Volume 14 | Issue 6 |

 Abstract

Recent advances in deep learning--particularly Convolutional Neural Networks (CNNs)--have opened transformative opportunities for automated object detection and classification in complex outdoor environments. Models from the YOLO (You Only Look Once) family, especially YOLOv8, have emerged as highly effective tools for real-time multi-class detection, offering an optimal balance between inference speed and detection accuracy. Unlike rule-based or template-matching methods, CNN-based architectures learn intricate visual features directly from data, enabling them to adapt to varying lighting conditions, occlusion from vegetation, and complex natural backgrounds [2]. This paper proposes a complete smart farm surveillance framework that combines YOLOv8-based object detection with automated Telegram Bot API notifications. The system processes live video feeds from farm cameras, identifies threats as they occur, captures annotated snapshots, and delivers instant alerts to farmers' smartphones. Key contributions include: o A fine-tuned YOLOv8 detection model trained on a diverse agricultural dataset covering humans, birds, and animals. o An integrated Telegram Bot alerting module providing real-time notifications with timestamped annotated images. o Deployment on resource-constrained edge devices (Raspberry Pi / NVIDIA Jetson Nano) demonstrating practical feasibility. o Comprehensive experimental evaluation under varied farm conditions confirming 94.7% precision and 93.1% recall.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

CNN, YOLOv8, Smart Farming, Object Detection, Telegram Alerts, Agricultural Surveillance, Real-Time Monitoring, Edge Computing, IoT, Deep Learning

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Microfinance has become an important tool for boosting rural economic development by offering financial services to low-income households often left out of the formal banking system. Through microcredit, savings, insurance, and financial education programs, microfinance helps poor households engage in productive activities, create jobs, and improve their living conditions. In India, the Self-Help Group (SHG)-Bank Linkage Program and Microfinance Institutions (MFIs) have greatly increased financial inclusion in rural areas. As of March 2024, over 144.22 lakh Self-Help Groups were connected to banks, with savings exceeding INR65,089 crore. More than 83 percent of these groups were made up of women, highlighting the significant role of microfinance in empowering women. This paper looks at how microfinance contributes to rural economic development, with a focus on reducing poverty, generating employment, developing agriculture, promoting entrepreneurship, and enhancing social empowerment. The study also addresses key challenges in the sector and offers policy recommendations for improving its effectiveness.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Microfinance, Rural Development, Poverty Alleviation, Financial Inclusion, Self-Help Groups, Entrepreneurship, Women's Empowerment

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Heart attack remains one of the leading causes of mortality worldwide, creating a critical need for early prediction and preventive healthcare strategies. HeartIntel: A Machine Learning Framework for Early Heart Attack Risk Prediction and Clinical Decision Support presents an intelligent system designed to identify individuals who may be at elevated risk of heart attack using clinical and physiological parameters. The framework utilizes patient information such as age, blood pressure, cholesterol level, chest pain characteristics, maximum heart rate, ST depression, and related health indicators for predictive analysis. Data preprocessing, feature transformation, and normalization techniques are applied to improve data quality and model performance. Machine learning algorithms including Logistic Regression, Random Forest, and Support Vector Machine are employed for classification and risk assessment. A web based interface enables efficient data entry, prediction generation, and result visualization. Experimental evaluation demonstrates reliable predictive capability, satisfactory accuracy, and timely identification of high risk cases. The system supports clinical decision making, promotes early intervention, enhances healthcare efficiency, reduces diagnostic delays, and contributes to improved patient safety and outcomes.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Heart Attack Risk Prediction, HeartIntel, Machine Learning, Clinical Decision Support, Logistic Regression, Random Forest, Support Vector Machine, Healthcare Analytics, Predictive Modeling, Risk Assessment.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Electrical poles are critical components of power distribution networks, yet faults such as current leakage, conductor breakage, and pole electrification can pose serious risks to public safety and infrastructure reliability. This paper presents ElectroSentrix, an IoT-based intelligent electrical pole fault detection and accident prevention framework designed to monitor pole conditions continuously and respond to hazardous situations in real time. The proposed system integrates voltage sensors, current sensors, an ATmega328 microcontroller, a NodeMCU communication module, an LCD display, and a voice alert unit to detect abnormal electrical conditions and issue timely warnings. Upon fault identification, the framework generates local audio notifications, displays fault information, and transmits alerts to authorized personnel through an IoT platform for rapid maintenance action. Automated monitoring minimizes manual inspection efforts and enhances operational efficiency while reducing the likelihood of electrocution incidents. Experimental implementation demonstrates reliable fault detection, prompt alert generation, and effective communication capabilities. The developed framework offers a cost-effective, scalable, and practical solution for improving electrical pole safety, ensuring public protection, and supporting smarter power distribution infrastructure.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Electrical Pole Safety, Internet of Things (IoT), Fault Detection, Accident Prevention, Smart Monitoring, NodeMCU, ATmega328, Voltage Sensor, Current Sensor, Public Safety.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Underground Cable Fault Detection, Arduino Uno, GSM Module, GPS Tracking, Fault Distance Locator, Power Distribution System, Smart Monitoring, Fault Localization.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

: Road traffic accidents remain one of the major causes of fatalities, injuries, and economic losses worldwide. This project aims to reduce the increasing number of road accidents by developing an AI-powered road safety system that proactively identifies accident-prone zones and provides real-time alerts to drivers. The proposed system utilizes five years of historical accident data (2021-2025) collected across twelve accident-related parameters, including fatalities, grievous injuries, accident intensity, and yearly accident trends. The accident records are mapped to precise geographical coordinates across Pune, Maharashtra, enabling spatial analysis and hotspot identification.The system integrates machine learning and geospatial technologies to improve road safety. K-Means clustering is employed to identify and visualize accident hotspots, achieving an optimal Silhouette Score of 0.5925 with three distinct clusters. Furthermore, an XGBoost classification model is trained using historical accident data from 2021-2023 to predict future accident-prone zones for 2024-2025. Experimental evaluation demonstrates strong predictive performance, achieving an accuracy of 92.08%, precision of 91.84%, recall of 100%, F1-score of 95.74%, and ROC-AUC of 97.78%.To enhance user awareness and safety, the identified accident-prone zones are displayed on an interactive Leaflet.js map as color-coded hotspots. Drivers receive visual alerts and audio notifications through the Web Speech API whenever they approach within a predefined distance of a hazardous zone. The system is further supported by a FastAPI backend and GPS-based location tracking for real-time operation. By combining accident hotspot detection, predictive analytics, and proactive driver alerts, the proposed framework contributes toward safer transportation networks, reduced accident risks, and improved decision-making for both drivers and traffic management authorities.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Accident-Prone Zone Identification, Road Safety, K-Means Clustering, XGBoost, Leaflet.js, Machine Learning, FastAPI, GPS Tracking, Real-Time Alerts.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Attendance management is an essential activity in educational institutions and organizations, where accurate record keeping directly influences administration, monitoring, and performance evaluation. Traditional attendance methods such as manual registers and card-based systems are often time consuming, error prone, and vulnerable to proxy attendance. This paper presents FaceTrack: An Intelligent Vision-Based Attendance Monitoring Framework Using Deep Facial Analytics, a smart attendance solution that automates the identification and recording process through facial recognition technology. The framework utilizes OpenCV for real-time face detection, dlib-based facial feature encoding for identity matching, and a Flask-powered web interface for attendance visualization and management. Captured facial images are processed and compared with stored facial encodings to accurately identify registered individuals and automatically generate attendance records with timestamps. Attendance information is maintained in both CSV files and a MySQL database to ensure reliable storage and easy retrieval. Experimental evaluation demonstrates high recognition accuracy, reduced processing time, and effective prevention of proxy attendance under standard indoor conditions. The proposed framework enhances operational efficiency, minimizes administrative workload, and provides a scalable, contactless, and cost-effective attendance monitoring solution for modern academic environments.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Attendance Management, Face Recognition, Computer Vision, OpenCV, Deep Facial Analytics, Flask, MySQL, Biometric Authentication, Automated Monitoring, Machine Learning.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Breastfeeding is widely recognized as the optimal method of infant feeding, offering significant nutritional, immunological, developmental, and health benefits for both infants and mothers. Despite its advantages, breastfeeding success is often affected by challenges related to positioning, attachment, and maternal comfort, which can lead to poor milk transfer, breastfeeding complications, and early cessation. This literature review aimed to examine and synthesize existing evidence on various breastfeeding methods and their effectiveness in promoting successful breastfeeding outcomes.A narrative literature review design was adopted. Relevant studies were retrieved from electronic databases, including pubmed, google scholar, cochrane Library, and Ssciencedirect, along with guidelines from the world health organization and United Nations Children's Fund. Peer-reviewed articles, systematic reviews, meta-analyses, randomized controlled trials, observational studies, and clinical guidelines published in English were included. The literature was analyzed thematically, focusing on different breastfeeding positions, factors influencing position selection, and their impact on maternal and infant outcomes.Findings indicate that breastfeeding success is strongly influenced by proper positioning and effective attachment, which facilitate efficient milk transfer, reduce complications, and enhance maternal comfort. Evidence suggests that no single breastfeeding position is universally superior; rather, effectiveness depends on maternal anatomy, infant developmental stage, mode of delivery, and clinical conditions. Positions such as the cross-cradle and football hold are particularly beneficial during early breastfeeding initiation, while side-lying and laid-back breastfeeding improve maternal comfort and recovery. Upright and specialized positions are useful in selected clinical situations, including reflux or prematurity. The review also highlights the essential role of nurses, midwives, and lactation consultants in providing breastfeeding education and individualized support.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

breastfeeding methods, breastfeeding positions, infant attachment, maternal comfort, breastfeeding support, lactation, nursing practice.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Allelopathy is a biological phenomenon in which plants release chemicals that affect the germination and growth of nearby plants. This study was conducted to evaluate the allelopathic effects of aqueous leaf extracts of Parthenium hysterophorus, Lantana camara, Ageratum conyzoides, and Azadirachta indica on seed germination and early seedling growth of Triticum aestivum (wheat). Leaf samples were shade-dried, powdered, and used to prepare extracts of 5%, 10%, and 15% concentrations. The experiment was carried out under controlled laboratory conditions using a completely randomized design with proper control and replications. In this study, parameters such as germination percentage, root growth, shoot growth, and seedling vigour index were evaluated. The results showed that all plant extracts had inhibitory effects, which increased with higher concentrations. Among them, Ageratum conyzoides exhibited the strongest inhibition on both root and shoot growth, followed by Lantana camara. Parthenium hysterophorus showed a moderate inhibitory effect, while Azadirachta indica showed no significant effect, indicating minimal allelopathic influence. Overall, the study demonstrates that allelopathic plants can negatively affect crop establishment and productivity. Therefore, proper management of these plants is necessary to improve crop growth and yield.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Allelopathy, seed germination, seedling growth, inhibitory effects

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The psychological well-being of prospective teachers plays a vital role in their academic success, professional preparation, and overall development. In recent years, increasing academic demands and changing educational environments have highlighted the importance of understanding the relationship between psychological well-being and academic engagement among teacher education students. The present study investigates the level of psychological well-being and academic engagement among prospective teachers and examines the relationship between these variables. A normative survey method was employed for the study. The sample consisted of prospective teachers enrolled in teacher education institutions. Standardized tools were used to measure psychological well-being and academic engagement. Descriptive statistics, t-test, ANOVA, and Pearson's Product Moment Correlation were utilized for data analysis. The findings revealed that prospective teachers exhibited moderate to high levels of psychological well-being and academic engagement. A significant positive relationship was found between psychological well-being and academic engagement, indicating that students with higher psychological well-being tend to demonstrate greater involvement, enthusiasm, and commitment to their academic activities. The study emphasizes the need for teacher education institutions to promote psychological well-being through supportive learning environments, counselling services, and well-being-focused interventions. Enhancing psychological well-being may contribute significantly to improving academic engagement and preparing competent future teachers.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Psychological Well-Being, Academic Engagement, Prospective Teachers, Teacher Education, Educational Psychology.

  License

Creative Commons Attribution 4.0 and The Open Definition