International Journal of Engineering and Manufacturing @ijem
Статьи журнала - International Journal of Engineering and Manufacturing
Все статьи: 623
Generation of Images from Text Using AI
Статья научная
Reading the words can be confusing, and it may be hard to picture what is happening. There are some circumstances where words can be misunderstood. It's much simpler to recognize text if it's displayed as an image. The use of visuals is proven to increase viewership and retention. Synthesizing realistic images automatically is a challenging undertaking, and even the most advanced artificial intelligence and machine learning algorithm has trouble meeting this standard. GANs (Generative Adversarial Networks) are just one example of a powerful neural network architecture that has shown promising results in recent years. Existing text-to-image methods can generate examples that generally reflect the meaning of the provided descriptions, but they often lack the necessary details and colorful object elements. The primary objective of our research was to explore diverse architectural methodologies with the intention of facilitating the generation of visual representations from textual descriptions. By delving into this investigation, we aimed to discover and examine various approaches that could effectively support the creation of visuals that accurately depict the content and context provided within written narratives. Our aim was to unlock new possibilities in the realm of visual storytelling by establishing a strong connection between language and imagery through innovative architectural techniques.
Бесплатно
Government Expenditures, Transfer Payments and Economic Growth
Статья научная
Incorporating a two-level government structure into an endogenous growth model, we distinguished between productive and non-productive government expenditures. With transfer payments considered, we showed that (1) there was an “Inverted U-shaped” relationship between the tax rate and the long-run economic growth, so was the relationship between the degree of fiscal decentralization and the long-run economic growth; (2) optimal ratios between productive and non-productive expenditures of two levels of governments, between transfer payments and other parts of expenditures of the state-level governments are needed to maximize the long-run economic growth.
Бесплатно
Статья научная
The basic difference between a sustainable aesthetically positive urban environment and an aesthetically negative one is in the way its component installations are rendered. The aesthetic positivity (AP) or aesthetic negativity (AN) of the whole is dependent on the aesthetics of the little parts that constitute it. Although may be functional, many electrical installations in Nigeria still lack considerable aesthetics mostly due to lack or laxity in the knowledge or practical application of basic design theories and principles. This study therefore examined how the application of design principles and theories used in graphic design can apply in electrical and design installations as a way of fostering a more aesthetic, yet functional and sustainable environments in developing West African countries using aesthetics as a key driver. Adopting a descriptive approach supported with direct observation, with a sample size of 320, respondents were purposively sampled in selected cities in Nigeria. The study showed a significant relationship between the application of graphic design theories and improved environmental aesthetics through the rendering of attractive-functional electrical/design installations. It also revealed that improved aesthetics of electrical/design installations limits negative interference which improves sustainability/safety in the built environment, hence serving as an abatement tool or technology for the alleviation of AN. This study therefore established the significance of the application of design theories and principles in achieving a more aesthetic, functional and sustainable environment, from the professionals’/ practitioners’ perspective.
Бесплатно
Green Functions for Sub-Laplacian on Half Spaces of the Heisenberg Group
Статья научная
Green functions for sub-Laplacian on the domains in the Heisenberg group are derived, which can be used to solve partial differential equations subject to specific initial conditions or boundary conditions. Then the integral formulas for sub-Laplace equation on characteristic and non-characteristic half spaces are given, respectively.
Бесплатно
Guiding Aid for Visually Impaired
Статья научная
Visual impairment is where the person either can’t see or his vision has weakened to large extent. There is no alternative technique for visually impairment, but to some extent it can be trim down with devices, smart sticks and sensors. Although many techniques are there for helping out through electronic travelling aid, cost effective and minimum hardware solution was the expectation by impaired. The device which can identify and classify the object ahead of impaired person is needed so that person can be prevented from the accident. In this paper, a unified model of YOLO (You Only Look Once) is used for detection of object ahead of camera. The proposed model is based on phenomena of detecting small object and good detection speed of yolov3 makes system more robust. Once detected, labeled objects name is converted from text to speech, so that blind person can be alerted from colliding with obstacles. This paper is one step in the direction to help them by exactly classifying, detecting and localizing target object along with providing voice based guideline. The proposed model has proved accuracy in many real time scenes.
Бесплатно
HKCHB: Meta-heuristic Algorithm for Task Scheduling and Load Balancing in Cloud-fog Computing
Статья научная
Cloud-fog computing has emerged as the contemporary approach for processing and analyzing Internet of Things applications due to its ability to offer remote resources. Cloud fog computing technology provides shared resources, information, and software packages, supporting distributed parallel systems in an open environment. It constructs and manages virtual machines to enhance efficiency and attractiveness. We have consistently strived to tackle challenges affecting the efficiency of cloud fog computing, including ineffective resource utilization and response times. The improvement of these challenges can be achieved through effective task scheduling and load balancing between Virtual Machines, this problem considered as NP-hard problem. This paper proposes a Hybrid K-means Clustering Honey Bee algorithm (HKCHB) to cluster Virtual Machines into two or more clusters. Subsequently, the hybrid Honey Bee algorithm is employed for task scheduling, enhancing load balance performance. The proposed algorithm is compared with other task scheduling and load balancing algorithms, including Round Robin, Ant Colony, Honey Bee, and Particle Swarm Optimization Algorithm, utilizing the CloudSim Simulator. The results demonstrate the superiority of the proposed algorithm, yielding the lowest response time. Specifically, the response time is reduced by 22.1%, and processing time is reduced by 47.9%, while throughput is increased by 95.4%. These improvements are observed under the assumption of multiple tasks in a heterogeneous environment, utilizing one or two Data Centers with Virtual Machines. This contribution gives the impression that network systems based on the Internet of Things and cloud fog computing will be improved in the future to operate within the framework of real-time systems with high efficiency.
Бесплатно
Hand Arm Vibration Alleviation of Motorcycle Handlebar using Particle Damper
Статья научная
Vibration induced in the vehicle affects the performance of the driver or rider because of reduction in comfort & safety level. In case of motorcycle, poor suspensions system & uneven road condition make driving difficult, these also introduces vibration. The vibrations are directly transferred to the body through the seat & handlebar. It has been seen that handlebar vibrations are more serious & creates physical problem to the rider. Particle damping technology is a derivative of impact damping with several advantages. Particle damping is the use of particles moving freely in a cavity to produce a damping effect. In this paper a passive damper using particle damping technique is designed and developed to reduce the hand arm vibrations (HAV). The experiments are planned and conducted using DOE. Optimum configuration of particle damper has been derived through this research work. Experimental tests shows by employing a particle damper the vibration amplitude is minimized significantly.
Бесплатно
Статья научная
The World is moving toward Smart traffic management and monitoring technologies. Vehicle detection and classification are the two important features of intelligent transportation system. Several algorithms for detection of vehicles such as Sobel, Prewitt, and Robert etc. but due to their less accuracy and sensitivity to noise they could not detect vehicles clearly. In this paper, a simple and rapid prototyping approach for vehicle detection and classification using MATLAB Xilinx system generator and Zedboard is presented. The Simulink model of vehicle detection and classification is designed using a complex canny edge detection algorithm for vehicle detection. The canny edge detection algorithm offers 91% accuracy as compared to its counterpart Sobel and Perwitt algorithms that offer 79.4% and 76.1% accuracy. The feature vector approach is used for vehicle classification. The proposed model is simulated and validated in MATLAB. The Canny edge detection and feature vector algorithms for vehicle detection and classification are synthesized through the Xilinx system generator in Zedboard. The proposed design is validated with the existing works. The implementation results reveal that the proposed system for vehicle detection and classification takes only 8 ns of execution time with a 128MHz clock, which is the lowest and optimum calculation period for the smart city.
Бесплатно
Hash Function Construction Based on RBFNN and Chaotic Mapping
Статья научная
One-way Hash function is not only widely used in the aspects of the digital signature, identity authentication and integrity checking, etc. but also the research hotspot in the field of contemporary cryptography. In this paper, it firstly utilized neural network and practiced the chaotic sequences produced by one-dimensional nonlinear mapping. And then, it constructed Hash function with cipherkey by means of altering sequences. One of the advantages of this algorithm is that neural network hides the chaotic mapping relations and make it difficult to obtain mapping directly. Simulation experiment showed that the algorithm have good unidirectionality and weak collision, and stronger confidentiality than the tradition-based Hash function, as well as easy to achieve.
Бесплатно
Healthcare System Technology using Smart Phones and Web Apps (Case Study Iraqi Environment)
Статья научная
In the Past a Few Years, smart devices like smart phones and tablets have radically change in many aspect, started from Entertainment to Shopping services to transfer Money and Banking, the next is Health Services. With the development in information technology And the big development in cloud computing Here smart phones have entered heavily in all aspects of health care. Now with the revolution of the smart devices (smart phones or tablets) and it's applications, there is many applications and tools are available started from attachments that allow to diagnose an infections and Now remotely and continuously monitor each heartbeat , blood pressure readings, the rate and depth of breathing, body temperature, oxygen concentration in the blood, glucose, brain waves, activity, mood, so the end result will be can reduce using of doctor ,also reduce the cost , and give us speed up and give power to patients , so make it possible for Patient to use portable devices (smart phones or tablet) to access their medical information, and achieve the goal to put information technology to work in health care and make the integration of health information technology into primary care .So the using information technology give us the good solution that won’t replace physicians. Health Information technology give the providers of health care to give better manage patient care. By making the health information are available electronically anytime and anywhere is needed, Health Information technology can help us to improve the quality Of Health , so can decrease the cost. Now, after all these advantages should shed light on the side of personal privacy And Hacking side that must have been tested all application and tools, All of these tools must be accurate and needs to be tested. So that must provide the highest level of security and privacy for the patients. After that. The application does not arrive the goal with 100% percent, according to the limitation that mentioned later.
Бесплатно
Heterostructure Silicon and Germanium Alloy Based Thin Film Solar Cell Efficiency Analysis
Статья научная
Thin film solar cell along with enhanced absorption property will be the best, so combination of SiGe alloy is considered. The paper presented here consists of a numerical model of Si/Si1−xGex heterojunction solar cell. The addition of Ge content to Si layer will affect the property of material. The research has investigated characteristics such as short circuit current density (Jsc), generation rate G , absorption coefficient (α), and open circuit voltage (Voc), power, fill factor (FF) with optimal Ge concentration. The speculative determination of appropriate germanium mole fraction is done to get the maximized thin-film solar cell efficiency.
Бесплатно
High Accuracy Swin Transformers for Image-based Wafer Map Defect Detection
Статья научная
A wafer map depicts the location of each die on the wafer and indicates whether it is a Product, Secondary Silicon, or Reject. Detecting defects in Wafer Maps is crucial in order to ensure the integrity of the chips processed in the wafer, as any defect can cause anomalies thus decreasing the overall yield. With the current advances in anomaly detection using various Computer Vision Techniques, Transformer Architecture based Vision models are a prime candidate for identifying wafer defects. In this paper, the performance of Four such Transformer based models – BEiT (BERT Pre-Training of Image Transformers), FNet (Fourier Network), ViT (Vision Transformer) and Swin Transformer (Shifted Window based Transformer) in wafer map defect classification are discussed. Each of these models were individually trained, tested and evaluated with the “MixedWM38” dataset obtained from the online platform, Kaggle. During evaluation, it has been observed that the overall accuracy of the Swin Transformer Network algorithm is the highest, at 97.47%, followed closely by Vision Transformer at 96.77%. The average Recall of Swin Transformer is also 97.54%, which indicates an extremely low encounter of false negatives (24600 ppm) in contrast to true positives, making it less likely to expose defective products in the market.
Бесплатно
Home Occupancy Classification Using Machine Learning Techniques along with Feature Selection
Статья научная
Monitoring systems for electrical appliances have gained massive popularity nowadays. These frameworks can provide consumers with helpful information for energy consumption. Non-intrusive load monitoring (NILM) is the most common method for monitoring a household’s energy profile. This research presents an optimized approach for identifying load needs and improving the identification of NILM occupancy surveillance. Our study suggested implementing a dimensionality reduction algorithm, popularly known as genetic algorithm (GA) along with XGBoost, for optimized occupancy monitoring. This exclusive model can masterly anticipate the usage of appliances with a significantly reduced number of voltage-current characteristics. The proposed NILM approach pre-processed the collected data and validated the anticipation performance by comparing the outcomes with the raw dataset’s performance metrics. While reducing dimensionality from 480 to 238 features, our GA-based NILM approach accomplished the same performance score in terms of accuracy (73%), recall (81%), ROC-AUC Score (0.81), and PR-AUC Score (0.81) like the original dataset. This study demonstrates that introducing GA in NILM techniques can contribute remarkably to reduce computational complexity without compromising performance.
Бесплатно
Hotspot sequence patterns with an improvement in spatial feature
Статья научная
Forest fires in Sumatra and Kalimantan resulted in degradation of peatlands significantly. The strong indicator of forest and land fires including in peatland can be identified using hotspots which occurred consecutively in 2 to 5 days. The previous studies have been conducted in mining sequence patterns on hotspot datasets in Sumatra and Kalimantan. However, those studies applied the sequential pattern algorithms on the datasets containing temporal and rough spatial features. This study aims to generate sequence pattern of hotspot datasets using the SPADE algorithm with the improvement of the spatial feature. The study results in 892 1-frequent sequences and 28 2-frequent sequence patterns at the minimum support of 0.02%. A total of 484 hotspots were found from the 28 2-frequents sequence patterns, most of which were occurred in September to November 2014 and 2015. Central Kalimantan, Riau, and South Sumatra are the area where hotspots mostly occurred in 2014 and 2015. The visualization module for hotspot sequences was successfully developed in two iterations using the JavaScript.
Бесплатно
Human Identification Using Foot Features
Статья научная
The goal of this paper is to investigate a new technique for human identification using foot features. This work can be mainly decomposed into image preprocessing, feature extraction and pattern recognition by Artificial Neural Network (ANN). Foot images are rarely of perfect quality. To obtain good minutiae extraction in foot with varying quality, we conducted preprocessing in form of image enhancement and binarization .To extract features from human foot based on shape geometry of foot boundaries by extracting 16 geometric features from a human foot image. The foot center has been determined, and then the distances between the center point and outer points are measured with different angles .The angles are from 30˚ to 360˚ by increment with 30˚ gradual. The 13th feature that can be extracted is the length of a foot which is defined as the distance between the top point of the foot and the bottom point. The 14th, 15th and 16th are three major features the width of the foot. The first width is passing through center point, therefore, the second widths of foot is measured from the upper part above the center point and third width from the region the center point under the center point at the bottom of the foot. Euclidean distance is used in the proposed system. Artificial Neural network used for recognition. MATLAB version 8.1(R2013a) and windows 7 with 32 bit is used to build the application and performed on pc of core i3 processor, and our test system on 40 persons, results were satisfactory up to more 92.5%.
Бесплатно
Hybrid Deep Learning-Based Automated Genre Classification of Assamese Regional Songs
Статья научная
This work aims to preserve and promote the rich musical heritage of Assam by developing an automated classification system for Assamese regional songs using a hybrid deep learning approach. This method not only modernizes the preservation of traditional music but also enhances its accessibility to a global audience for integrating AI with cultural conservation. Five genres of Assamese songs—Bihu, Kamrupiya Lokageet, Goalporiya Lokageet, Borgeet, and Naam—are considered in this study. By leveraging Convolutional Neural Networks (CNNs) and advanced audio feature extraction techniques such as Mel-Frequency Cepstral Coefficients (MFCCs) and spectrograms, a hybrid model combining VGG16 and ResNet50 is developed. This fusion utilizes the strengths of both architectures, enhancing the model’s performance and accuracy. Following the process, it is observed that two distinctly different genres, Bihu and Borgeet, are accurately categorized by the proposed model, while the remaining three show slight labeling inconsistencies.
Бесплатно
Hybrid Feature Fusion and Bayesian-Optimized Ensemble Learning for Robust Citrus Disease Detection
Статья научная
Detecting citrus diseases at an early stage is very important for ensuring fruit quality and minimizing production losses, as well as for raising awareness about sustainable agriculture. As a solution to this problem, the authors of this paper propose a hybrid feature-based citrus disease classification system that integrates deep learning representations, handcrafted descriptors, feature selection, and ensemble learning, all of which are tuned via Bayesian optimization. We perform tests on two real-world citrus disease datasets that differ greatly in nature: a four-class lemon dataset and a two-class orange dataset. Both datasets were collected under quite different environmental conditions, so they show diverse disease symptoms and varying background complexity. Deep semantic features were obtained by running a pretrained ResNet50 network. In addition to those, other complementary handcrafted features, such as color, texture, spatial, and statistical features, were extracted from the segmented infected areas. Together, the hybrid feature vector of 2082 dimensions was subjected to an optimization method known as Neighborhood Component Analysis (NCA). This technique selects 300 features that are most effective for discrimination while at the same time ensuring the preservation of class separability and the minimization of redundancy. For the classification task, two classifiers, namely Random Forest (RF) and Bayesian-Optimized Random Forest (BORF), were employed. The latter is based on Bayesian optimization to locate the model hyperparameters. To measure the model's performance in an unbiased manner, five-fold cross-validation was performed. Based on the experimental results, BORF can improve classification accuracy on the lemon dataset from 90.42% to 93.75% and on the orange dataset from 95.42% to 95.92% compared to the baseline RF classifier. Cross-validation mean accuracies of the proposed system were 93.75 ± 0.82% and 95.92 ± 0.47% for the lemon and orange datasets, respectively. Receiver Operating Characteristic (ROC) analysis provided class-specific area under the curve (AUC) values of 0.975 and 0.972 for the orange dataset, with a macro-averaged AUC of about 0.94 for the multi-class lemon dataset. The combination of hybrid feature fusion, NCA-based feature optimization, and Bayesian-optimized ensemble classification leads to enhanced discriminative power, greater robustness, and better generalization performance for the system in citrus disease identification in a real-world agricultural setting, as demonstrated by the results.
Бесплатно
Hybrid Quantum-Classical Framework for Computational Mental Energy from Multichannel EEG Streams
Статья научная
This paper presents a hybrid quantum-classical framework for real-time estimation of cognitive engagement from multichannel electroencephalography (EEG) using a new operational indicator called Computational Mental Energy (CME). The proposed approach integrates signal preprocessing (windowing, filtering, spectral feature extraction), spectral feature extraction, a 4-qubit variational quantum classifier (VQC) for flow-state probability estimation, and a metaheuristic optimization loop for balancing predictive quality and quantum resource cost. CME is defined as a window-level function of aggregated spectral energy, task complexity, and estimated flow probability, measured in a dedicated signal-energy unit called Vernik (Vn), with session-level aggregation rules. The system supports quantum-only, classical-only, and hybrid inference modes and is designed for streaming deployment with wearable EEG devices and server-side inference services. A single-subject pilot study involving eight cognitive activities and EEG recordings from a Muse Athena headband demonstrates that the hybrid mode (μ = 0.6) achieves 0.914 AUROC for flow-state detection, compared to 0.548 for the standalone quantum model, while reducing prediction variance by 40.9%. Validation on the IBM Marrakesh 156-qubit Heron r2 quantum processor shows strong agreement between simulator and hardware results (r = 0.869, MAE = 0.045), confirming the practical feasibility of execution on current quantum hardware. Across activities, CME rates differed significantly, with approximately a nine fold gap between coding and resting states, illustrating the framework’s ability to capture activity-dependent cognitive demand. The proposed architecture provides a reproducible pipeline for EEG-based cognitive-state analytics, resource-aware quantum inference, and future adaptive human-computer interaction systems.
Бесплатно
Hybrid Solar Power Plant in Saint Martin's Island can Enlarge Tourist Attraction in Bangladesh
Статья научная
Saint Martin's Island is the best tourist spot in Bangladesh and one of the most beautiful tourist places in the world. But the accommodation facilities are not suitable for tourists. The supply of electricity in the hotels is only 4-5 hours from the generator. For the geographical position, the electricity cannot supply from the mainland grid and the cost of electricity is so high and is not favourable to the environment. In this paper, Hybrid system of photovoltaic (PV), diesel generator, battery for generating electricity in the Saint Martin's Island is analyzed for 18 hotels. The main objective of the present study is to determine the optimum size of Hybrid system which can fulfil the requirements of 528 kWh/day primary load with 125 kW peak for 18 hotels in this island. By using HOMER (Hybrid Optimization Model for Electric Renewables) software an optimum model is established for the renewable system. The aim is to configure a renewable system with low interest and low energy cost. The diagrams and tables which show prices and performances of the types of equipment on the optimum model are also presented. The result shows that PV (185 kW), diesel generator (105 kW), converter (96 kW) and 615 piece batteries of the Hybrid system is most commercially reliable and least cost of energy is about 19.48Tk per kWh or $ 0.253 per kWh ($1=77Tk) with total net present cost $ 624,391 or 48,078,107TK. The emission of CO2 is very low in this Hybrid system.
Бесплатно
Статья научная
Precise mapping of water bodies is crucial for flood monitoring, disaster risk and response reduction, as well as sustainable water resource management. In this paper, we introduce a deep learning model for effective segmentation of rivers, lakes, and reservoirs from high-resolution Gaofen-2 satellite images. Leveraging the Five-Billion-Pixels dataset-more than 5 billion annotated pixels for 24 land cover classes—our approach solves the problem of segmenting water bodies on various terrains and environmental conditions. The proposed U-Net and ViT-UNet models, with the former employing Vision Transformers to enhance global context perception. For enhancing generalization, the dataset is augmented using Albumentations and flipping, rotation, and scaling transformations. Hybrid loss functions of Dice Loss, Binary Cross-Entropy, and Focal Loss are employed to handle class imbalance, especially for slender river segments. The ViT-UNet model attained 98.8% pixel accuracy, which mirrors its ability to preserve fine detail and large-scale spatial pattern. Mixed-precision training and the AdamW optimizer has enhanced the computational efficiency. Further, demonstrates the potential of transformer-based segmentation models for remote sensing achieved accuracy of 98% for environmental risk management and decision support in disaster-prone areas.
Бесплатно
