Special Issue on Innovation in Computing, Engineering Science & Technology 2023

The detection and classification of fish is a prevalent and fascinating area of study. Numerous researchers develop skills in fish recognition in both aquatic and non-aquatic environments, which is beneficial for population control and the aquaculture industry, respectively. Rarely is research conducted to optimize the recognition of fish with diverse backgrounds. This paper proposes a method for fish recognition that uses the landmarking technique to optimize YOLO version 4 to detect and classify fish with varying background conditions, making it applicable for both underwater and terrestrial fish recognition. The proposed procedure was evaluated on the Bringham Young University (BYU) dataset containing four different fish species. The final test results indicate that the detection accuracy had reached 96.60% with an average confidence score of 99.67% at the 60% threshold. The outcome is 4,94% better than the most common traditional labeling method.

Read more…

Currently, the quality and quantity of product must be inspected before transporting. Currently the popular unsealing box product inspecting is performed by weighing the box where the errors occur according to the tolerance of the weighting machine and tolerance weight of the product. On the other hand, the quantity of product can be inspected automatically using the image processing and recognition where the sealed box needs to be unpacked which is difficult to be implemented in practice. Then, the error in product transportation causes a loss profit of for the vendor and losing the reliability from customers. In this paper, we proposed a new volume estimation technique to estimate the product quantity in the sealed box using RSSI with machine learning for improving the monitoring accuracy. The proposed system includes one transmitter on the top and five receivers at bottom of the package. Based on practical environment, we align the product’s pattern inside the boxed package into two cases including horizontal/vertical aligned pattern and random pattern. In the experiment, we compare the volume estimation accuracy of five machine learning techniques including linear regression, logistic regression, Naïve Bayes, KNN, and NN. From the results, the NN method provide the highest volume estimation accuracy among others and consumes the shortest estimation time. NN presents accuracy as 99.4% and consumes 6.51 milliseconds of estimation time. Moreover, for protecting the products from the delivery process, shockproof material must be put to cover the product in the box. Three shockproof types are considered in our experiment such as bubble wrap, paper, and airbag and our proposed system is considered all kinds of shockproof situations. The suggested method can estimate the volume of products without necessitating their opening or destruction. The suggested approach is also resistant to the impacts of packaging cushioning material.

Read more…

The quality of the electric power system must be maintained properly, one of which is voltage. Under certain operating conditions, the bus voltage may drop below its minimum level, called bad voltage. A large reactive load can cause a voltage drop across the bus or the location of the bus is far from the generator stations, so that the line impedance has a large value. One technique to increase the voltage is through compensation technique. This paper presents the determination of the minimum statistical VAR compensation for increasing the voltage to the minimum safety limit. The methodology creates a compensating model for bad bus voltages. Electrical quantities (voltage, power and system losses) are calculated through the power flow technique. The compensating capacity is increased until the voltage rise reaches its minimum security limit. The simulation results on the IEEE 9 bus system show that the voltage increases on all buses with minimal compensation on buses 5 and 8, and can save up to 1.37 MW of power

Read more…

This article explains the fundamental principles of the economic policy that are integrated into the decentralized public platform Waterfall. The platform has a DAG (Directed Acyclic Graph) based system architecture and is designed to develop decentralized applications and financial services. The main goal of this work is to create a favorable environment that incentivizes positive behavior from each network participant and from the system as a whole. Economic leverages ensure general equilibrium to provide an optimal data replication ratio, attack protection, and affordable transaction fees. Although this model of tokenomic is designed explicitly for the current version of the Waterfall platform named Salto Collazo, the presented approaches possess the potential to be applied across a broad spectrum of decentralized public platforms, owing to their inherent transparency and a set of tuned parameters.

Read more…

The article presents a simulation of obstacle detection based on noise-free optical flow images for motion control of mobile robots. The detection of hazardous areas in optical flow images is accomplished by dividing multiple layers of optical flow vectors into equal parts. Based on the results of calculating the average magnitude of the vectors in the divided parts and using a solution of comparing these average magnitudes with each other, the robot can figure out obstacle position to avoid and guide the robot to a safe direction.The experiments are simulated on Matlab program to test the performance of the system. The simulated office environment with many obstacles randomly arranged along the corridors is used to test the ability to recognize obstacles to avoid. Simulation results related to different obstacle density scenarios are analyzed to demonstrate the stability of obstacle detection from the noise-free optical flow images.

Read more…

Soft robots have softer, more flexible, and more compliant characteristics than traditional rigid robots. These qualities encourage more secure relationships between people and machines. Nevertheless, traditional robots continue to rule the commercial sector. Due to the high cost of gripper production, soft robots are very far from being commercially feasible. This research focuses on fabricating a soft robotic gripper with the potential for mass production using injection molding technology. The material used for manufacture is Thermoplastic Elastomer (TPE). This study gives an injection molding optimization strategy by using Moldex 3D simulation to minimize production time for soft grippers. Furthermore, using an Ansys workbench, this study simulated soft gripper deflections with variable pressures by finite element analysis and then compared it with the actual experiment. The simulation results of TPE warpage volume shrinkage are 11.969% and 11.96% in the molding experiment. Thus, the shrinkage and warpage for the simulation and actual experiment are similar. According to the simulation outcome, the success of TPE hollow injection molding facilitates soft gripper creation. The maximum pressure used in the FEM simulation of the bending experiment was achieved at the pressure of 50 kPa with 152.02 mm of deformation and compared to the experimental data, 145,03 mm. This error is less than 5%. Finally, a better soft gripper design was achieved by Ansys simulation, and the soft gripper appears to be ready for mass-produced by TPE injection molding.

Read more…

Soft robots have softer, more flexible, and more compliant characteristics than traditional rigid robots. These qualities encourage more secure relationships between people and machines. Nevertheless, traditional robots continue to rule the commercial sector. Due to the high cost of gripper production, soft robots are very far from being commercially feasible. This research focuses on fabricating a soft robotic gripper with the potential for mass production using injection molding technology. The material used for manufacture is Thermoplastic Elastomer (TPE). This study gives an injection molding optimization strategy by using Moldex 3D simulation to minimize production time for soft grippers. Furthermore, using an Ansys workbench, this study simulated soft gripper deflections with variable pressures by finite element analysis and then compared it with the actual experiment. The simulation results of TPE warpage volume shrinkage are 11.969% and 11.96% in the molding experiment. Thus, the shrinkage and warpage for the simulation and actual experiment are similar. According to the simulation outcome, the success of TPE hollow injection molding facilitates soft gripper creation. The maximum pressure used in the FEM simulation of the bending experiment was achieved at the pressure of 50 kPa with 152.02 mm of deformation and compared to the experimental data, 145,03 mm. This error is less than 5%. Finally, a better soft gripper design was achieved by Ansys simulation, and the soft gripper appears to be ready for mass-produced by TPE injection molding.

Read more…

In recent years, the adoption of ICT education has increased in educational settings. Research and development of educational support robots have garnered considerable interest as a promising approach to inspire and engage students. Conventional robots provide learning support through button operations by the learners. However, the frequent need for button operation to request support may lead to a tedious impression on the learner and lower the efficiency of the learning process. Therefore, in this study, we developed a Perplexion Estimation Method that estimates the learner’s state of perplexity by analyzing their facial expressions and provides autonomous learning support. We verified the impact of a teacher-type robot (referred to as the proposed robot) that autonomously provides learning support by estimating the learners’ perplexity states in joint learning with university students. The results of a subject experiment showed that the impression of the proposed robot was not different from that of the conventional robot. However, the proposed robot demonstrated the ability to provide optimal support timing compared to the conventional robot. Based on these results, it is expected that the utilization of the perplexion estimation method with teacher-type robots can create a learning environment similar to human-to-human interaction.

Read more…

Over the past ten years, deep learning models have considerably advanced research in artificial intelligence, particularly in the segmentation of medical images. One of the key benefits of medical picture segmentation is that it allows for a more accurate analysis of anatomical data by separating only pertinent areas. Numerous studies revealed that these models could make accurate predictions and provide results that were on par with those of doctors. In this study, we investigate different methods of deep learning with medical image segmentation, like the V-net and U-net models. Improve the V-net model by adding attention in 2D with a decoder to get high performance through the training model. Using tumors of severe forms, size, and location, we downloaded the BRAST 2018 data set from Kaggle and manually segmented structural T1, T1ce, T2, and Flair MRI images. To enhance segmentation performance, we also investigated several benchmarking and preprocessing procedures. It’s significant to note that our model was applied on Colab-Google for 35 epochs with a batch size of 8. In conclusion, we offer a memory-effective and effective tumor segmentation approach to aid in the precise diagnosis of oncological brain diseases. We have tested residual connections, decoder attention, and deep supervision loss in a comprehensive ablation study. Also, we looked for the U-Net encoder and decoder depth, convolutional channel count, and post-processing approach.

Read more…

Early warning systems have had a significant impact on society by providing timely information to mitigate the effects of natural disasters. To enhance early warning capabilities, researchers are exploring the use of digital terrestrial television systems to broadcast alerts across large urban and rural areas. In this research project, the aim is to integrate the global early warning protocol CAP (Common Alerting Protocol) into the existing ISDB-T standard, alongside the standard’s Emergency Warning Broadcasting System (EWBS). This integration will enable the creation of a hybrid system, allowing various global emergency alert devices that utilize the CAP protocol to be activated through the Digital Television signal. To achieve this, a CAP to EWBS translator was developed as part of the design proposal prior to transmission. Additionally, transmitters compliant with both the full-seg and one-seg ISDB-T standards were designed to support the CAP protocol. These transmitters utilize SDR (Software-Defined Radio) cards of the Adalm Pluto type. The CAP protocol, encoded in XML format, was transmitted through the ISDB-T transport stream using the DSM-CC data transmission protocol. By incorporating the CAP protocol into the ISDB-T standard and utilizing the DSM-CC data transmission protocol, this research project aims to enhance the early warning capabilities of digital terrestrial television systems.

Read more…

The human resource development of robotics and automation in the smart factory is an important factor in “Thailand 4.0” roadmap, which is following the industry 4.0 model. To pursue this goal of Thailand 4.0 roadmap of labor development, the effective and intuitive training system must be easy to understand. This study proposes the implementation of augmented reality (AR) technology for training purposes due to its ability to visualize real-time invisible data, such as device status. This involves the development of the AR-based visual programming interface with an educational demonstration site (demo site). The AR-based training system is started with animation content explaining the smart factory concept, followed by hands-on learning using Microsoft HoloLens 2 and IoT hardware devices in demo site. The demo site using an MQTT protocol, simulates an automated packing line in the smart factory. The hardware status is published in real-time to the MQTT broker. This approach enables users to comprehend the interconnected relationship between data and hardware functionality and allowing them to create their own programs to control the IoT hardware in the demo site. With this training, 22 targeted users have successfully grasped the smart factory concept and its correlation with hardware functionality. The block-based visual programming employed in the system enables easy comprehension of robot commands. Moreover, the AR application provides a smooth display at 48-60 frames per second. The proposed system’s usability and value for specific tasks received high scores, ranging from 4 to 5 points, confirming its effectiveness for the targeted users. The proposed AR-based training system has proved that it has benefit for human resource development in the robotics and automation sector following the Thailand 4.0 roadmap. The proposed system empowered users to understand the smart factory concept and its practical implementation, leading to create new ideas for integrating AR and smart factory concepts into their manufacturing in the future.

Read more…

Unmanned aerial vehicles (UAVs) are becoming increasingly popular for both civil and military applications. Unmanned aerial vehicles can be categorized into two categories: rotary-wing and fixed-wing. Due to its capacity to fly long distances and carry substantial payloads, fixed-wing UAVs are gaining popularity and are currently utilized for various tasks. However, when confronted with disturbances such as weather or wind gusts, fixed-wing UAVs can rapidly lose stability, leading to a loss of lift and stalling. Consequently, it is vital to ensure the stability of fixed-wing UAVs. For the longitudinal stability management of a fixed wing unmanned aerial vehicle, the design and modeling of a feedback controller, including a PI controller, PID controller, and Fuzzy logic controller, are discussed in this article. MATLAB/SIMULINK©2021 will be used to design the control system and compare the response of each controller during the simulation. The controller’s response to various input formats will indicate its capacity to regulate the system’s behavior. Our results indicate that the fuzzy logic controller was superior to the PI and PID controllers at controlling the system’s response according to the desired or input behavior.

Read more…

Condition assessment of medium voltage assets is essential to ensure reliability and cost-effective operation of power distribution networks. This article presents a literature review of condition assessment of medium voltage assets related to a distribution system in a non-interconnected zone in Colombia, namely, power transformers, photovoltaic systems, switchgear, lines and cables, and instrument transformers. Advanced search rules are formulated to obtain bibliographic records of relevant academic literature from the database Scopus. The retrieved data are analyzed quantitatively to provide insights on the current state of research on the topic. Next, the most relevant academic papers for each medium voltage asset are selected and analyzed in a critical review along with more diverse literature including standards, technical reports, and white papers obtained through complementary searches. The results of the review show that several approaches have been formulated for condition assessment of medium voltage assets, ranging from traditional diagnostic methods to advanced artificial intelligence-based approaches. Moreover, research on some assets is already mature including power transformers, and photovoltaic systems, whereas other assets have been incipiently studied such as distribution and instrument transformers. Therefore, the need for deeper condition assessment research for these critical assets is highlighted. Research gaps are identified in the standardization and integration of condition assessment tools for distribution system operators.

Read more…

In a challenging and increasingly technological world, it is important to promote critical thinking, multidisciplinary problem solving, and collaboration through STEAM education; however, there are important economic, administrative, and especially pedagogical manage- ment limitations for its implementation at the secondary level. Therefore, this paper presents systematic recommendations for an effective and sustainable implementation of STEAM educa- tion in educational institutions through the Gradual Multidisciplinary Model (GMM), which allows the identification and specific adaptation of STEAM knowledge through the topic of logic gates related to the representations of disjunction and conjunction in Boolean algebra (university content) to its physical representation in Minecraft (high school content). The quasi-experimental method allows to evaluate the results through the application of a pre-test designed to measure logical-mathematical thinking and a post-test designed to measure the level of understanding of practical skills and the students’ perception of the learning experience. The results obtained by t-student show that there is a significantly high difference between the means and suggest that the educational intervention orchestrated by the GMM had a significant impact on the performance and skills of the participants, since different levels of understanding (gradualness) and perception of the concepts related to logic gates could be identified; while the qualitative assessment shows the group’s willingness and enthusiasm to work with a practical and meaningful activity using Minecraft, which allows them to specifically apply their skills in science, technology, engineering, art and mathematics.

Read more…

In a challenging and increasingly technological world, it is important to promote critical thinking, multidisciplinary problem solving, and collaboration through STEAM education; however, there are important economic, administrative, and especially pedagogical manage- ment limitations for its implementation at the secondary level. Therefore, this paper presents systematic recommendations for an effective and sustainable implementation of STEAM educa- tion in educational institutions through the Gradual Multidisciplinary Model (GMM), which allows the identification and specific adaptation of STEAM knowledge through the topic of logic gates related to the representations of disjunction and conjunction in Boolean algebra (university content) to its physical representation in Minecraft (high school content). The quasi-experimental method allows to evaluate the results through the application of a pre-test designed to measure logical-mathematical thinking and a post-test designed to measure the level of understanding of practical skills and the students’ perception of the learning experience. The results obtained by t-student show that there is a significantly high difference between the means and suggest that the educational intervention orchestrated by the GMM had a significant impact on the performance and skills of the participants, since different levels of understanding (gradualness) and perception of the concepts related to logic gates could be identified; while the qualitative assessment shows the group’s willingness and enthusiasm to work with a practical and meaningful activity using Minecraft, which allows them to specifically apply their skills in science, technology, engineering, art and mathematics.

Read more…

In this paper, we delve into the pressing necessity for proficient battery cell balancing, an imperative in the context of the escalating adoption of renewable energy and electric vehicles. While traditional methodologies, including the passive technique, offer a straightforward and cost-effective solution, they compromise on efficiency. The active technique, though superior in efficiency, is hindered by its innate restriction of transferring energy solely to proximate cells, thus prolonging the balancing process. To address these limitations, we introduce a novel near-field coupling method centered on a meticulously designed resonant coil with an emphasis on achieving a larger Q-factor, a pivotal factor for enhanced battery cell balancing. This augmented Q-factor not only propels our approach past the passive method in efficiency but also catalyzes rapid balancing by enabling wireless energy transfer to cells regardless of their relative positioning. Validating our theoretical insights, we developed physical coil prototypes and adopted a Series-Parallel circuit configuration, steered by the resonant coil’s Q-factor. Preliminary experiments with three batteries substantiate our claim, showcasing that our proposed technique achieves cell balancing with approximately double the speed of conventional strategies.

Read more…

Many designs for robot arms exist. Here we present an affordable revolute arm, capable of executing simple pick-and-place tasks. The arm employs a double parallelogram structure, which ensures its endpoint angle in the plane of the upper arm remains fixed without the need for additional actuation. Its limbs are fabricated from circular tubes made from bonded carbon fiber, to ensure low moving mass while maintaining high rigidity. All custom structural elements of the arm are produced via 3D printing. We employ worm-drive DC motor actuation to ensure that stationary configurations are maintained without the necessity of continuous motor power. Our discussion encompasses an analysis of the arm’s kinematics. A simulation of the arm’s operation was carried out in MATLAB, revealing key operational metrics. In conclusion, we achieved extrinsic endpoint position tracking by implementing its inverse kinematics and PID control using a microcontroller. We also demonstrate the arm’s functionality through simple movement tracking and object manipulation tasks.

Read more…

A robot development has spread widely for various purposes. It is difficult to create a control program for an autonomous mobile robot manually. Therefore, an automatic design of the control program for an autonomous mobile robot is proposed in this research. The autonomous mobile robot is created with LEGO MINDSTORMS EV3, and the control program for the au- tonomous mobile robot is designed using Grammatical Evolution (GE). Grammatical Evolution (GE), which is one of the evolutionary computations, is designed to generate a program or a program fragment satisfying the design objective. PyBullet is used with GE to simulate the behavior of the robot. A robot traveling along a trajectory was considered as an example. GE can generate the control program of the robot behavior of a robot vehicle traveling along a trajectory. The computer simulation reveals the robot can travel along a designated line. Since there is a reality gap between the simulator and the real environment, the parameters of the vehicle robot such as produced power and sensor sensitivity are calibrated to reduce the gap. Comparison of the computer simulation and the experimental result shows that the reproducibility of the vehicle trajectory in the real environment is high.

Read more…

An ensemble method is a Machine Learning (ML) algorithm that aggregates the predictions of multiple estimators or models. The purpose of an ensemble module is to provide better predictive performance than any single contributing model. This can be achieved by producing a predictive model with reduced variance using bagging, and bias using boosting.
The Tree-Based Ensemble Models with Decision Tree (DT) as base model is the most frequently used. On the other hand, there are some individual Machine Learning algorithms that can provide more competitive predictive power to the ensemble models. It is a problem, and this issue is addressed here. This work has two parts. The first one presents a Projective Decision Tree (PA) based on purity measure. Next node criterion (CNN) is also used for node decision making. In the second part, two sets of algorithms for predictive performance are presented. The Tree-Based Ensemble model includes bagging and boosting for homogeneous learners and a set of known individual algorithms. Comparison of two sets is performed for accuracy. Furthermore, the changes of bagging and boosting ensemble performance under various hyperparameters are also investigated. The datasets used are the sonar and the Breast Cancer Wisconsin (BCWD) from UCI site. Promising results of the proposed models are accomplished.

Read more…

Daily life is characterized by a great explosion of abundance of information available on the internet and social media. Smart technology has radically changed our lives, giving a leading role to social media for communication, advertising, information and exchange of opinions. Managing this huge amount of data by humans is an almost impossible task. Adequacy of summarizing texts is therefore urgently needed, in order to offer people knowledge and information avoiding time-consuming procedures. Various text summarization techniques are already widely used. Artificial intelligence techniques for automated text summarization are a major undertaking. Due to the recent development of neural networks and deep learning models like Transformers, we can create more efficient summaries. This paper reviews text summarisation approaches on social media and introduces our approach towards a summarization system using transformers.

Read more…

In the context of adopting cloud Infrastructure-as-a-Service (IaaS), prospective consumers need to consider a wide array of both business and technical factors associated with the service. The development of an intelligent tool to aid in the assessment of IaaS offerings is highly desirable. However, the creation of such a tool requires a robust foundation of domain knowledge. Thus, the focus of this paper is to introduce an ontology specifically designed to characterize IaaSs from the consumer’s perspective, enabling informed decision-making. The ontology additionally serves two purposes of other relevant parties besides the consumers. Firstly, it empowers IaaS providers to better tailor their services to align with consumer expectations, thereby enhancing their competitiveness. Additionally, IaaS partners can play a pivotal role in supporting both consumers and providers by understanding the protocol outlined in the ontology that governs interactions between the two parties. By applying principles of ontological engineering, this study meticulously examined the various topics related to IaaS as delineated in existing cloud taxonomies. These topics were subsequently transformed into a standardized representation and seamlessly integrated through a binary integration approach. This process resulted in the creation of a comprehensive and cohesive ontology that maintains semantic consistency. Leveraging Protégé, this study successfully constructed the resultant ontology, comprising a total of 340 distinct classes. The study evaluated the syntactic, semantic, and practical aspects of the ontology against a worldwide prominent IaaS. The results showed that the proposed ontology was syntactically and semantically consistent. Furthermore, the ontology successfully enabled not only the assessment of a real leading IaaS but also queries to support automation tool development.

Read more…

Mental stress assessment remains riddled with biases caused by subjective reports and individual differences across societal backgrounds. To objectively determine the presence or absence of mental stress, there is a need to move away from the traditional subjective methods of self-report questionnaires and interviews. Previously, it has been evidence that EEG Oscillations can discriminate mental states, for instance, stressed and non-stressed. However, it is still not clear in which range of EEG oscillations the neural activities are associated with the mental states. This paper presents a wavelet-based EEG feature extraction method for the classification of mental stress using machine learning classifiers. An EEG dataset of 22 participants was used to test the performance of the proposed wavelet-based feature extraction method. The dataset includes both stress and control conditions, and the stress condition has multiple levels of stress, starting from low, mild, and high stress. The Daubechies mother wavelet of the fourth order was used to separate the EEG oscillations into 7 levels for the extraction of the absolute powers. Whereas Fast Fourier Transform were implemented to obtain the average power of the oscillations. The features were then used in support vector machine, decision tree, linear discriminant analysis and artificial neural network classifiers. A comparison between the classifiers using average power, absolute power, and a combination of both is provided. The EEG alpha, theta, and beta frequency bands showed promising results for the classification of mental stress vs. control conditions by achieving an average accuracy of 95% using the decision tree. The results of the proposed method suggest the potential use of wavelet analysis for mental stress detection despite FFT performing better. The proposed method has the potential to be used in Computer-Aided Diagnosis (CAD) systems for mental stress assessment in the future alongside the discovery of significant wave bands in relation to mental stress detection.

Read more…

Liver cancer is a major contributor to cancer-related mortality both in the United States and worldwide. A range of liver diseases, such as chronic liver disease, liver cirrhosis, hepatitis, and liver cancer, play a role in this statistic. Hepatitis, in particular, is the main culprit behind liver cancer. As a consequence, it is decisive to investigate the correlation between hepatitis and symptoms using statistic inspection. In this study, we inspect 155 patient data possessed by CARNEGIE-MELLON UNIVERSITY in 1988 to prognosticate whether an individual died from liver disease using supervised machine learning models for category and connection rules based on 20 different symptom attributes. We compare J48 (Gain Ratio) and CART (Classification and Regression Tree), two decision tree classification algorithms elaborate from ID3 (Iterative Dichotomiser 3), with the Gini index in a Java environment. The data is preprocessed through normalization. Our study demonstrates that J48 outperforms CART, with an average accuracy rate of nearly 87% for the complete specimen, cross-validation, and 66% training data. However, CART has the supreme accurate rate in all samples, with an accuracy rate of 90.3232%. Furthermore, our research indicates that removing the conjunction attribute of the Apriori algorithm does not impact the results. This research showcases the potential for physician and researchers to apply brief machine learning device to attain accurate outcomes and develop treatments based on symptoms.

Read more…

In this paper, we have developed a gas injection-type prefabricated lifting device based on location tracking to efficiently lift the human body in the event of an accident that occurs underwater on the sea or land. The efficiency of the lifting system is very important to ensure the golden time of the rescue and the safety of divers in the event of casualties underwater. Divers performing underwater safety rescue operations must endure up to 30 minutes with two air vents, and always consider the safety accident environment due to difficulty in securing visibility or high flow rates due to underwater turbidity. Particularly, there are many cases where life is threatened by hypothermia in the water. Therefore, both divers and the deceased need location tracking connected to the lifting device, and a fast and efficient lifting system was studied in underwater activities. The monitoring device uses a communication speed of 115.2 kbps from the sensor to the monitoring, and a communication speed of 2.4 kbps from the controller to the receiving unit. The gas injection-type prefabricated lifting device with a high elastic structure is lightweight and portable, and which consists of a baggy bag with minimal components to increase usage and work efficiency based on the instinctive behavior of divers. Accordingly, the entrance element design combining a bow and hinge that maintains a moment of force with TPU-based materials, a balanced design using weight balancing technology of a network structure, an SMB linkage design that induces water surface rise through gas injection, and an underwater experiment.

Read more…

The advent of laparoscopic surgery has increased the need to incorporate simulator-based training into traditional training programs to improve resident training and feedback. However, current training methods rely on expert surgeons to evaluate the dexterity of trainees, a time-consuming and subjective process. Through this research, we aim to extend the use of object detection in laparoscopic training by detecting and tracking surgical tools and objects. In this project, we trained YOLOv7 object detection neural networks on Fundamentals of Laparoscopic Surgery pattern-cutting exercise videos using a trainable bag of freebies. Experiments show that YOLOv7 has a mAP score of 95.2, 95.3 precision, 94.1 Recall, and 78 accuracy for bounding boxes on a limited-size training dataset. This research clearly demonstrates the potential of using YOLOv7 as a single-stage real-time object detector in automated tool motion analysis for the assessment of the resident’s performance during training.

Read more…

Advances in Internet of Things (IoT) and wireless technologies are revolutionizing various sectors, including environment, education, healthcare, industry, etc. In the same dynamic, as the world population constantly evolves, solutions based on such technologies need to be proposed to improve the agricultural sector. Senegalese agriculture, primarily rain-fed and based on both cash crops and subsistence food crops, faces challenges from climate change, environmental impacts, and traditional methods. Lack of information on good practices, new efficient and low-cost farming methods, and access to seeds, along with the COVID-19 pandemic, further complicates the situation. This research work proposes a smart farming management system that uses IoT technology, named MbaïMi, whose main objective is to improve Senegalese agriculture through new digital technologies for sustainable agriculture, environmental benefits and food self-sufficiency. It provides a real- time decision support tool that offers farmers best seeds to choose at the beginning of work and helps optimize water resources. Additionally, the system includes features like sending SMS/advice, graphical visualizing of temperature and humidity changes for greenhouse cultivation control. The development of a web and mobile application, deployment of a prototype in southern Senegal, and validation case studies demonstrate the system’s potential to improve farmers’ yields and daily hard work to meet the challenges of sustainable agriculture, environmental benefits and food self-sufficiency.

Read more…

In this study, two healthy women in their 40s wore Oura Rings on both fingers and verified left and right differences in sleep index. Bed time, sleep time, rem sleep, sleep score, and heart rate had small left and right differences in both subjects, and a high correlation of 0.8 or more was observed(subjectA:P<0.001,subjectB:P<0.01). Subject B showed a significant difference in the mean values of sleep latency, awakening time and sleep efficiency, and one finger was overestimated(P<0.05). Although the detection accuracy of sleep stages differs depending on the stage, it was shown that the influence of left and right differences is particularly small for bed time, sleep time, rem sleep, sleep score and heart rate.

Read more…

A resume is a prevalent and generally employed method for individuals to showcase their proficiency and qualifications. It is typically composed using diverse customized, personalized methods in multiple inconsistent formats (such as pdf, txt, doc, etc.). Screening candidates based on the alignment of their resume with a set of job requirements is typically a labori- ous, challenging, time-intensive, and resource-intensive endeavor. This work is crucial for extracting pertinent information and valuable attributes that indicate acceptable applicants. This study aims to improve the candidate career-matching process for human resource (HR) departments by implementing automation and increasing efficiency. Machine learning (ML) and natural language processing (NLP) techniques are applied to infer and analyze comparative semantic resume attributes. Using semantic data comparisons, the ranking support vector ma- chine (RankSVM) algorithm is subsequently applied to rank these resumes based on attributes. RankSVM detects tiny differences among candidates and assigns unique scores, resulting in an improved ranking of candidates based on their suitability for job requirements and from the best to worst match for the vacancy. The experiment and performance comparison results show that the proposed comparative ranking, which relies on semantic descriptions, outperforms the standard ranking based on regular scores in distinguishing candidates and distributing resumes across the ranks with an accuracy of up to 92%. Eventually, we obtained a list of the top ten candidates out of 228 technical specialists’ resumes.

Read more…

This work enhances the LoRaWAN server framework, focusing on an innovative approach for robust security and dynamic data visualization in network management. Migrating from RVC4 to AES encryption, it fortifies the network’s defense against cyber threats, a crucial advancement in IoT security. Furthermore, the integration with Grafana’s mapping plugin capitalizes on geolocation data, a strategic element for network oversight and IoT data analysis. In-depth configuration and application of this plugin are explored, revealing substantial benefits for network administrators and end-users. Expanded discussions, backed by new experimental data, illustrate the real-world efficacy of these technological improvements. The research substantially enriches the understanding of LoRaWAN’s technological evolution, addressing vital aspects of IoT security and geolocation integration. The outcomes are expected to resonate significantly within academic circles and practical domains, particularly in reinforcing IoT and network security. This work marks a significant stride in the progression of LoRaWAN technologies, with implications that extend well into the broader landscape of network management and security in the digital age.

Read more…

The aim of this study is to develop an advanced framework that not only recognize the dominant facial emotion, but also contains modules for gesture recognition and text-to-speech recognition. Each module is meticulously designed and integrated into unified system. The implemented models have been revised, with the results presented through graphical representations, providing prevalent emotions in facial expressions, body language, and dominant speech/voice analysis. Current research, to identify the dominant facial emotion, involves two distinct approaches that autonomously determine the primary emotional label among seven fundamental emotions found in the input data: anger, disgust, happiness, fear, neutral expressions, sadness and surprise. The dataset utilized comprises over 292680 images sourced from the benchmark datasets FER-2013 and CK, enriched by images sourced from the Google search engine, along with 80 videos obtained during dedicated sessions, used for training and testing purposes. The Residual Masking Network (resmasknet) and CNN architectures are used as pre-trained models in this analysis. Resmasknet and CNN were chosen considering their superior performance compared to other algorithms found in the literature. The CNN architecture comprises 11 blocks, with each block containing a linear operator followed by ReLU or max-pooling layer. Starting with a convolutional layer that uses 32 filters and an 11x11x3 input, followed by a 3×3 max-pooling layer with a step of 2, the next layer includes a convolutional layer that uses 16 filters of size 9x9x16. The Residual Masking Network, contains four residual masking blocks operating on different feature sizes, each consisting a residual layer and masking block. The network initiates with a 3×3 convolution layer, followed by 2×2 max-pooling, effectively downsizing the image to 56×56. Successive transformations within four residual masking blocks generate different maps of sizes 56×56, 28×28, 14×14, and 7×7, culminating in an average pooling layer and a fully connected SoftMax output layer. The significance of this project lies in its focus on a comprehensive analysis of emotions and movements characteristic of Asian and European facial expressions. Showing promising accuracy rates, the proposed solutions achieve 75.2% accuracy for Asian and 86.6% for European individuals. This performance demonstrates the potential of this multidisciplinary framework in understanding and interpreting different facial expressions in different cultural settings.

Read more…

The cybercriminal utilized the skills and freely available tools to breach the networks of internet-connected devices by exploiting confidentiality, integrity, and availability. Network anomaly detection is crucial for ensuring the security of information resources. Detecting abnormal network behavior poses challenges because of the extensive data, imbalanced attack class nature, and the abundance of features in the dataset. Conventional machine learning approaches need more efficiency in addressing these issues. Deep learning has demonstrated greater effectiveness in identifying network anomalies. Specifically, a recurrent neural network model is created to recognize the serial data patterns for prediction. We optimized the hybrid model, the convolutional neural network combined with Bidirectional Long-Short Term Memory (BLSTM), to examine optimizers (Adam, Nadam, Adamax, RMSprop, SGD, Adagrad, Ftrl), number of epochs, size of the batch, learning rate, and the Neural Network (NN) architecture. Examining these hyperparameters yielded the highest accuracy in anomaly detection, reaching 98.27% for the binary class NSL-KDD and 99.87% for the binary class UNSW-NB15. Furthermore, recognizing the inherent class imbalance in network-based anomaly detection datasets, we explore the sampling techniques to address this issue and improve the model’s overall performance. The data imbalance problem for the multiclass network anomaly detection dataset is addressed by using the sampling technique during the data preprocessing, where the random over-sampling methods combined with the CNN-based BLSTM model outperformed by producing the highest performance metrics, i.e., detection accuracy for multiclass NSL-KDD and multiclass UNSW-NB15 of 99.83% and 99.99% respectively. Evaluation of performance, considering accuracy and F1-score, indicated that the proposed CNN BLSTM hybrid network-based anomaly detection outperformed other existing methods for network traffic anomaly detection. Hence, this research contributes valuable insights into selecting hyperparameters of deep learning techniques for anomaly detection in imbalanced network datasets, providing practical guidance on choosing appropriate hyperparameters and sampling strategies to enhance model robustness in real-world scenarios.

Read more…

Read more…