3% for phases A and B, respectively. This research was carried out for various operating conditions of the drive system. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

28 pages, 7894 KiB

Open AccessArticle

Enhancing UAV Security Against GPS Spoofing Attacks Through a Genetic Algorithm-Driven Deep Learning Framework

by Abdallah Al-Sabbagh, Aya El-Bokhary, Sana El-Koussa, Abdulrahman Jaber and Mahmoud Elkhodr

Information 2025, 16(2), 115; https://doi.org/10.3390/info16020115 - 7 Feb 2025

Viewed by 1055

Abstract

Unmanned Aerial Vehicles (UAVs) are increasingly employed across various domains, including communication, military, and delivery operations. Their reliance on the Global Positioning System (GPS) renders them vulnerable to GPS spoofing attacks, in which adversaries transmit false signals to manipulate UAVs’ navigation, potentially leading [...] Read more.

Unmanned Aerial Vehicles (UAVs) are increasingly employed across various domains, including communication, military, and delivery operations. Their reliance on the Global Positioning System (GPS) renders them vulnerable to GPS spoofing attacks, in which adversaries transmit false signals to manipulate UAVs’ navigation, potentially leading to severe security risks. This paper presents an enhanced integration of Long Short-Term Memory (LSTM) networks with a Genetic Algorithm (GA) for GPS spoofing detection. Although GA–neural network combinations have existed for decades, our method expands the GA’s search space to optimize a wider range of hyperparameters, thereby improving adaptability in dynamic operational scenarios. The framework is evaluated using a real-world GPS spoofing dataset that includes authentic and malicious signals under multiple attack conditions. While we discuss strategies for mitigating CPU resource demands and computational overhead, we acknowledge that direct measurements of energy consumption or inference latency are not included in the present work. Experimental results show that the proposed LSTM–GA approach achieved a notable increase in classification accuracy (from 88.42% to 93.12%) and the F1 score (from 87.63% to 93.39%). These findings highlight the system’s potential to strengthen UAV security against GPS spoofing attacks, provided that hardware constraints and other limitations are carefully managed in real deployments. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

24 pages, 7108 KiB

Open AccessArticle

Explainable AI Using On-Board Diagnostics Data for Urban Buses Maintenance Management: A Study Case

by Bernardo Tormos, Benjamín Pla, Ramón Sánchez-Márquez and Jose Luis Carballo

Information 2025, 16(2), 74; https://doi.org/10.3390/info16020074 - 21 Jan 2025

Viewed by 890

Abstract

Industry 4.0, leveraging tools like AI and the massive generation of data, is driving a paradigm shift in maintenance management. Specifically, in the realm of Artificial Intelligence (AI), traditionally “black box” models are now being unveiled through explainable AI techniques, which provide insights [...] Read more.

Industry 4.0, leveraging tools like AI and the massive generation of data, is driving a paradigm shift in maintenance management. Specifically, in the realm of Artificial Intelligence (AI), traditionally “black box” models are now being unveiled through explainable AI techniques, which provide insights into model decision-making processes. This study addresses the underutilization of these techniques alongside On-Board Diagnostics data by maintenance management teams in urban bus fleets for addressing key issues affecting vehicle reliability and maintenance needs. In the context of urban bus fleets, diesel particulate filter regeneration processes frequently operate under suboptimal conditions, accelerating engine oil degradation and increasing maintenance costs. Due to limited documentation on the control system of the filter, the maintenance team faces obstacles in proposing solutions based on a comprehensive understanding of the system’s behavior and control logic. The objective of this study is to analyze and predict the various states during the diesel particulate filter regeneration process using Machine Learning and explainable artificial intelligence techniques. The insights obtained aim to provide the maintenance team with a deeper understanding of the filter’s control logic, enabling them to develop proposals grounded in a comprehensive understanding of the system. This study employs a combination of traditional Machine Learning models, including XGBoost, LightGBM, Random Forest, and Support Vector Machine. The target variable, representing three possible regeneration states, was transformed using a one-vs-rest approach, resulting in three binary classification tasks where each target state was individually classified against all other states. Additionally, explainable AI techniques such as Shapley Additive Explanations, Partial Dependence Plots, and Individual Conditional Expectation were applied to interpret and visualize the conditions influencing each regeneration state. The results successfully associate two states with specific operating conditions and establish operational thresholds for key variables, offering practical guidelines for optimizing the regeneration process. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

12 pages, 1582 KiB

Open AccessArticle

Detection of the Leg-Crossing Position Using Pressure Distribution Sensor and Machine Learning

by Emi Yuda, Tomoki Ando and Yutaka Yoshida

Information 2024, 15(12), 810; https://doi.org/10.3390/info15120810 - 17 Dec 2024

Viewed by 879

Abstract

Humans often cross their legs unconsciously while sitting, which can lead to health problems such as shifts in the center of gravity, lower back pain, reduced blood circulation, and pelvic distortion. Detecting unconscious leg crossing is important for promoting correct posture. In this [...] Read more.

Humans often cross their legs unconsciously while sitting, which can lead to health problems such as shifts in the center of gravity, lower back pain, reduced blood circulation, and pelvic distortion. Detecting unconscious leg crossing is important for promoting correct posture. In this study, we investigated the detection of leg-crossing postures using machine learning algorithms applied to data from body pressure distribution sensors. Pressure data were collected over 180 s from four male subjects (25.8 ± 6.29 years old) under three conditions: no leg crossing, right-leg crossing, and left-leg crossing. Seven classifiers, including support vector machine (SVM), random forest (RF), and k-nearest neighbors (k-NN), were evaluated based on accuracy, recall, precision, and specificity. Among the tested methods, k-NN demonstrated the highest classification performance, suggesting it may be the most effective approach for identifying leg-crossing postures in this study. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

20 pages, 5100 KiB

Open AccessArticle

Neurophysiological Approach for Psychological Safety: Enhancing Mental Health in Human–Robot Collaboration in Smart Manufacturing Setups Using Neuroimaging

by Arshia Arif, Zohreh Zakeri, Ahmet Omurtag, Philip Breedon and Azfar Khalid

Information 2024, 15(10), 640; https://doi.org/10.3390/info15100640 - 15 Oct 2024

Viewed by 1337

Abstract

Human–robot collaboration (HRC) has become increasingly prevalent due to innovative advancements in the automation industry, especially in manufacturing setups. Although HRC increases productivity and efficacy, it exposes human workers to psychological stress while interfacing with collaborative robotic systems as robots may not provide [...] Read more.

Human–robot collaboration (HRC) has become increasingly prevalent due to innovative advancements in the automation industry, especially in manufacturing setups. Although HRC increases productivity and efficacy, it exposes human workers to psychological stress while interfacing with collaborative robotic systems as robots may not provide visual or auditory cues. It is crucial to comprehend how HRC impacts mental stress in order to enhance occupational safety and well-being. Though academics and industrial interest in HRC is expanding, safety and mental stress problems are still not adequately studied. In particular, human coworkers’ cognitive strain during HRC has not been explored well, although being fundamental to sustaining a secure and constructive workplace environment. This study, therefore, aims to monitor the mental stress of factory workers during HRC using behavioural, physiological and subjective measures. Physiological measures, being objective and more authentic, have the potential to replace conventional measures i.e., behavioural and subjective measures, if they demonstrate a good correlation with traditional measures. Two neuroimaging modalities including electroencephalography (EEG) and functional near-infrared spectroscopy (fNIRS) have been used as physiological measures to track neuronal and hemodynamic activity of the brain, respectively. Here, the correlation between physiological data and behavioural and subjective measurements has been ascertained through the implementation of seven different machine learning algorithms. The results imply that the EEG and fNIRS features combined produced the best results for most of the targets. For subjective measures being the target, linear regression has outperformed all other models, whereas tree and ensemble performed the best for predicting the behavioural measures. The outcomes indicate that physiological measures have the potential to be more informative and often substitute other skewed metrics. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

22 pages, 3158 KiB

Open AccessArticle

Sensitivity Analysis of Traffic Sign Recognition to Image Alteration and Training Data Size

by Arthur Rubio, Guillaume Demoor, Simon Chalmé, Nicolas Sutton-Charani and Baptiste Magnier

Information 2024, 15(10), 621; https://doi.org/10.3390/info15100621 - 10 Oct 2024

Viewed by 1781

Abstract

Accurately classifying road signs is crucial for autonomous driving due to the high stakes involved in ensuring safety and compliance. As Convolutional Neural Networks (CNNs) have largely replaced traditional Machine Learning models in this domain, the demand for substantial training data has increased. [...] Read more.

Accurately classifying road signs is crucial for autonomous driving due to the high stakes involved in ensuring safety and compliance. As Convolutional Neural Networks (CNNs) have largely replaced traditional Machine Learning models in this domain, the demand for substantial training data has increased. This study aims to compare the performance of classical Machine Learning (ML) models and Deep Learning (DL) models under varying amounts of training data, particularly focusing on altered signs to mimic real-world conditions. We evaluated three classical models: Support Vector Machine (SVM), Random Forest, and Linear Discriminant Analysis (LDA), and one Deep Learning model: Convolutional Neural Network (CNN). Using the German Traffic Sign Recognition Benchmark (GTSRB) dataset, which includes approximately 40,000 German traffic signs, we introduced digital alterations to simulate conditions such as environmental wear or vandalism. Additionally, the Histogram of Oriented Gradients (HOG) descriptor was used to assist classical models. Bayesian optimization and k-fold cross-validation were employed for model fine-tuning and performance assessment. Our findings reveal a threshold in training data beyond which accuracy plateaus. Classical models showed a linear performance decrease under increasing alteration, while CNNs, despite being more robust to alterations, did not significantly outperform classical models in overall accuracy. Ultimately, classical Machine Learning models demonstrated performance comparable to CNNs under certain conditions, suggesting that effective road sign classification can be achieved with less computationally intensive approaches. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Graphical abstract

47 pages, 17094 KiB

Open AccessArticle

Short-Term Water Demand Forecasting from Univariate Time Series of Water Reservoir Stations

by Georgios Myllis, Alkiviadis Tsimpiris and Vasiliki Vrana

Information 2024, 15(10), 605; https://doi.org/10.3390/info15100605 - 3 Oct 2024

Cited by 1 | Viewed by 1137

Abstract

This study presents an improved data-centric approach to short-term water demand forecasting using univariate time series from water reservoir levels. The dataset comprises water level recordings from 21 reservoirs in Eastern Thessaloniki collected over 15 months via a SCADA system provided by the [...] Read more.

This study presents an improved data-centric approach to short-term water demand forecasting using univariate time series from water reservoir levels. The dataset comprises water level recordings from 21 reservoirs in Eastern Thessaloniki collected over 15 months via a SCADA system provided by the water company EYATH S.A. The methodology involves data preprocessing, anomaly detection, data imputation, and the application of predictive models. Techniques such as the Interquartile Range method and moving standard deviation are employed to identify and handle anomalies. Missing values are imputed using LSTM networks optimized through the Optuna framework. This study emphasizes a data-centric approach in deep learning, focusing on improving data quality before model application, which has proven to enhance prediction accuracy. This strategy is crucial, especially in regions where reservoirs are the primary water source, and demand distribution cannot be solely determined by flow meter readings. LSTM, Random Forest Regressor, ARIMA, and SARIMA models are utilized to extract and analyze water level trends, enabling more accurate future water demand predictions. Results indicate that combining deep learning techniques with traditional statistical models significantly improves the accuracy and reliability of water demand predictions, providing a robust framework for optimizing water resource management. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

16 pages, 2706 KiB

Open AccessArticle

Classification of Moral Decision Making in Autonomous Driving: Efficacy of Boosting Procedures

by Amandeep Singh, Yovela Murzello, Sushil Pokhrel and Siby Samuel

Information 2024, 15(9), 562; https://doi.org/10.3390/info15090562 - 11 Sep 2024

Cited by 2 | Viewed by 1452

Abstract

Autonomous vehicles (AVs) face critical decisions in pedestrian interactions, necessitating ethical considerations such as minimizing harm and prioritizing human life. This study investigates machine learning models to predict human decision making in simulated driving scenarios under varying pedestrian configurations and time constraints. Data [...] Read more.

Autonomous vehicles (AVs) face critical decisions in pedestrian interactions, necessitating ethical considerations such as minimizing harm and prioritizing human life. This study investigates machine learning models to predict human decision making in simulated driving scenarios under varying pedestrian configurations and time constraints. Data were collected from 204 participants across 12 unique simulated driving scenarios, categorized into young (24.7 ± 3.5 years, 38 males, 64 females) and older (71.0 ± 5.7 years, 59 males, 43 females) age groups. Participants’ binary decisions to maintain or change lanes were recorded. Traditional logistic regression models exhibited high precision but consistently low recall, struggling to identify true positive instances requiring intervention. In contrast, the AdaBoost algorithm demonstrated superior accuracy and discriminatory power. Confusion matrix analysis revealed AdaBoost’s ability to achieve high true positive rates (up to 96%) while effectively managing false positives and negatives, even under 1 s time constraints. Learning curve analysis confirmed robust learning without overfitting. AdaBoost consistently outperformed logistic regression, with AUC-ROC values ranging from 0.82 to 0.96. It exhibited strong generalization, with validation accuracy approaching 0.8, underscoring its potential for reliable real-world AV deployment. By consistently identifying critical instances while minimizing errors, AdaBoost can prioritize human safety and align with ethical frameworks essential for responsible AV adoption. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

25 pages, 4636 KiB

Open AccessArticle

Application of Multi-Source Remote Sensing Data and Machine Learning for Surface Soil Moisture Mapping in Temperate Forests of Central Japan

by Kyaw Win, Tamotsu Sato and Satoshi Tsuyuki

Information 2024, 15(8), 485; https://doi.org/10.3390/info15080485 - 15 Aug 2024

Cited by 3 | Viewed by 2740

Abstract

Surface soil moisture (SSM) is a key parameter for land surface hydrological processes. In recent years, satellite remote sensing images have been widely used for SSM estimation, and many methods based on satellite-derived spectral indices have also been used to estimate the SSM [...] Read more.

Surface soil moisture (SSM) is a key parameter for land surface hydrological processes. In recent years, satellite remote sensing images have been widely used for SSM estimation, and many methods based on satellite-derived spectral indices have also been used to estimate the SSM content in various climatic conditions and geographic locations. However, achieving an accurate estimation of SSM content at a high spatial resolution remains a challenge. Therefore, improving the precision of SSM estimation through the synergies of multi-source remote sensing data has become imperative, particularly for informing forest management practices. In this study, the integration of multi-source remote sensing data with random forest and support vector machine models was conducted using Google Earth Engine in order to estimate the SSM content and develop SSM maps for temperate forests in central Japan. The synergy of Sentinel-2 and terrain factors, such as elevation, slope, aspect, slope steepness, and valley depth, with the random forest model provided the most suitable approach for SSM estimation, yielding the highest accuracy values (overall accuracy for testing = 91.80%, Kappa = 87.18%, r = 0.98) for the temperate forests of central Japan. This finding provides more valuable information for SSM mapping, which shows promise for precision forestry applications. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

19 pages, 2777 KiB

Open AccessArticle

Fabric Defect Detection in Real World Manufacturing Using Deep Learning

by Mariam Nasim, Rafia Mumtaz, Muneer Ahmad and Arshad Ali

Information 2024, 15(8), 476; https://doi.org/10.3390/info15080476 - 11 Aug 2024

Cited by 7 | Viewed by 7045

Abstract

Defect detection is very important for guaranteeing the quality and pricing of fabric. A considerable amount of fabric is discarded as waste because of defects, leading to substantial annual losses. While manual inspection has traditionally been the norm for detection, adopting an automatic [...] Read more.

Defect detection is very important for guaranteeing the quality and pricing of fabric. A considerable amount of fabric is discarded as waste because of defects, leading to substantial annual losses. While manual inspection has traditionally been the norm for detection, adopting an automatic defect detection scheme based on a deep learning model offers a timely and efficient solution for assessing fabric quality. In real-time manufacturing scenarios, datasets lack high-quality, precisely positioned images. Moreover, both plain and printed fabrics are being manufactured in industries simultaneously; therefore, a single model should be capable of detecting defects in all kinds of fabric. So training a robust deep learning model that detects defects in fabric datasets generated during production with high accuracy and lower computational costs is required. This study uses an indigenous dataset directly sourced from Chenab Textiles, providing authentic and diverse images representative of actual manufacturing conditions. The dataset is used to train a computationally faster but lighter state-of-the-art network, i.e., YOLOv8. For comparison, YOLOv5 and MobileNetV2-SSD FPN-Lite models are also trained on the same dataset. YOLOv8n achieved the highest performance, with a mAP of 84.8%, precision of 0.818, and recall of 0.839 across seven different defect classes. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

13 pages, 4630 KiB

Open AccessArticle

AquaVision: AI-Powered Marine Species Identification

by Benjamin Mifsud Scicluna, Adam Gauci and Alan Deidun

Information 2024, 15(8), 437; https://doi.org/10.3390/info15080437 - 27 Jul 2024

Cited by 2 | Viewed by 3200

Abstract

This study addresses the challenge of accurately identifying fish species by using machine learning and image classification techniques. The primary aim is to develop an innovative algorithm that can dynamically identify the most common (within Maltese coastal waters) invasive Mediterranean fish species based [...] Read more.

This study addresses the challenge of accurately identifying fish species by using machine learning and image classification techniques. The primary aim is to develop an innovative algorithm that can dynamically identify the most common (within Maltese coastal waters) invasive Mediterranean fish species based on available images. In particular, these include Fistularia commersonii, Lobotes surinamensis, Pomadasys incisus, Siganus luridus, and Stephanolepis diaspros, which have been adopted as this study’s target species. Through the use of machine-learning models and transfer learning, the proposed solution seeks to enable precise, on-the-spot species recognition. The methodology involved collecting and organising images as well as training the models with consistent datasets to ensure comparable results. After trying a number of models, ResNet18 was found to be the most accurate and reliable, with YOLO v8 following closely behind. While the performance of YOLO was reasonably good, it exhibited less consistency in its results. These results underline the potential of the developed algorithm to significantly aid marine biology research, including citizen science initiatives, and promote environmental management efforts through accurate fish species identification. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

16 pages, 21213 KiB

Open AccessArticle

A Lightweight Face Detector via Bi-Stream Convolutional Neural Network and Vision Transformer

by Zekun Zhang, Qingqing Chao, Shijie Wang and Teng Yu

Information 2024, 15(5), 290; https://doi.org/10.3390/info15050290 - 20 May 2024

Cited by 1 | Viewed by 1570

Abstract

Lightweight convolutional neural networks are widely used for face detection due to their ability to learn local representations through spatial induction bias and translational invariance. However, convolutional face detectors have limitations in detecting faces under challenging conditions like occlusion, blurring, or changes in [...] Read more.

Lightweight convolutional neural networks are widely used for face detection due to their ability to learn local representations through spatial induction bias and translational invariance. However, convolutional face detectors have limitations in detecting faces under challenging conditions like occlusion, blurring, or changes in facial poses, primarily attributed to fixed-size receptive fields and a lack of global modeling. Transformer-based models have advantages on learning global representations but are insensitive to capture local patterns. To address these limitations, we propose an efficient face detector that combines convolutional neural network and transformer architectures. We introduce a bi-stream structure that integrates convolutional neural network and transformer blocks within the backbone network, enabling the preservation of local pattern features and the extraction of global context. To further preserve the local details captured by convolutional neural networks, we propose a feature enhancement convolution block in a hierarchical backbone structure. Additionally, we devise a multiscale feature aggregation module to enhance obscured and blurred facial features. Experimental results demonstrate that our method has achieved improved lightweight face detection accuracy with an average precision of 95.30%, 94.20%, and 87.56% across the easy, medium, and hard subdatasets of WIDER FACE, respectively. Therefore, we believe our method will be a useful supplement to the collection of current artificial intelligence models and benefit the engineering applications of face detection. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

17 pages, 1313 KiB

Open AccessArticle

Using Generative AI to Improve the Performance and Interpretability of Rule-Based Diagnosis of Type 2 Diabetes Mellitus

by Leon Kopitar, Iztok Fister, Jr. and Gregor Stiglic

Information 2024, 15(3), 162; https://doi.org/10.3390/info15030162 - 12 Mar 2024

Viewed by 3145

Abstract

Introduction: Type 2 diabetes mellitus is a major global health concern, but interpreting machine learning models for diagnosis remains challenging. This study investigates combining association rule mining with advanced natural language processing to improve both diagnostic accuracy and interpretability. This novel approach has [...] Read more.

Introduction: Type 2 diabetes mellitus is a major global health concern, but interpreting machine learning models for diagnosis remains challenging. This study investigates combining association rule mining with advanced natural language processing to improve both diagnostic accuracy and interpretability. This novel approach has not been explored before in using pretrained transformers for diabetes classification on tabular data. Methods: The study used the Pima Indians Diabetes dataset to investigate Type 2 diabetes mellitus. Python and Jupyter Notebook were employed for analysis, with the NiaARM framework for association rule mining. LightGBM and the dalex package were used for performance comparison and feature importance analysis, respectively. SHAP was used for local interpretability. OpenAI GPT version 3.5 was utilized for outcome prediction and interpretation. The source code is available on GitHub. Results: NiaARM generated 350 rules to predict diabetes. LightGBM performed better than the GPT-based model. A comparison of GPT and NiaARM rules showed disparities, prompting a similarity score analysis. LightGBM’s decision making leaned heavily on glucose, age, and BMI, as highlighted in feature importance rankings. Beeswarm plots demonstrated how feature values correlate with their influence on diagnosis outcomes. Discussion: Combining association rule mining with GPT for Type 2 diabetes mellitus classification yields limited effectiveness. Enhancements like preprocessing and hyperparameter tuning are required. Interpretation challenges and GPT’s dependency on provided rules indicate the necessity for prompt engineering and similarity score methods. Variations in feature importance rankings underscore the complexity of T2DM. Concerns regarding GPT’s reliability emphasize the importance of iterative approaches for improving prediction accuracy. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

Review

Jump to: Research

14 pages, 1414 KiB

Open AccessReview

The Use of AI in Software Engineering: A Synthetic Knowledge Synthesis of the Recent Research Literature

by Peter Kokol

Information 2024, 15(6), 354; https://doi.org/10.3390/info15060354 - 14 Jun 2024

Cited by 4 | Viewed by 9011

Abstract

Artificial intelligence (AI) has witnessed an exponential increase in use in various applications. Recently, the academic community started to research and inject new AI-based approaches to provide solutions to traditional software-engineering problems. However, a comprehensive and holistic understanding of the current status needs [...] Read more.

Artificial intelligence (AI) has witnessed an exponential increase in use in various applications. Recently, the academic community started to research and inject new AI-based approaches to provide solutions to traditional software-engineering problems. However, a comprehensive and holistic understanding of the current status needs to be included. To close the above gap, synthetic knowledge synthesis was used to induce the research landscape of the contemporary research literature on the use of AI in software engineering. The synthesis resulted in 15 research categories and 5 themes—namely, natural language processing in software engineering, use of artificial intelligence in the management of the software development life cycle, use of machine learning in fault/defect prediction and effort estimation, employment of deep learning in intelligent software engineering and code management, and mining software repositories to improve software quality. The most productive country was China (n = 2042), followed by the United States (n = 1193), India (n = 934), Germany (n = 445), and Canada (n = 381). A high percentage (n = 47.4%) of papers were funded, showing the strong interest in this research topic. The convergence of AI and software engineering can significantly reduce the required resources, improve the quality, enhance the user experience, and improve the well-being of software developers. Full article

(This article belongs to the Special Issue Machine Learning and Artificial Intelligence with Applications)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Machine Learning and Artificial Intelligence with Applications

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Benefits of Publishing in a Special Issue

Published Papers (18 papers)

Research

Review

Further Information

Guidelines

MDPI Initiatives

Follow MDPI