The research extensively examines mechanical testing and optimization of basalt fiber-reinforced nylon (BF-nylon) materials. All samples were produced by FFF 3D printer, specifically the Delta Wasp company’s model 2040 industrial X. Crucial mechanical evaluations such as compression testing, tensile testing, Izod impact testing and wear resistance analysis were performed to analyze the composite’s stiffness, strength, toughness and wear resistance. Minimal wear rate were 0.0293 mm³ m⁻¹, while highest compression, impact and tensile resistance strengths reached were 48.2 MPa, 58.4 MPa, and 40.79 J m⁻¹, respectively, in accordance with the major process parameters outlined in experimental matrix. Data obtained at experimental evaluation was employed to evaluate and train ANN model. The efficiency of BF-nylon were enhanced through (MOGA) multi-objective optimization employing a genetic algorithm. The impact resistance, wear rate, compression strength, and tensile strength, of 0.016275777 mm³/m, 33.11 J/m, 56.70 MPa, and 47.40 MPa, have been attained at layer height 335 mm, infill density 90
The Smart Roadways system transforms electric vehicle (EV) infrastructure by integrating Wireless Power Transmission (WPT) and IoT-based billing to enable seamless, on-the-go charging. A wireless power transmitter embedded in the road transfers energy to an EV’s receiver coil, ensuring continuous power without stopping at charging stations. The EV system, powered by an ESP32 microcontroller, interfaces with a battery, voltage sensor, LCD, buzzer, motor driver, and DC geared motors for real-time monitoring and control. The vehicle can be activated via a touch sensor, and upon entering the charging lane, it automatically begins wireless charging. Charging details, including voltage and cost, are displayed and monitored in real time via Blynk IoT integration. This innovation reduces range anxiety, enhances EV adoption, and contributes to a sustainable and intelligent transportation ecosystem.
The image data has grown fast in fields like social media, healthcare, and e-commerce and this has enhanced the demand of quick and precise image retrieval systems. Conventional methods of retrieval that have relied on handcrafted methods of feature extraction and pixel-wise comparison are costly in computer time and cannot withstand changes in image quality, lighting, and view point. To resolve these drawbacks, this project adopts an effective content-based image retrieval framework with a Vision Transformer (ViT) to extract features and FAISS (Facebook AI Similarity Search) to conducted fast similarity search. The system gets compact numerical representations of features of images as a result of a deep learning model and constructs a scale index to search similarity. The FAISS IndexIVFFlat architecture allows to search the nearest-neighbor approximationally with poor scale and low query time by grouping image features into clusters and lessening the search space at query time. Metadata of the indexed images are stored in the system and the top-k most similar images are retrieved according to the distance scores. This method is accurate and fast thus appropriate when dealing with large scale image databases. The system relies on the IndexIVFFlat structure of FAISS to cluster image features into regions which is a significant reduction of the search space that can support meaningful approximate nearest-neighbor queries. Results indicate that the system can be effective in separating useful and useless images that have high similarity (up to 0.819) and low distances (0.220) that ensure the accuracy and scalability of the system to image databases of large scales.
Cyberattacks have increased in impact and complexity and require more intelligent solutions to quickly and efficiently detect an attack. With increasing threats and sophistication of cyber-attack, there is a great need for timely and accurate real-time detection of malicious network activity with intelligent intrusion detection systems. The existing security-based model based on rules cannot detect new attacks or the vast amount of data contained within the amount of network traffic. A combined framework of some classical machine learning algorithms with Variational Quantum Classifier (VQC) is suggested as Quantum Machine Learning (QML) based Cyberattack Detection Framework in this paper as a solution. The proposed system employs a comprehensive preprocessing pipeline to generate a compact representation of high-dimensional network traffic features using feature normalization and Principal Component Analysis (PCA), enabling efficient cyberattack classification. The performance of the quantum classifier is compared to multiple machine learning models such as Random Forest, Support Vector Machine using Radial Basis Function (RBF) kernel and Extreme Gradient Boosting. A web platform is built with the Flask framework to provide secure user authentication, prediction services, attack risk assessment, historical record-keeping and PDF reporting. The framework can be used for both predicting benign and malicious network traffic and can also be used to provide levels of confidence and risk-level interpretation. The proposed system - that features scalable web deployment and advanced classification methods, such as machine learning, quantum inspired methods - is a large scale solution for intelligent cyber threat detection and decision support in the current network environment. Experimental tests revealed that the highest accuracy was achieved by using the Random Forest classifier with accuracy of 99.77% and the AUC-ROC as 0.9999, and the highest accuracy was achieved by the Variational Quantum Classifier with accuracy of 67.18%, which was able to help detect adversarial cyberattacks successfully thereby confirming the feasibility of quantum-assisted cyberattack detection.
Diabetic retinopathy (DR) is one of the biggest causes of vision loss in the world, however, retinal image analysis can help prevent permanent vision loss by allowing DR to be detected early. Automated analysis of retinal blood vessels plays a critical role in diabetic retinopathy screening, and this paper presents a DR analysis with a focus on the U-Net frameworks and its use to segment blood vessels from wide field retinal images. Our model uses a unique and modified U-Net model that has both batch normalization (BN) and dropout regularization (DR) to address over and out of sample variability. It is hosted on a back end and dual front end offer DRIVE support. The proposed model achieved a sensitivity of 83.92 % and an AUC of 0.983, demonstrating strong discrimination between vessel and background pixels. We do find blood vessel segmentation is still optimal in thin blood vessels. We find our model is both computationally efficient and clinically trustworthy which makes our segmentation of retinal vessels optimal in automated screening for diabetic retinopathy (DR) programs.
