Proceedings of the 7th International Conference on Communications and Cyber Physical Engineering

Table of Contents

1. Navigating Complexity: The Role of Tableau in Driving Data-Driven Decision Making

   Kesavulu Poola, Pavakumari, J. Anil Kumar, Akkyam Vani

   This chapter delves into the pivotal role of Tableau in driving data-driven decision-making, emphasizing its user-friendly interface and powerful visualization capabilities. It explores how Tableau enables organizations to leverage big data for competitive advantage, with a focus on its benefits such as quick data processing, interactive dashboards, and storytelling features. The text also examines the impact of Tableau on organizational performance, highlighting improvements in data accuracy, decision-making speed, and overall business outcomes. Through case studies and practical examples, it demonstrates how Tableau has democratized data analytics, making it accessible to a broader audience. Additionally, the chapter provides a detailed methodology for using Tableau, including steps for importing and cleaning data, and creating various types of visualizations. It also presents an in-depth analysis of data science salaries across different countries, roles, and expertise levels, offering valuable insights into the data science job market. The conclusion underscores Tableau's transformative potential in fostering innovation and agility in today's competitive landscape.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

2. IoT Based Solar Powered Multipurpose Autonomous Robot for Effective Farming: A Prototype

   Morampudi Rajitha, Ch. Shravani, P. Rajesh Kumar, A. Keerthi, M. Navya, D. Meghana

   This chapter delves into the development and application of an IoT-based, solar-powered multipurpose autonomous robot designed to enhance farming efficiency. The robot, controlled via an Arduino ATmega328 and powered by solar energy, is equipped to perform essential agricultural tasks such as seeding, plowing, and watering. The text provides a detailed overview of the robot's components, including the solar panel, DC motors, WiFi module, and L293D motor driver, and explains how these elements work together to automate farming processes. It also discusses the challenges faced by modern agriculture, such as climate change, water scarcity, and soil degradation, and how this robot can help address these issues. The chapter includes practical results from prototype testing, demonstrating the robot's ability to accurately sow seeds and manage water distribution. The conclusion highlights the potential of this technology to reduce labor and time requirements, ultimately improving productivity and sustainability in agriculture.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

3. Revolutionizing Healthcare Logistics: Elevating Drug Accountability Through GIS-Enabled Tracing

   Padmavati E. Gundgurti, Poornima Gottimukkala, Nitisha Thallapally, Vaanisha Praveen Mahendrakar

   This chapter delves into the critical challenges faced by pharmaceutical supply chains, including the proliferation of counterfeit drugs and the lack of transparency. It highlights the limitations of conventional tracing methods and introduces Geographic Information Systems (GIS) and blockchain technology as innovative solutions. The text explores the benefits of integrating GIS-enabled tracing with blockchain, such as real-time visibility, immutable records, and enhanced regulatory compliance. It also discusses the technical complexities and challenges associated with this integration. The chapter presents a proposed system architecture using Ethereum blockchain and smart contracts, detailing the roles of various stakeholders and the flow of drug supply chain processes. Additionally, it showcases results from a simulation of vehicle tracking using GIS technology. The conclusion emphasizes the need for further research to enhance the efficacy of the proposed approach, underscoring its potential to combat counterfeit drugs and improve transparency in pharmaceutical supply chains.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

4. Enhancing Image Captioning Using Deep Learning

   Sireesha Vikkurty, Nagaratna P. Hegde, Abhijith Koppula, Yagnan Reddy Nimma

   This chapter delves into the intersection of computer vision and natural language processing, focusing on the advancements in image captioning through deep learning techniques. The study introduces an attention-based Encoder-Decoder architecture that combines convolutional features from pre-trained ImageNet models with object features from the Flick8K dataset. The research methodology involves a detailed process of image feature extraction, text preprocessing, and model training using CNN and LSTM networks. The proposed architecture is evaluated using the BLEU score, with ResNet outperforming VGG16 and Xception in generating accurate and contextually appropriate captions. The chapter also discusses future work, including the potential for fine-tuning CNN models and incorporating an attention module to further enhance performance. This comprehensive exploration provides valuable insights into the current state and future directions of image captioning technology.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

5. Enhanced Road Safety Through Video-Based Helmet Violation Detection

   Susmith Reddy Duggimpudi, Rohan Reddy Solipuram, Mukesh Kottur, Saroja Kumar Rout, Nilamadhab Mishra, S. Eswar Reddy

   This chapter explores the implementation of a video-based helmet violation detection system to improve road safety, particularly in regions with high motorcycle traffic. The system utilizes Faster R-CNN, a deep learning model, to accurately detect helmet violations in real-time surveillance footage. The article delves into the methodology, including data preprocessing, model training, and performance evaluation, demonstrating the system's effectiveness in challenging environments. It also compares various computational vision systems, emphasizing the superiority of the proposed solution. The results indicate promising accuracy and reliability, underscoring the potential of automated helmet detection in reducing motorcycle accidents and promoting compliance with safety regulations.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

6. GreenWatch: Revolutionizing Farming with Plant Disease Detection by CNN

   Aadithya Vikram Budarapu, John Hyde Gaddam, Ram Boyedi, Thoshan Kumar Muthyala, Sunanda Yadla

   This chapter explores the critical role of early disease detection in improving agricultural productivity, particularly in India where farming is a significant economic sector. It delves into various deep learning models, including InceptionV3, ResNet50, and ResNet152v2, which are trained on extensive plant image datasets to identify diseases with high accuracy. The ensemble model, combining these architectures, achieves an impressive accuracy of 98.29%, revolutionizing crop protection. The project also emphasizes the importance of user-friendly technology, bridging the gap between complex deep learning and everyday plant care. By uploading an image, users receive diagnoses and treatment recommendations, empowering them to care for their plants effectively. The chapter discusses the strategic intent behind the technology choices and their collective impact on accessibility and user empowerment, ultimately contributing to sustainable agricultural practices and global food security.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

7. Modelling of Li-Ion Battery Pack and Simulation of BMS

   M. Chakravarthy, Dheeraj Gurijala, Bhargavi Nagavarapu, Sannihith Peruka

   This chapter delves into the modeling of Li-ion battery packs using Thevenin's equivalent circuit model, which effectively captures the dynamic and static characteristics of batteries. The study utilizes a high-capacity Lithium-ion battery pack with specific nominal capacity, voltage, and current ratings, critical for accurate performance and safety assessments. The pulse discharge method is employed to extract key parameters such as internal ohmic resistance and polarization resistance and capacitance, essential for battery simulation. The chapter also focuses on the simulation of Battery Management Systems (BMS), highlighting its core functions: status monitoring, state estimation, and optimization management. A detailed control circuit using universal logic gates is developed to manage charge and discharge processes, ensuring safe and efficient battery operation. The relationship between state of charge (SoC) and open circuit voltage (OCV) is explored, with various techniques applied for accurate SoC estimation. The chapter concludes with the successful implementation of the Thevenin model and the BMS control circuit, providing a versatile and cost-effective solution for battery management.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

8. Performance Analysis of Effective Battery Management System for Electric Vehicles

   M. T. L. Gayatri, Odela Sathwika

   This chapter delves into the performance analysis of an effective battery management system for electric vehicles, focusing on the integration of supercapacitors to enhance system efficiency and durability. The text explores the dynamic power management system, highlighting the role of supercapacitors in handling high charging and discharging current peaks, thereby extending the battery pack's longevity. It also discusses the use of a fuzzy logic controller for regulating DC link voltage and the implementation of a bidirectional DC-DC converter for optimal power flow. The chapter presents simulation results that demonstrate the system's performance under various conditions, including irradiance and temperature variations. Additionally, it covers the control algorithms used for maximum power point tracking (MPPT) and the overall system output. The conclusion emphasizes the necessity of incorporating supercapacitors in electric vehicle battery designs to meet peak power demands and improve system durability.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

9. Revolutionizing Rice Farming: A Hybrid Deep Learning Approach for Automated Detection of Plant Diseases from Leaf Images

   B. Sarada, Siva Sankar Namani, K. Thirupathi Rao

   This chapter explores the implementation of a hybrid deep learning system for the automated detection of rice plant diseases from leaf images. The study focuses on three major diseases: bacterial blight, brown spot, and leaf blast, which significantly impact rice yields. The system employs deep learning models such as DarkNet19, MobileNetV2, and ResNet18 for feature extraction, followed by traditional machine learning classifiers like SVM, KNN, and Ensemble methods for disease classification. The research highlights the importance of early disease detection in mitigating yield losses and improving crop management. The proposed method achieves high classification accuracy, demonstrating its potential as a valuable tool for farmers and agriculturalists. The study also discusses the future enhancements, including real-time disease identification and classification, to further improve the system's practicality and effectiveness.

   AI Generated

   This summary of the content was generated with the help of AI.

   Show original text

10. DDoS Detection Using ML Algorithm

    Ashok Kumar Nanda, S. Harshavardhan Reddy, S. Rahul, V. Karthikeshwar

    This chapter delves into the critical role of machine learning in detecting and mitigating Distributed Denial-of-Service (DDoS) attacks, a growing threat in the digital landscape. It explores various machine learning algorithms, including Random Forest, XGBoost, and K-Nearest Neighbors, and their effectiveness in identifying and classifying network traffic as benign or malicious. The text provides a detailed workflow of the machine learning process, from data preprocessing to model validation, highlighting the importance of feature engineering and data splitting. It also discusses the performance metrics used to evaluate the models, such as accuracy, precision, recall, and the F1 score. The chapter concludes with a comparison of the algorithms, showing that Random Forest achieved the highest accuracy of 98%, followed by Decision Tree at 96% and K-Nearest Neighbors at 95%. This comprehensive analysis underscores the potential of machine learning in enhancing network security and protecting against evolving cyber threats.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

11. A Review on Intelligent Exam Monitoring System Using Deep Learning

    T. Neha, T. Vennela, N. Ravithreni, A. Ch. S. Kanakadurga, T. Sneha

    This chapter delves into the critical issue of exam integrity and the role of deep learning in combating cheating. It explores various deep learning models and systems designed to detect and prevent cheating behaviors, such as copying, using unauthorized aids, and unauthorized communication. The review highlights the limitations of traditional monitoring methods and the advantages of deep learning techniques, which can analyze large datasets with high speed and accuracy. Key topics include the use of computer vision and deep learning for suspicious activity detection, the integration of facial recognition and eye tracking for automated proctoring, and the development of hybrid deep learning techniques for comprehensive cheating detection. The chapter also discusses the challenges and future directions in this field, emphasizing the need for continuous improvement to ensure academic integrity. The conclusion underscores the potential of deep learning to revolutionize exam monitoring and maintain the integrity of offline exams.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

12. Revolutionizing Web Security: The Efficiency of a Reusable CAPTCHA Security Engine

    Yamsani Vaishnavi, Touria Tanazzum, Hari Bharadwaj, S. Ranjithreddy, Kottu Santosh Kumar, Saroja Kumar Rout

    This chapter delves into the critical role of CAPTCHA technology in distinguishing human users from automated bots, focusing on its application in enhancing online security. It explores various CAPTCHA methods, including text-based and image-based CAPTCHAs, and their respective AI challenges. The text provides a detailed overview of the implementation process, highlighting the integration of CAPTCHA within a job application web application built with Flask. It also discusses the ethical considerations and privacy implications of CAPTCHA development. The chapter concludes with a discussion on the effectiveness of traditional CAPTCHA methods and the potential of future vision-based CAPTCHAs, offering insights into the ongoing efforts to balance security, usability, and privacy in online platforms.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

13. An Empirical Analysis of Indian Banks’ Spatial Efficiency Based on Performance and Production Approaches

    Putha Viswanatha Kumar, Vulichi Pavankumari, Venkataramana Musala, S. Damodharan

    This chaptere presents an empirical analysis of the spatial and intertemporal efficiency of Indian banks from 2005 to 2021. The study employs traditional and window-based Data Envelopment Analysis (DEA) methods to evaluate the efficiency of public, private, and foreign banks. Key topics include the use of six different efficiency types: Deposit Mobilization Efficiency (DME), Fund Conversion Efficiency (FCE), Off-Balance Sheet Activities Efficiency (OBAE), Cost Revenue Management Efficiency (CRME), Production Approach Efficiency (PAE), and Intermediate Approach Efficiency (IAE). The analysis reveals that public sector banks (PSBs) generally exhibit higher overall efficiency compared to private sector banks (PVTs) and foreign banks (FBs), with notable variations in specific efficiency types. The study also highlights the impact of ownership structure on bank efficiency, with PSBs showing more consistent performance due to unified government policies. Additionally, the chaptere provides detailed descriptive statistics and heatmaps for each bank type, offering a comprehensive overview of their efficiency scores. The results suggest that while some banks perform above average in certain efficiency types, there is still room for improvement, particularly in areas like production approach efficiency and cost revenue management. Overall, the study offers valuable insights into the efficiency dynamics of the Indian banking sector, with implications for policymakers, researchers, and industry leaders.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

14. Convolutional Neural Networks (CNN): Handwritten Digit Recognition

    Aluka Madhavi, Samala Nandini, Potlakayala Deepthi, Manchala Bhavani, Kasapaka RubenRaju, Bomma Reddy Sindhuja

    This chapter delves into the world of Convolutional Neural Networks (CNNs) and their exceptional ability to recognize handwritten digits. It begins by highlighting the importance of this task in various applications, such as digitizing historical data and automating postal mailings. The chapter then explores the architecture of CNNs, including their convolutional, pooling, and fully connected layers, and how these components work together to extract and classify features from raw pixel data. Preprocessing techniques, such as normalization, noise reduction, and contrast enhancement, are discussed to improve the quality of input data and enhance model performance. The chapter also compares CNNs with other machine learning algorithms, demonstrating their superior accuracy and efficiency. Furthermore, it reviews related work in the field, providing insights into various methodologies and their findings. The chapter concludes with a look at future research avenues, including novel network architectures, improved noise resistance, and enhanced model interpretability. By reading this chapter, professionals will gain a comprehensive understanding of CNNs for handwritten digit recognition and practical insights into implementing and improving these systems.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

15. Sentiment Analysis and Classification Using Convolutional Neural Networks

    Bomma Reddy Sindhuja, Aluka Madhavi, Samala Nandini, Potlakayala Deepthi, Manchala Bhavani, Kasapaka RubenRaju

    This chapter delves into the application of Convolutional Neural Networks (CNNs) for sentiment analysis and classification, a technique that has shown promising results in understanding public opinion, consumer feedback, and social media trends. The text covers the preprocessing steps essential for effective sentiment analysis, including text tokenization, word embeddings, and data augmentation. It also explores advanced techniques such as pre-trained word embeddings, attention mechanisms, and ensemble methods to enhance model performance. The chapter compares CNNs with other deep learning models like Recurrent Neural Networks (RNNs) and Long Short-Term Memory (LSTM) networks, highlighting the strengths and weaknesses of each. Additionally, it discusses the potential for future improvements, such as integrating multi-modal information and leveraging transfer learning techniques. The text concludes that CNNs, originally developed for image recognition, have been successfully adapted to process textual data, capturing both local and global semantic information for accurate sentiment classification.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

16. Deep Learning Based MPPT Control for 200 Watts PV System in Electrical Vehicles

    M. T. L. Gayatri, G. Laxmi Prasanna

    This chapter delves into the integration of deep learning techniques to enhance the performance of photovoltaic (PV) systems in electric vehicles. It focuses on a deep learning-based Maximum Power Point Tracking (MPPT) control system designed for a 200-watt PV system. The chapter covers the architecture of the neural network, which includes input, hidden, and output layers, and utilizes backpropagation for training. It also discusses the simulation and results of the MPPT integration, highlighting its ability to optimize power generation and efficiency under dynamic environmental conditions. The chapter concludes with a detailed analysis of the system's performance, demonstrating its effectiveness in improving energy harvesting and overall system reliability.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

17. High Gain DC-DC Converter with Dual Input for Low Power Applications

    B. Harshini, B. Naga Swetha, R. Geshma Kumari, K. Sravani, M. Naga Jyothi, O. Sobhana

    This chapter delves into the design and implementation of a high-gain DC-DC converter with dual input, specifically tailored for low power applications. The focus areas include the problem identification of low voltage outputs from renewable energy sources and the need for efficient conversion to higher voltages. The proposed solution involves a novel converter design that integrates a two-stage booster with a switched inductor construction, resulting in a high voltage gain and minimal switching losses. The chapter also explores the use of a voltage multiplier circuit and a dual input two-switch boost converter, comparing their performance with the proposed high-gain DC-DC converter. The results demonstrate significant improvements in voltage gain and efficiency, making the proposed converter a promising solution for energy storage and renewable energy applications. The detailed analysis and comparative study provide a comprehensive understanding of the advantages and potential applications of the proposed design.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text

18. Nonlinear Companding Scheme PAPR Reduction of OFDM Signals

    T. Y. Melligeri, Rajkumar L. Biradar

    This chapter delves into the challenges posed by high PAPR in OFDM signals and introduces a novel non-linear companding scheme to address this issue. The text explores the existing PAPR reduction methods, highlighting their limitations, and presents a detailed formulation of the proposed system. Simulation results demonstrate the superior performance of the new technique in terms of BER and PAPR, even without de-companding at the receiver. The chapter also discusses the practical implementation of the proposed method using a graphical user interface in MATLAB, showcasing its application in image and audio transmission. The findings suggest that the proposed companding system can deliver satisfactory BER performance without the need for de-companding, making it a promising solution for enhancing the efficiency and reliability of OFDM-based communication systems.

    AI Generated

    This summary of the content was generated with the help of AI.

    Show original text