In today's mechanical industries, accurate measurement and proper inspection are essential for maintaining product quality. Various gauges are used in workshops to check the dimensions, alignment, and tolerances of machine components. However, in many small and medium-scale industries, gauges are not properly arranged, which increases searching time, leads to improper handling, and may result in inaccurate inspection. This project focuses on the design and development of a Design & Manufacturing of welding Fixture to address these problems. The fixture is designed to provide a systematic arrangement, easy access, and safe storage of different types of gauges in one place. Mild steel (IS 2062) was selected as the manufacturing material due to its good strength, durability, weld ability, and low cost. The developed gauge board fixture reduces inspection time and improves measurement accuracy. It enhances workplace organization and minimizes damage to gauges.

Read More...

ResQr is a simple and useful system designed to help people during emergencies by using QR code technology. In many situations like accidents or sudden health problems, a person may not be able to share their identity or medical details. This can delay proper help and treatment. The ResQr system solves this problem by storing important information such as name, blood group, medical conditions, and emergency contact numbers in a secure database. Each user gets a unique QR code that can be printed or saved on their phone. In an emergency, anyone can scan this QR code using a smartphone to quickly view the person's details and contact their family or doctors. The system also allows quick actions like calling emergency contacts. This solution is easy to use, low-cost, and helps in saving time during critical situations. It can improve emergency response and increase the chances of saving lives.

Read More...

An adaptive filter is a self-adjusting digital signal processor that minimizes error signals and adapts to dynamic environments. It uses algorithms like Least Mean Squares (LMS) and Recursive Least Squares (RLS) for tasks like noise cancellation and system identification. Traditional adaptive algorithms are effective with Gaussian noise but struggle with sudden disruptions. This research focuses on developing robust adaptive filters to manage various noise types, utilizing MATLAB models that demonstrate superior performance in real-time applications. For creating noisy signals four types of noises considered AWGN (Additive White Gaussian Noise), Johnson, random and uniform are considered. A hybrid LMS-RLS filter combines the stability of LMS with the quick response of RLS, performing better in non-stationary conditions with improved precision and reduced computational demand.

Read More...

Online product reviews play an important role in influencing customers’ purchasing decisions on e-commerce platforms. However, the increasing presence of fake or misleading reviews has significantly reduced the reliability of these systems and created challenges for both users and businesses in making trustworthy decisions. This paper presents a machine learning–based approach for identifying fake product reviews by analyzing textual patterns and review characteristics. The proposed system uses natural language processing techniques to preprocess review data and applies feature extraction methods such as TF-IDF to transform textual information into numerical form suitable for classification. Multiple classification algorithms are trained and evaluated to distinguish genuine reviews from deceptive ones effectively. The performance of the system is measured using evaluation metrics such as accuracy, precision, recall, and F1-score. Experimental results demonstrate that machine learning techniques can successfully detect suspicious reviews with satisfactory accuracy and reliability. The proposed approach can help e-commerce platforms improve the authenticity of the review, improve customer trust, and help users make better purchasing decisions. Future improvements may include the integration of deep learning models and real-time detection mechanisms for enhanced Conference performance.

Read More...

Comparators are fundamental components in analog-to-digital converters (ADCs). The overall performance of an ADC strongly depends on the speed, power consumption, and accuracy of the comparator. Traditional CMOS-based comparators suffer from leakage current and short channel effects when implemented in deep submicron technologies. To address these challenges, this paper presents a three-stage dynamic comparator implemented using 20-nm FinFET technology. The proposed design incorporates a tail transistor in the latch stage to reduce power consumption and improve energy efficiency. The circuit was designed and simulated using Cadence Virtuoso with FinFET device models. Simulation results show significant improvement in power efficiency compared with conventional three-stage comparator architectures, making the proposed design suitable for high-speed and low-power ADC applications.

Read More...

The rapid expansion of Internet of Things (IoT) devices has introduced significant security challenges due to their limited computational capabilities and increased exposure to cyber threats. Detecting anomalies in IoT networks has become essential for maintaining system integrity and preventing unauthorized activities. This paper presents a comparative analysis of various machine learning techniques for anomaly detection using secondary data derived from existing research studies. Techniques such as Random Forest, Decision Tree, Naïve Bayes, Support Vector Machine (SVM), and Neural Networks are evaluated based on performance, computational complexity, and suitability for resource-constrained environments. The analysis highlights that lightweight models provide a practical balance between efficiency and detection accuracy, whereas deep learning techniques offer improved performance at the cost of higher resource consumption. The study also emphasizes the importance of selecting appropriate models based on system requirements and available computational resources.

Read More...

This study evaluates bagasse ash as a partial replacement for cement to reduce environmental impact. The ash was ground to match the fineness of Portland cement and analyzed for its chemical composition. Concrete mixes were prepared for 30 MPa strength with varying water-cement ratios and bagasse ash replacements of 5%, 10%, and 15%. Tests on fresh and hardened concrete showed that optimal mix proportions improved strength and reduced cement use, leading to cost savings. However, compressive strength decreased beyond 10% replacement. Overall, all mixes satisfied ASTM C618 requirements.

Read More...

Due to its performance above diesel fuel with little or no engine modification, biodiesel is a useful fuel today. This paper discusses waste cooking oil biodiesel production and attributes. Different Waste Cooking oil biodiesel-diesel blends and loads can be tested to improve biodiesel-fueled diesel engine performance. Experimental work uses a single-cylinder four-stroke CI engine. Taguchi design with Minitab software will be utilised to test different load and blend ratios for brake power, fuel consumption, thermal efficiency, and pollution. Lab tests were done as advised by the software. Finally, in CI engine, Diesel, waste cooking-oil biodiesel, and diesel are tested for performance and emission parameters and optimisation.

Read More...

Travel planning often becomes difficult because users need to use different applications for navigation, itinerary planning, and travel information. Most existing travel applications provide limited features and do not support complete trip planning in one place. To solve this problem, this paper presents TravelEase, a travel planning system developed for web and mobile platforms. The system allows users to make travel arrangements such as hotel booking, transportation, and selecting tourist attractions on a single platform. Studies on online tourism systems discuss a lot about booking service systems at hotel reservations, transportation reservations, or booking tour tickets. However, most of these studies were carried out only on individual processes and were not integrated. This study built a web service for travel package booking that allows travelers to book transportation, hotels, and travel tickets in one transaction. The system allows users to find routes, create travel itineraries, and receive assistance through chatbot and voice features. TravelEase is designed using modern web and mobile technologies to provide easy access and better user experience. The system helps users save time and reduces the effort required for planning trips by combining important travel features into a single platform.

Read More...

Face recognition systems are increasingly used for authentication in various applications such as banking, mobile security, and access control. However, these systems are highly vulnerable to spoofing attacks, including the use of printed photographs, video replays, and masks. To address this issue, this paper proposes a Face Liveness Detection System that can effectively distinguish between real human faces and fake rep- resentations. The proposed system utilizes computer vision and machine learning techniques to analyze dynamic facial features such as eye blinking, facial movements, and texture patterns. A dataset containing both real and spoofed facial inputs is used to train and evaluate the model. The system operates in real-time using a webcam and provides accurate classification of live and fake inputs. Experimental results demonstrate improved detection accuracy and robustness against common spoofing methods. This approach enhances the reliability and security of face recognition systems, making it suitable for real-world biometric authentication applications.

Read More...

This paper presents the design and development of a four-wheel drive (4WD) robotic vehicle integrated with a rack and pinion steering mechanism and Bluetooth-based control system. The system aims to improve traction, steering accuracy, and stability compared to conventional robotic platforms. Four DC motors are used to drive each wheel independently, ensuring uniform torque distribution. The steering mechanism converts rotational motion into linear displacement using a rack and pinion system for precise control. An Arduino Uno microcontroller processes wireless commands received via an HC-05 Bluetooth module to control movement. The proposed system demonstrates improved mobility, efficient control, and cost-effective design suitable for educational and research applications.

Read More...

Predictive analytics has become a critical component in modern healthcare systems, enabling early disease detection and improved clinical decision-making. This study proposes a machine learning-based predictive analytics model for early disease detection using patient clinical data. Multiple supervised learning algorithms including Logistic Regression, Decision Tree, Random Forest, and Support Vector Machine were implemented and evaluated. The dataset used for the study consists of patient health indicators such as age, blood pressure, cholesterol level, and glucose level. Data preprocessing techniques such as normalization, missing value handling, and feature selection were applied before model training. Experimental results show that the Random Forest model achieved the highest accuracy compared to other algorithms. The proposed predictive model can assist healthcare professionals in identifying high-risk patients at an early stage, thereby improving treatment outcomes and reducing healthcare costs.

Read More...

Accurate psychological evaluation for adventure sports participation is an increasingly critical requirement within modern occupational and sports medicine. Conventional assessment methodologies, predominantly reliant on self-reported questionnaires and structured clinical interviews, are inherently susceptible to subjective bias, social desirability effects, and an inability to capture real-time physiological reactions under authentic fear-inducing conditions. This paper presents the Neuro-Adaptive Phobia Diagnostic System (NPDS), a comprehensive and integrated framework that synergistically combines immersive Virtual Reality (VR) simulation, continuous physiological signal acquisition, and unsupervised machine learning for objective and reproducible phobia detection. The proposed system immerses candidates in photorealistic, scenario-specific VR environments engineered to evoke common phobic responses—including acrophobia (heights), claustrophobia (confined spaces), aquaphobia (deep water), nyctophobia (darkness), and hylophobia (dense forests). Concurrently, cardiac activity, specifically heart rate, is monitored via wearable biosensors and transmitted in real time through the Lab Streaming Layer (LSL) protocol. Anomaly detection is performed using the Isolation Forest algorithm, which constructs a personalized physiological baseline during an initial calibration phase, thereby enabling precise differentiation between transient stress elevations and genuine phobic autonomic responses. To bridge the translational gap between raw anomaly scores and actionable clinical intelligence, NPDS incorporates a Retrieval-Augmented Generation (RAG) module that retrieves contextually relevant medical literature from a curated vector database and synthesises structured, human-readable diagnostic reports. A dual-interface web platform—built on Next.js and FastAPI—facilitates real-time monitoring for clinicians and accessible summary reports for patients. Experimental evaluation conducted across multiple simulated phobia scenarios demonstrates that NPDS achieves reliable anomaly detection, seamless VR-physiological synchronization, and clinically interpretable outputs. The system represents a significant step toward objective, scalable, and bias-free phobia screening in adventure sports medical clearance contexts.

Read More...

Critical infrastructure faces increasing risks from both physical hazards and cyber threats. This paper proposes C3S, an integrated cyber–civil security framework combining structural engineering, geotechnical analysis, sustainable design, and ISO 27001–aligned cybersecurity. The system incorporates AI-based predictive models to identify vulnerabilities across physical and digital domains. Simulation results demonstrate improved structural resilience, reduced environmental impact, and enhanced cybersecurity performance. C3S provides a scalable and unified approach for securing modern cyber-physical infrastructure.

Read More...

Autonomous vehicles rely heavily on multi-sensor systems to perceive and interact with their environment. However, these sensors are vulnerable to malicious attacks such as spoofing, jamming, and adversarial manipulation. This study proposes a hybrid framework integrating residual-based anomaly detection and machine learning classification for effective detection and isolation of sensor attacks. Using a simulation-based experimental design in the CARLA environment, the proposed model achieved high detection accuracy (96.4%) and isolation accuracy (93.1%) across multiple attack scenarios. The findings highlight the importance of integrating statistical and learning-based approaches to enhance system resilience. The study contributes to both theoretical and practical advancements in autonomous vehicle security.

Read More...

Efficient and clean energy charging infrastructures are necessary due to the rising demand for sustainable energy solutions and the fast expansion of electric cars (EVs). A smart hybrid renewable energy based wireless power transfer (WPT) system that uses wind and solar photovoltaic (PV) to power the grid and electric vehicle charging stations is proposed in this project. Through the integration of various renewable energy sources, the system guarantees a consistent and dependable energy supply, lessens the impact of intermittent power production, and decreases reliance on traditional electricity generated from fossil fuels. In addition to offering a secure and hassle-free way for electric vehicles to charge wirelessly, the system is engineered to maximize energy consumption via smart energy management. A hybrid renewable energy generating unit, energy storage components, power conditioning units, and a resonance WPT module for electric vehicles make up the suggested system. Solar panels, wind turbines, battery storage, and WPT transmitters all contribute to a constant stream of real-time data that is collected by an Internet of Things (IoT) control and monitoring platform. Integration with grid demand response systems, remote monitoring, fault detection, adaptive load management, predictive charging schedules, and the Internet of Things (IoT) system are all made possible. To maximize system efficiency and minimize energy losses, this link guarantees that renewable energy is routed effectively between grid support and EV charging. The topic of wireless energy is the focus of this study. What we call "wireless power transfer" or "wireless energy transmission" really refers to the movement of electrical current from a generator to an appliance or other device that uses it, all without the need of physical wires. Power transmission methods that employ time-varying electro-magnets are collectively referred to by this name. When running cables between devices is too much of a hassle, too risky, or just not an option, wireless transmission is a lifesaver. One equipment, known as a transmitter, is linked to a power source—the mains power line, solar panels, or wind turbines—and uses electromagnetic fields to send electricity over space to another device, or devices, that can convert it back into electric power.

Read More...

Mental health issues are rising due to stress, lifestyle changes, and limited access to timely support. This paper presents an AI Mental Wellness App an intelligent platform to monitor, understand, and improve user mental well-being. The system uses Python with Django for the backend, HTML/CSS/JS/Bootstrap for the web interface, Flutter for Android, and MySQL for data management. It includes three modules: User (registration, wellness exercises, mini-games, AI chatbot, recommendations, doctor views, feedback), AI (sentiment analysis, facial emotion detection, speech-to-text for emotional understanding), and Admin (login, user management, feedback response, game maintenance). By integrating AI with mobile and web technologies, the app offers accessible, proactive mental health support, enabling early detection of distress and connecting users to professionals. Initial testing shows improved user engagement and emotional awareness. Future work includes clinical validation and multilingual support.

Read More...

This project presents an AI-powered system for identifying fraudulent profiles using behavioral analysis. With the rapid growth of online platforms, fake profiles have become a major concern, leading to misinformation, scams, and security threats. The proposed system analyzes user behavior patterns such as posting frequency, interaction style, language usage, and network connections to detect anomalies. Machine learning algorithms, including classification models and anomaly detection techniques, are employed to distinguish between genuine and fake profiles. The system leverages historical and real-time data to improve accuracy and adaptability. By integrating natural language processing and pattern recognition, it identifies suspicious activities that are difficult to detect using traditional methods. The model is trained on diverse datasets to ensure robustness and scalability. Experimental results demonstrate high accuracy and efficiency in detecting fraudulent accounts. This approach enhances platform security, reduces malicious activities, and provides a reliable solution for social media and professional networking platforms.

Read More...

Automatic floor cleaning robots have become increasingly relevant in modern households and commercial spaces for their ability to reduce human labor and improve cleaning efficiency. This paper presents the design and development of an autonomous floor cleaning robot utilizing ESP microcontroller technology. The system integrates sensors, motor control, and wireless communication to enable efficient cleaning with minimal human intervention. The robot is equipped with obstacle detection, path planning, and cleaning mechanisms, controlled via a mobile application. This approach aims to produce a cost-effective, customizable, and efficient cleaning solution suitable for indoor environments. This project presents the design and implementation of an Automatic Floor Cleaning Robot based on the ESP32 microcontroller. The primary objective is to automate indoor floor cleaning with minimal human intervention, improving convenience and efficiency in homes and offices. This system reduces manual effort, saves time, and can be further enhanced with features like wet mopping, floor mapping, and IoT-based scheduling.

Read More...

Access to safe drinking water is essential for public health, yet conventional water quality monitoring systems are often slow, manual, and lack real-time accessibility. This paper presents Aqua Trace, an intelligent water quality monitoring and management platform that integrates mobile and web technologies with machine learning for water classification, contamination mapping, lab-oratory report delivery, and predictive filter recommendation. The system uses Random Forest for water quality classification, Linear Regression for filter prediction, and K-Nearest Neighbour interpolation for contamination visualization. Developed using Django REST Framework, Flutter, MySQL, and Google Maps API, Aqua Trace improves efficiency, transparency, and public accessibility in smart water management. Experimental results confirm faster reporting, accurate contamination detection, and enhanced decision-making for water safety.

Read More...

The present era very vast development occurred in the field of construction, specially in concrete technology, Concrete is the most versatile material due to its continuous demand. The present research work focused on one such product, i.e. Micro silica Micro ilica is a highly effective pozzolanic material widely used in modern concrete technology. It is an ultrafine by-product of silicon and ferrosilicon alloy production, characterized by its very small particle size and high amorphous silica content. This study investigates the effects of micro-silica on concrete properties by incorporating it in varying proportions from [5%, 10%, 15%, 20%, 25%] by weight of cement. Key parameters examined include workability, compressive strength, and splitting tensile strength. The results indicate that the improvement is mainly due to its pozzolanic reaction with calcium hydroxide, forming additional calcium silicate hydrate (C-S-H) gel, and resulting in a denser and stronger matrix. In fresh concrete, micro-silica improves cohesiveness and reduces segregation and bleeding.

Read More...

This research focuses on the enhancement and optimization of weld bead geometry in AISI 316L austenitic stainless steel using the Flux-Activated Gas Tungsten Arc Welding (A-TIG) process. Although conventional TIG welding is widely used due to its superior weld quality and control, it is limited by shallow penetration, which often requires multiple welding passes. This leads to increased heat input, higher distortion, and reduced process efficiency. The ATIG technique overcomes these limitations by applying an activating flux on the workpiece surface, which modifies weld pool behavior through arc constriction and reversal of Marangoni convection, thereby improving penetration characteristics. In the present study, silicon dioxide (SiO₂) flux was applied on 6 mm thick AISI 316L plates to investigate its influence on weld bead geometry. The experimental design was developed using the Taguchi L18 orthogonal array, considering key welding parameters such as welding current, travel speed, and pulse frequency. The quality of welds was evaluated in terms of depth of penetration, bead width, and angular distortion. In addition, conventional TIG welding was performed under similar conditions to provide a comparative assessment. The experimental results indicate that A-TIG welding with SiO₂ flux significantly enhances penetration depth and improves the aspect ratio of the weld bead compared to conventional TIG welding. The improvement is attributed to increased heat concentration at the weld center due to arc constriction effects. Statistical analysis using Signal-to-Noise (S/N) ratios and Analysis of Variance (ANOVA) revealed that welding current is the most influential parameter affecting penetration, followed by travel speed and pulse frequency. The optimal parameter combination was identified and validated through confirmation experiments. It can be concluded that the application of SiO₂ flux in A-TIG welding is an effective and economical approach for achieving deeper penetration, reduced distortion, and improved welding efficiency in AISI 316L stainless steel. The findings of this study provide valuable insights for industrial applications where high-quality, single-pass welding is required, particularly in sectors such as petrochemical, marine, and power generation industries.

Read More...

This project presents an AI-powered Bone Fracture Detection system that analyzes X-ray images using advanced deep learning models and web technologies. The system integrates multiple object detection models, including YOLOv8, Faster R-CNN, and SSD, to accurately identify fracture regions in medical images. A Flask-based web application is developed to provide an interactive interface for uploading X-rays, selecting models, and visualizing detection results. Image preprocessing techniques such as noise reduction and contrast enhancement are applied to improve detection performance. The system also incorporates AI-generated medical reports using Gemini, offering detailed fracture analysis, severity assessment, and clinical recommendations. Additionally, features like grid overlay and confidence threshold adjustment enhance usability and precision. This solution aims to assist medical professionals by providing quick, reliable, and automated fracture detection, thereby improving diagnostic efficiency and reducing human error in radiographic analysis.

Read More...

This paper introduces a context aware framework of multimodal narrative generation that allows personalized story generation with the help of structured conditioning of large language models (LLMs). The proposed approach differs in that it incorporates the hierarchy of context modeling, where demographic factors, including age, interest type, and tone preference, control emotional interpretation, whereas visual scene descriptions control vocabulary and structure of the environmental context. The architecture uses guided prompt engineering with a massive contextually constrained instruction-tuned language model (LLaMA-3.3-70B) to generate stories under explicitly defined contextual constraints. In order to test the efficacy of personalization, we implement controlled experiments in three settings, which are a non-personalized control group, a child-focused profile, and an adult-focused dark-fantasy profile. Every setup is tested in 15 independent generations to cover stochastic random variation in the outputs of LLM. The personalization effects are measured in terms of a multi-dimensional assessment system that takes into consideration lexical diversity (Type-Token Ratio) and average sentence length, sentiment polarity, and Shan-non entropy. Experimental findings show statistically consistent personalization behavior: there are greater lexical diversity (0.846 vs. 0.743 in the case of baseline) and longer sentence structures, and sentiment polarity between adult-oriented and child-oriented narratives exhibits directional change according to demographic conditioning (+0.256 in the case of child narratives and -0.196 in the case of adult dark-fantasy narratives) in accordance with the demographic conditioning. The hierarchical fusion behavior is also demonstrated by the multimodal experiments where the same dark visual contexts appear positively when interpreted with child profiles and negative, horror-oriented narrative when interpreted with adult profiles. These results confirm that context-sensitive prompting in a structured form allows lexical, structural, and emotional adjustment to be measured in multimodal narrative generation systems, and it is a scaled and assessment-based way of approaching personalised generative AI.

Read More...

Urban flooding in Kerala can no longer be understood only as the outcome of intense rainfall, it increasingly reflects the cumulative effect of human alteration of land, drainage systems, and ecological buffers and other human-induced modifications to the urban fabric. This reviewed literature case studies consistently evaluates the variability of anthropogenic variables that reduce the capacity of urban landscapes to absorb, store, and convey water safely and that regulate contemporary urban flood inundation patterns. This paper also develops a clear clarity on how these variables influence runoff, storage, and exposure. Evidence drawn from the case studies of Okitipupa, Bhopal, Bangalore, Mumbai, Hyderabad, and Surat to identify recurring anthropogenic parameters, explain how they modify urban hydrology, and consolidate them into a practical indicator framework for flood vulnerability analysis. The cases consistently show that flood severity is controlled not only by rainfall intensity but also by built-up expansion, wetland loss, less vegetation cover, surface sealing, floodplain encroachment, blocked drainage, poor land-use regulation, waste dumping, and high population exposure. The study further explains how GIS, remote sensing, census data, and field observations can be combined to quantify these drivers through normalization, weighting, and composite vulnerability indices. By synthesizing findings from the provided case study, the paper highlights the role of human-induced modifications to the urban landscape in exacerbating flood risks and by addressing these parameters through efficient planning can make urban flood resilient cities.

Read More...