Browse
Recent Submissions
Item Allergy and asthma trigger app(NHCE, 2025) Manoj CK 1NH21EC088; Mohammed Faizan 1NH21EC069; Zarhan Yaqoob 1NH21CS270; Vishal N 1NH21CS262Allergies and asthma are prevalent and millions of Allergies and asthma chronic health conditions globally, significantly affecting individuals' quality of life. environmental triggers, such as air pollution, seasonal allergens, and abrupt weather changes, which can provoke severe reactions in sensitive individuals. Pollutants like PM2.5, PM10, NO2, and CO, commonly found in urban and industrialized areas, exacerbate respiratory symptoms, making it essential for those affected to monitor their surroundings closely. Unfortunately, the rising levels of pollution and unpredictable climate shifts have made it increasingly difficult for people to avoid these triggers. Advancements in mobile technology and machine learning issue through intelligent applications. By integrating real-time environmental data, such as air quality indices and pollutant concentrations, with predictive algorithms, these apps can offer users a proactive way to manage their respiratory health. data and environmental patterns to predict hazardous air quality conditions, while mobile platforms ensure that this information is readily accessible to users wherever they are. Such apps not only alert users to high-risk situations but also provide personalized recommendations, empowering individuals to take preventive actions, such as times or using protective measures like masks and air purifiers.Item Leveraging Advanced Deep Learning techniques For Chest X-ray Tuberculosis Detection(NHCE, 2025) Hashim khan GJ 1NH21CS096; Kandimalla Rishik 1NH21CS123; Priyanshi Bharvesh 1NH21EC121; Sindhu Bhargavi 1NH21EC152Tuberculosis (TB) continues to be a major global health issue, especially in low-income and developing regions, where it claims around 1.5 million lives annually, despite being both preventable and treatable. Traditional TB diagnosis through chest radiographs, analyzed by skilled radiologists, is often time-consuming, prone to variability in interpretation, and vulnerable to misclassification, particularly with other diseases that have similar radiologic features. These challenges are even more pronounced in rural and underserved areas where access to experienced radiologists is limited. To tackle these issues, this project proposes the creation of an automated deep learning-based system for TB detection using chest X-rays. By harnessing the power of Convolutional Neural Networks (CNNs), the system aims to enhance both diagnostic accuracy and accessibility. The proposed system intends to streamline the detection process by learning deep features directly from raw image data, eliminating the need for manual feature extraction, which is typically required in conventional machine learning models. Previous studies have shown the efficacy of CNNs in medical image analysis, with numerous models achieving impressive accuracy rates. This project builds on these advancements to develop a robust deep learning model specifically designed for TB detection, focusing on improving classification performance, consistency, and reliability. Key advancements in TB detection have been driven by the evolution of sophisticated CNN architectures and the availability of extensive annotated datasets, which have enabled the development of highly accurate models. This project will leverage cutting-edge deep learning techniques and rigorously evaluate the model's performance in various clinical contexts, ensuring its robustness across different healthcare settings. The primary goal is to create a system that not only matches but exceeds the diagnostic capabilities of human radiologists, offering a vital tool for early, precise TB detection, particularly in resource-constrained environments. In addition to improving diagnostic efficiency, this system aims to provide a scalable solution that can be widely implemented across diverse healthcare systems, especially in regions with limited medical infrastructure. The system’s integration into clinical workflows could transform TB screening, making it faster, more reliable, and more accessible. By enhancing early detection capabilities, this project seeks to improve patient outcomes, reduce transmission rates, and contribute significantly to global efforts aimed at eradicating tuberculosis.Item Smart pill reminder(NHCE, 2025) Nischitha M 1NH21IS103; Meghana P 1NH21IS198; A Leela Sagar 1NH21EC001; R Rahul 1NH21EC124Nowadays many elderly people live alone, some with disabilities, who find it difficult to take care of themselves. Most of the elders have a couple of chronic illnesses, and they use capsules to stabilize their health status but old age patients forget to take pills on proper time or as per the prescription which causes certain health issues for patients having permanent diseases like diabetes, blood pressure, breathing problems, heart problems, cancer diseases etc. Taking proper medicine at proper time is necessary to become healthy, but failure of that can create big trouble for a patient. Medical errors are occurred due to the fact that patients and caretakers have to deal with sorting time and dosages. To overcome this, we propose a smart medicine box for those people who regularly take medicines, and the prescription of their medicine is very long as it is hard to remember. Our medicine box contains three sub pill boxes. Caregiver can set up time for these three sub pill boxes. Pill boxes are pre-loaded in the system which patients need to take at given time which reduces caregiver’s responsibility towards giving the correct and timely consumption of medicines. When time of pill is set, pillbox will remind to take pill at a particular time and the pills required to take at that time comes out to the user to avoid confusion among medicines. The medicine box has a rectangular shape having three compartments for pill. The DC motor is used for automatic dispensing of the pill. An LCD display is used to display the medicine name and the time at which the medicine to be taken. Microcontroller interface with LCD display, LED, DC motor and speaker are used to provision for visual and audio notifications are to be provided. A vibrator sensor is used to notify the specific box especially for the blind people who can sense it for the intake of medicine at the appropriate time.Item Earthquake Prediction and Alerting System Using Machine Learning and Raspberry Pi(NHCE, 2025) Likhitha M1NH21EE051; Likhith Kumar N 1NH22EC405; Santhosh 1NH22EC411; Karishma Palekar 1NH22CS405; Mohan Kumar S 1NH22CS412This report presents a novel method for earthquake warning and prediction that makes use of machine learning and inexpensive hardware integration with a Raspberry Pi. With the help of an MPU6050 accelerometer, vibration sensor, and BMP180 sensor, the suggested system may continuously monitor environmental and seismic parameters like temperature, pressure, altitude, acceleration, and vibrations. Magnitude, Peak Ground Acceleration (PGA), and Root Mean Square (RMS), three crucial seismic characteristics, are obtained from the constant stream of data produced by these sensors. A rich and varied dataset is created using the gathered data, which serves as the basis for training and assessing different machine learning algorithms. To determine which model has the best accuracy for predicting earthquakes, the system thoroughly evaluates various models. After that, the chosen model is installed on a Raspberry Pi, which allows for real-time prediction using an integrated Flask server. In addition to processing real- time data, this server helps the system run independently and deliver alerts on time. The system's connection with Twilio's messaging API, which permits the automatic sending of SMS messages depending on observed seismic activity and earthquake magnitude, is one of its most notable features. The purpose of these alerts is to give users advance notice so they can take preventative measures like leaving the area or finding cover. The system demonstrates the possibility of developing an affordable and scalable earthquake prediction and alerting system by fusing machine learning with Internet of Things technology. Future improvements, such adding more sensor types or extending deployment to other areas, are made possible by its modular architecture. This study emphasizes how important technology is in reducing the dangers of natural catastrophes and how it may improve public safety, especially in areas where earthquakes are common. In order to minimize the loss of life and property in earthquake-prone locations, this system is an essential step in closing the gap between scientific discoveries and real-world disaster preparedness implementations.Item Design and development of a high-efficient nanowire fet for emerging logic devices(NHCE, 2025) Harshavardhan PR 1NH21EC066; Niranjan sharma SH 1NH21EC110; Harshitha S 1NH21EE037; Likhitha M 1NH21EE051The exponential growth of semiconductor technology has driven the need for advanced transistor designs to overcome the limitations of traditional planar Metal-Oxide-Semiconductor Field-Effect Transistors (MOSFETs). As device dimensions shrink to sub-nanometer scales, critical challenges such as short-channel effects (SCEs), increased leakage currents, and elevated power dissipation arise, hampering performance and increasing manufacturing complexities. In response, Gate-All-Around (GAA) FETs have emerged as a revolutionary architecture, offering superior electrostatic control, reduced SCEs, and enhanced scalability. These attributes make GAA FETs ideal candidates for next-generation logic devices. Among the innovative implementations of GAA FETs, nanowire technology stands out as a game-changer. By utilizing cylindrical or quasi-cylindrical channels completely surrounded by the gate, nanowire FETs deliver unparalleled gate control, minimized leakage, and improved switching performance. This design not only supports extreme miniaturization but also ensures energy efficiency and high-frequency operation, aligning perfectly with the demands of modern applications such as artificial intelligence, 5G, and IoT. Despite these architectural advancements, there exists a gap in understanding the influence of different channel materials on the performance of GAA nanowire FETs. Silicon (Si), a widely used semiconductor material, offers excellent manufacturability and well-established fabrication processes. However, wide-bandgap materials like Silicon Carbide (SiC) and Gallium Nitride (GaN) provide unique advantages, such as higher breakdown voltages, superior thermal conductivity, and lower leakage currents, making them attractive for high-power and high-frequency applications. This project, "Design and Development of a High-Efficient Nanowire FET for Emerging Logic Devices," conducts a comprehensive comparative analysis of GAA nanowire FETs with Si, SiC, and GaN as channel materials. The study evaluates critical performance parameters, including threshold voltage (Vth), subthreshold slope (SS), and the on/off current ratio (Ion/Ioff). Advanced simulation techniques are used to explore the interplay between material properties, device design, and scalability. The findings underscore the importance of material selection in optimizing nanowire GAA FETs for specific applications. While Si remains a versatile option for general-purpose logic devices, SiC and GaN demonstrate superior performance for high-power, high-frequency applications. This research bridges the gap between architecture and material innovation, paving the way for high-efficiency, scalable, and sustainable logic devices that surpass the limitations of conventional MOSFETs. The outcomes contribute significantly to the advancement of semiconductor technology, enabling novel computing and IoT applications. .Item Drip IV rate Monitor(NHCE, 2025) Anjay B 1NH21EC136; Sharan Prasad M 1NH21EC145Intravenous (IV) therapy is a cornerstone of modern medical treatments, providing a direct route for administering fluids, medications, and nutrients to patients. This method ensures rapid and efficient delivery, bypassing the digestive system to achieve immediate therapeutic effects. Despite its critical importance, the traditional manual monitoring of IV fluid levels is prone to human error, leading to potential risks such as air embolism, dehydration, and treatment delays. Additionally, manual monitoring demands significant time and attention from healthcare providers, diverting resources from other essential tasks. Addressing these challenges, this project presents the design and implementation of an intelligent IV monitoring system aimed at automating the process, enhancing accuracy, and ensuring timely alerts when IV fluid levels are low. The core of the intelligent IV monitoring system is an ESP32 microcontroller unit (MCU), a versatile and powerful component known for its high performance and low power consumption. The ESP32 MCU integrates various input and output components, forming a comprehensive system capable of monitoring and managing IV fluid levels autonomously. At the heart of the monitoring mechanism is a 3KG load cell, which accurately measures the weight of the IV bottle. The load cell's output is a low-voltage analog signal, requiring amplification to ensure precise data capture. To achieve this, the signal is amplified by an HX711 amplifier, a specialized component designed for load cell applications. The amplified data is then transmitted to the ESP32 MCU via a uni-directional SPI (Serial Peripheral Interface) protocol, ensuring seamless and efficient communication between the hardware components. To accommodate different IV therapy requirements, the system incorporates three control buttons, allowing users to select from three standard IV bottle sizes: 250ml, 500ml, and 1 liter. This functionality is crucial for calibrating the system according to the specific IV bottle in use, ensuring accurate monitoring regardless of the bottle size. Once the appropriate bottle size is selected, the ESP32 processes the incoming data, calculating the remaining IV fluid volume based on the weight measurements.Item Human Body Micromotion Sensing and Detection Using 24GHz mmWaves(NHCE, 2025) Harsh Jashvantbhai Valaki 1NH21IS060; Vinayak B Sohar 1NH21IS178; Aryaman Prajin 1NH21EC023; Siddharth Rajesh 1NH21EC150; Vinayak S Bangarshetra 1NH21EC184This paper presents a detailed case study on the application of 24GHz millimeter-wave (mmWave) technology for micro-motion sensing detection. Millimeter-wave sensing has gained significant attention due to its ability to detect subtle human movements with high precision, offering superior resolution compared to conventional RF-based sensing systems. The 24GHz frequency band, widely used in various industrial and consumer applications, provides a balance between signal range, resolution, and power consumption, making it suitable for micro-motion detection in smart environments. The study investigates the performance of 24GHz mmWaves in detecting micro-motions such as hand gestures, breathing patterns, and small body movements in indoor environments. By leveraging the Doppler effect and variations in radio signal reflection, the system can accurately detect and classify human motion without requiring complex sensor installations. This paper discusses key challenges, including signal attenuation, environmental interference, and calibration issues. Solutions such as beamforming and multiple-input multiple-output (MIMO) techniques are proposed to mitigate these challenges, ensuring robust detection in real-world applications. The experimental results demonstrate that 24GHz mmWave technology can achieve high detection accuracy and low latency, making it suitable for applications in healthcare, security, and smart homes. Additionally, this paper highlights the potential for integrating 24GHz mmWave sensors with IoT frameworks to enable real-time data processing and enhanced automation in smart environments. The findings underscore the feasibility of using 24GHz mmWave technology as a cost-effective and efficient solution for precise motion sensing in diverse applications.Item Voice command systems for drivers(NHCE, 2025) Vinay K 1NH21EC182; Rahul S 1NH21EC126; M Sai Vamsi 1NH21AI135; Likhiteswar Reddy V 1NH21AI114This project introduces a Bluetooth-enabled, voice command system tailored for drivers, allowing hands-free control over essential vehicle functions, particularly beneficial for tasks such as reversing. Utilizing Bluetooth technology, the system seamlessly connects a smartphone or other Bluetooth-enabled device to the vehicle’s control module, enabling users to issue voice commands in any language to perform specific tasks. The system’s architecture includes a microphone for capturing voice input, a speech processing unit for command recognition, and a control unit that interfaces with the vehicle's motor mechanisms. By employing natural language processing techniques, the system ensures accurate recognition of diverse languages and dialects, enhancing accessibility for a broad range of users. Designed with safety and convenience in mind, this innovative solution minimizes physical interaction with vehicle controls, promoting a more intuitive and safer driving experience. The proposed system’s effectiveness and accuracy have been evaluated through extensive testing in varied acoustic environments, demonstrating its potential to improve driving accessibility and usability. This project explores the technical challenges faced, the solutions implemented, and the broader implications for future vehicle automation and smart mobility technologies.Item Vision Assist(NHCE, 2025) Rakshit D 1NH21EC127; Tiba Sidra 1NH21EC163; Sirisha P 1NH21IS153; Zeba Syed Farooq 1NH21IS192Vision Assist is an advanced system designed to provide visually impaired individuals with a reliable and intelligent aid for navigating their environment safely and independently. Utilizing cutting-edge deep learning algorithms, this system interprets real-time camera input to deliver accurate and actionable feedback. It is fully compatible with smartphones, ensuring ease of access and usability for a wide range of users. The system offers a comprehensive suite of features tailored to the unique needs of visually impaired users. Object detection identifies and classifies objects in the user’s surroundings, helping them understand their immediate environment. Obstacle detection ensures safety by alerting users to potential hazards in their path. Currency recognition enables users to manage financial transactions confidently, while text detection, powered by Optical Character Recognition (OCR), allows them to access and interact with printed materials. To further enhance its functionality, Vision Assist incorporates Smart Glasses that can read text aloud, providing real-time auditory feedback for an intuitive and hands-free experience. The integration of GPS navigation offers step-by-step guidance for outdoor travel, ensuring users can move through unfamiliar areas with confidence and ease. By combining these advanced features, Vision Assist is a transformative tool that enhances accessibility and fosters greater independence for visually impaired individuals. Its intelligent design bridges the gap between visual impairment and the need for autonomy, making it a valuable solution for improving mobility, safety, and overall quality of life.Item Smart Hydroponics system(NHCE, 2025) Shashank CR 1NH21EC146; Bhimaraj K Naikar 1NH2215402; Sumanth Rrai 1NH21EC158; Darshan V 1NH2115047Hydroponics refers to the art of growing plants in water without soil. Nutrients for the plants are supplied to the roots in the form of solution that can be either in the form of static or flowing. Hydroponics can be cultivated both in green house and glass house environment. The limitation in green house environment is to maintain the temperature, pressure, humidity value at a particular level. In addition to that, monitoring and maintaining PH value is another challenge. Manual monitoring is in practice which is a very trivial task else the plants may die out. This project, focuses on two tasks, the first one is to automate the greenhouse environment monitoring, and the subsequent one is automation of PH level, temperature, turbidity and electrical conductivity maintenance. In order to do so a sensor network is established. IOT is used to transfer the retrieved data to the internet (mass storage) and mobile app is developed to communicate the current status. So that monitoring & maintenance will be easier. This Hydroponics system requires less manual intervention, water and space than traditional agricultural systemsItem Design of a power efficient 16-bit posit Multiplier and its application in Image contrasting(NHCE, 2025) Vishwass R 1NH21EC188; Yashasvi linga Reddy 1NH21EC190; Srinivas Abhinay Gandla 1NH21EE112The ever-increasing demand for energy-efficient computing has driven the need for innovative approaches to arithmetic logic design. Traditional floating-point arithmetic units, including multipliers and Multiply-Accumulate (MAC) units, have long been essential for applications ranging from artificial intelligence and machine learning to image processing and scientific computation. However, these units are inherently power-hungry and require significant silicon area, which limits their suitability for modern systems that prioritize energy efficiency. Addressing these challenges, this project focuses on the design and implementation of a 16-bit posit multiplier. The posit number system, a promising alternative to IEEE-754 floating-point representation, offers a highly flexible structure that dynamically adjusts precision based on input data. This characteristic allows the posit architecture to achieve a superior trade-off between computational accuracy, power efficiency, and hardware area, making it an ideal candidate for next generation arithmetic designs. The 16-bit posit multiplier designed in this project incorporates advanced optimization techniques to enhance performance. The radix-4 Booth algorithm is utilized to streamline the multiplication process by reducing the number of partial products, thereby minimizing switching activity and power consumption. Additionally, a selective activation mechanism is implemented, which dynamically limits the number of output bits based on the precision required for the given computation. This not only reduces unnecessary hardware utilization but also further decreases dynamic power consumption. The design was modelled and simulated using Verilog HDL in the Xilinx Vivado environment, where functionality was verified, and post-synthesis results were obtained. These optimizations demonstrate the potential of posit arithmetic in achieving superior performance with reduced power and area requirements compared to traditional floating-point systems.Item Smart gloves that speak(NHCE, 2025) Bhuvan Singh Rajput 1NH21EE024; Aneelkumar Madarkhandi 1NH21EE015; Nishanth Valliyappan K 1NH21EC111; Naresh Kumar 1NH21EC106In daily encounters, effective communication is essential, but those with speech and hearing impairments have several difficulties. To solve this, we created Smart Gloves that Speak, a creative and reasonably priced way to enable smooth communication. These gloves help people with communication challenges by translating basic hand motions into visual and aural outputs. Our system combines an Arduino Uno microcontroller, an LCD screen, a speaker, and inexpensive components like aluminium foil to create circuits. The way the technology operates is by using the touch of fingertips implanted in metal to detect motions. When a circuit is finished, it activates preset commands that are shown on the LCD and uses a voice output module to translate these motions into speech. The solution keeps top functionality while cutting production costs to only ₹1,000 by doing away with the requirement for pricey sensors like flex sensors. With up to 1,024 unique combinations that correspond to various programmed messages, the gloves can understand a wide variety of motions. This enables a broad range of use cases and easy modification. People are able to utilize the gloves efficiently in their daily lives because of its ergonomic and lightweight design. Furthermore, the use of gesture recognition technology facilitates wireless connectivity and improves real-time communication. This initiative offers a significant way to encourage accessibility and inclusion. We show how technology can empower and change lives by giving them an affordable and user-friendly communication tool. A viable strategy for removing obstacles and encouraging more social integration is provided by the Smart Gloves that Speak.Item Iot based accurate weather forecast at micro level(NHCE, 2025) Deepak H 1NH22EC403; Mahesh L Herakal 1NH22EC407; Pavan R 1NH22CE403; Praveen B 1NH22CE404The development of an Internet of Things (IoT)-based weather monitoring and forecasting system is presented in this study. The system integrates multiple sensors, such as the DHT11 for temperature and humidity, LDR module for Light Intensity, BMP280 for atmospheric pressure, and rain sensors, with an ESP8266 microcontroller. Solar power integration ensures sustainability and energy efficiency, making the system suitable for deployment in both rural and urban areas. Weather forecasting is a crucial component of environmental monitoring that supports informed decision-making in a variety of sectors, including agriculture, urban planning, and disaster management. The system analyzes past data trends to produce weather predictions for the next five hours using machine learning techniques, namely the Random Forest model used in Google Colab. The effectiveness, accuracy, and viability of the suggested IoT-based system are assessed by contrasting it with conventional machine learning models used in weather forecasting. The IoT-integrated strategy bridges the gap between environmental sensing and predictive analytics by providing real-time data collecting and processing, whereas standalone ML models mostly rely on vast amounts of historical data and computational resources to provide reliable predictions. This integration guarantees that the forecasts are updated constantly as new data becomes available and improves forecast precision. Key findings show that the IoT-based system offers a sustainable, scalable, and affordable solution for micro-level weather forecasting, tackling important issues including energy dependence, data accessibility, and adaptation to different geographic situations. While the modular design makes it simple to replace with more sensors or cutting-edge components, the use of renewable energy sources, such as solar panels, guarantees continuous operation in distant areas. The comparative analysis emphasizes the benefits of integrating machine learning with IoT over using separate ML models. Although machine learning models are excellent at evaluating large datasets and spotting intricate patterns, their high computing costs and dependence on previous data restrict their use in real-time. By combining real-time sensor data with predictive algorithms, the Internet of Things-based system gets over these restrictions and produces forecasts that are more precise and context-specific. Additionally, the cloud-based design of the system makes it easier to retrieve data and predictions remotely, improving scalability and usability for a variety of applications. According to the study's findings, IoT-based weather forecasting systems enable data-driven decision-making at the micro level and constitute a paradigm change in environmental monitoring. Further developments, such the addition of edge computing for localized processing and sophisticated neural network models for enhanced predictive capabilities, are made possible by the combination of IoT and machine learning technology. Future studies might investigate these possibilities to improve system performance and broaden its scope of use, supporting international initiatives to address climate-related issues.Item IOT-Based Smart Bin Monitoring System(NHCE, 2025) Ishita dayal 1NH21EC071; Vaibhav Kumar Shah 1NH21EC169; Krishna Kunal 1NH21IS081; Mannat Atwal1NH21IS091The rapid urbanization and population growth in cities have intensified the challenges associated with waste management. Traditional methods of waste collection, which rely on fixed schedules and routes, often result in inefficiencies such as overflowing bins, underutilized collection resources, and increased operational costs. These limitations demand innovative solutions to ensure efficient waste collection, segregation, and monitoring. The Smart Garbage Dustbin System offers an automated and technologically advanced solution for efficient waste management. This system emphasizes three key areas: waste segregation, energy estimation, and real-time monitoring. By utilizing a network of sensors—including moisture sensors to detect wet waste, gas sensors to identify organic decomposition, and ultrasonic sensors to measure bin fill levels—the system ensures accurate and effective waste classification. The integration of microcontrollers such as Arduino and communication modules like ESP8266, GPS, and GSM enhances its capability to function seamlessly. The system segregates waste into dry and wet categories using the input from moisture and gas sensors, enabling more effective recycling and disposal. Real-time monitoring is achieved through ultrasonic sensors that continuously measure the fill level of garbage bins, reducing the chances of overflow and improving collection efficiency. Data collected from the sensors is transmitted to a central server via the ESP8266 module, ensuring secure and real-time updates. The central server processes the data and displays it on a user-friendly web interface, allowing users and waste management authorities to monitor the bin’s location, fill level, and overall status in real-time. The integration of GPS enables precise location tracking, while GSM ensures reliable data transmission even in remote areas. This paper provides a comprehensive analysis of the Smart Garbage Dustbin System, detailing its system design, hardware-software integration, and application in modern smart waste management. The study also explores the scalability of the system for large urban areas and highlights its potential for reducing environmental impact, enhancing operational efficiency, and promoting sustainable practices in waste management. IoT based Smart Bin Monitoring System.Item Robotic arm development for prosthetics using 3D Printing(NHCE, 2025) Varsha M 1NH21EC173; Moksha S M 1NH21EC096; Harshith R Pattankodi 1NH21ME023; Mohammad Irfan 1NH21ME045Prosthetic devices have the power to change lives, offering individuals with limb loss the chance to regain their movements and perform daily activities with ease. However, many current options are either too expensive, difficult to use, or lack the natural responsiveness that users need to feel truly connected to their prosthetics. This project tackles these challenges by designing a 3D-printed robotic arm that mimics human arm movements in real time, controlled by a wireless glove. The glove, equipped with flex sensors, captures the subtle movements of the user’s hand and fingers and sends these signals wirelessly to the robotic arm. The arm then mirrors these movements seamlessly, creating a natural experience. By using accessible components like Arduino boards, flex sensors, and breadboards, along with cost-effective 3D printing technology, we’ve built a system that is both functional and affordable. This project is designed with a focus on prosthetic applications, aiming to make advanced prosthetic technology more accessible and user-friendly. Our goal is to create a solution that not only restores movement but also empowers individuals by giving them a device that feels like a true extension of themselves. Through this work, we hope to bridge the gap between complex, high-cost prosthetics and simple, affordable solutions, making a meaningful impact on the lives of those who rely on them.Item Real time speech to braille conversion tablet for visually impaired(NHCE, 2025) Riti S Devatha 1NH21EC079; M Chandana 1NH21EC084; K Rakesh 1NH21ME031; M Sanjay Balaji 1NH21ME036The Real-Time Speech-to-Braille Conversion Tablet is an innovative assistive device designed to bridge the accessibility gap for individuals with dual sensory impairments— visual and hearing disabilities. Recognizing the challenges faced by these individuals, including limited access to Braille resources, high costs of existing technologies, and barriers to real-time information, this project aims to deliver a cost-effective, portable, and functional solution. The tablet converts spoken English into tactile Braille characters, allowing users to access real-time auditory input through touch. The core of the system features a dynamic Braille cell operated by solenoid actuators controlled by an ESP32 microcontroller. Speech input is captured via a microphone, processed through Python-based speech-to-text algorithms, and subsequently converted into Braille representations. The system updates the single Braille cell every five seconds, displaying successive characters in a sequence. This design ensures portability and cost- efficiency while maintaining functionality. Key objectives of the project include addressing the high cost of traditional Braille systems by using affordable components like solenoid linear motors, enabling real-time speech-to-text conversion, and enhancing accessibility to educational and communication resources. The device fosters independence and reduces reliance on caregivers by offering a reliable, user-friendly interface tailored for visually and hearing-impaired individuals. Additional features include a scalable design, with potential future expansions to support multi- language speech recognition, multi-cell Braille displays for faster reading, and integration with other assistive technologies. By broadening access to information, the Speech-to-Braille Conversion Tablet is poised to transform the learning and communication experiences of visually and hearing-impaired individuals. Its innovative approach to affordability, portability, and real-time functionality establishes a new benchmark for assistive technology in this domain.Item Smart Player Performance Analysis(NHCE, 2025) Athul Satheesh 1NH21EC025; Chirag V 1NH21EC037; Ashlesh Ujjwal Tandulkar 1NH21IS031; Faizan Abdul Azeez 1NH21IS055The project “Smart Player Performance Analysis”, which is themed under ‘Digital Identity and Learning Analytics’ focusses on Smart Player Performance Analysis provides an innovative platform that seamlessly blends professional football analytics with interactive user engagement. At its core, the platform offers an extensive database of football teams and their players, showcasing detailed personal performance metrics such as speed, strength, dribbling, and heading abilities. By providing this repository of real-world player statistics, it enables users to explore the intricacies of professional football analysis while serving as a benchmark for evaluating their own performance. This platform goes beyond traditional analytics tools by empowering users—whether aspiring athletes, enthusiasts, or fans—to assess their skills and compare them with those of professional players. Users can manually input their data or leverage real-time performance metrics collected through IoT sensors. The analysis process is streamlined via cloud integration, utilizing platforms like ThingSpeak to ensure efficient data handling. By comparing individual performance with professional benchmarks, the platform provides personalized recommendations for skill improvement and suggests the most suitable playing positions, thereby offering an enriched football experience. One of the unique aspects of this project is its focus on accessibility and engagement. Unlike existing tools such as Wyscout, which are often complex and require subscriptions, this platform democratizes football analytics, making professional-grade insights accessible to everyone. Through its intuitive Streamlit-based interface, it fosters a user-friendly environment that simplifies the process of analyzing and visualizing data. Features such as the "Your Analysis" tool enhance user interaction by predicting potential, identifying areas of improvement, and fostering a sense of community among users. The project aims to go beyond technical functionality, seeking to create an engaging platform that encourages fan participation and skill development. By bringing professional football analytics to a broader audience, it bridges the gap between grassroots and eliteItem Earthquake Prediction and Alerting System Using Machine Learning and Raspberry Pi(NHCE, 2025) Likhith kumar N 1NH22EC405 Santhosh 1NH22EC411 Karishma Palekar 1NH22CS405 Mohan Kumar S 1NH22CS412This report presents a novel method for earthquake warning and prediction that makes use of machine learning and inexpensive hardware integration with a Raspberry Pi. With the help of an MPU6050 accelerometer, vibration sensor, and BMP180 sensor, the suggested system may continuously monitor environmental and seismic parameters like temperature, pressure, altitude, acceleration, and vibrations. Magnitude, Peak Ground Acceleration (PGA), and Root Mean Square (RMS), three crucial seismic characteristics, are obtained from the constant stream of data produced by these sensors. A rich and varied dataset is created using the gathered data, which serves as the basis for training and assessing different machine learning algorithms. To determine which model has the best accuracy for predicting earthquakes, the system thoroughly evaluates various models. After that, the chosen model is installed on a Raspberry Pi, which allows for real-time prediction using an integrated Flask server. In addition to processing real- time data, this server helps the system run independently and deliver alerts on time. The system's connection with Twilio's messaging API, which permits the automatic sending of SMS messages depending on observed seismic activity and earthquake magnitude, is one of its most notable features. The purpose of these alerts is to give users advance notice so they can take preventative measures like leaving the area or finding cover. The system demonstrates the possibility of developing an affordable and scalable earthquake prediction and alerting system by fusing machine learning with Internet of Things technology. Future improvements, such adding more sensor types or extending deployment to other areas, are made possible by its modular architecture. This study emphasizes how important technology is in reducing the dangers of natural catastrophes and how it may improve public safety, especially in areas where earthquakes are common. In order to minimize the loss of life and property in earthquake-prone locations, this system is an essential step in closing the gap between scientific discoveries and real-world disaster preparedness implementationsItem Pothole detection System(NHCE, 2025) Shruti Shukla 1NH20EC191; Haripriya P 1NH21EC064; Ketan Mane 1NH21IS089; Krishna Goyal 1NH21IS210In today's fast-paced world, finding relevant and personalized travel routes, points of interest (POIs), and recommendations that cater to individual preferences and needs is a significant challenge. Traditional map applications provide generic suggestions, often leading to suboptimal travel experiences, wasted time, and missed opportunities for discovering valuable and interesting locations. To address this issue, we propose the development of a personalized map recommendation system leveraging artificial intelligence (AI) and machine learning (ML) technologies. The system aims to analyze user behavior, preferences, and contextual data to provide tailored recommendations for routes, attractions, dining options, and other POIs. By incorporating advanced ML algorithms and real-time data analysis, the solution enhances the user's travel experience, making it more efficient, enjoyable, and personalized. Key features include adaptive route optimization based on traffic conditions, weather, and time of day, as well as proactive suggestions for stops based on user habits and preferences. Our model achieved a high recommendation accuracy of 90.3%, with precision and recall rates of 88.7% and 89.2%, respectively. User studies indicated a 92.5% satisfaction rate, highlighting the system's effectiveness in improving travel experiences and uncovering new places. The contextual adaptation feature significantly improved user satisfaction during adverse conditions, such as suggesting indoor activities on rainy days. Challenges include ensuring data privacy and security, handling diverse user preferences, and maintaining scalability as the user base grows. Future work will focus on enhancing the model's adaptability, integrating with existing navigation systems, and expanding the dataset to include more diverse road conditions and user behaviors. The personalized map recommendation system not only revolutionizes how individuals navigate and explore but also opens new possibilities for creating smarter, more responsive applications in various domains, including retail, healthcare, and smart cities.Item RealTalk(NHCE, 2025) Cedric Baby 1NH21EC033; Satvik Anguluri 1NH21EC142; Ravin Raina 1NH21AI084; Uday Singh 1NH21AI110Language is one of the most fundamental ways humans communicate, but linguistic diversity often poses challenges in understanding and accessibility. This project aims to bridge these gaps by developing a robust multilingual translation system powered by state-of-the-art Marian MT models from Hugging Face's transformers library. The system supports translations from English into 13 diverse languages: Hindi, Tamil, Malayalam, Kannada, Telugu, Gujarati, Bengali, French, German, Russian, Japanese, Chinese, and Lebanese. By combining pretrained machine translation models with an efficient and scalable implementation, the system ensures translations are not only linguistically accurate but also contextually relevant, capturing cultural nuances wherever possible. The modular design enables straightforward integration of additional languages, while its flexibility accommodates varying linguistic complexities and syntactical structures. This capability positions the system as a versatile tool for education, healthcare, business, and cultural preservation. A major innovation of this project lies in its ability to address the limitations of pretrained models for underrepresented languages like Lebanese by setting a foundation for custom model training. The project explores solutions to technical challenges such as memory optimization, model compatibility, and accurate handling of idiomatic expressions. The inclusion of future-facing goals, such as real-time translation and enhanced contextual understanding, underscores its scalability and relevance in an increasingly globalized world.