Journal of Research in Computer Science and Engineering

Authors: Prof. K. V. Hulle, Shreyash Kudache, Aniket Awati, Saransh Patil

Abstract: In today's fast-paced market, predicting future profitability is crucial for making informed business decisions. By analyzing historical data on sales, costs, and market trends, machine learning techniques help uncover patterns that can forecast profit ability out comes accurately. This platform helps to preset textile products globally. It also aims to engage visitors through interactive features, facilitate online shopping, offer customer support, and highlight the company's commitment to sustainability the platform delivers a user-friendly and responsive interface that enables textile businesses to easily access.

Keywords: Textile, Business Profitability, Forecasting, Machine Learning, Advertisement, Products

Author: Rajesh Kumar Lohani

Abstract: In this paper, a comparative study on eight feature matching algorithms: SIFT, ORB, KAZE, AKAZE, Dense SIFT, DAISY, BRISK and FREAK were presented. Feature matching is an important aspect of computer vision used for key point detection and matching across im- ages to perform tasks such as object recognition. Each importantly has different features and performance metrics, making them suitable for cer- tain uses. I have evaluated these algorithms with two main criteria: the computation time, and the number of key points matched between two images. Based on the experimental results, there appear to be substantial performance differences among these algorithms that reveal key features of their relative strengths and weaknesses. This performance analysis will help choose which algorithm to use based on requirements like speed and accuracy.

In future work, I will apply this analysis to different data sets and fur- ther discuss the advantages and disadvantages of each feature-matching algorithm in other application scenarios.

Keywords: Feature Detection • SIFT (Scale-Invariant Feature Trans- form) • ORB (Oriented FAST and Rotated BRIEF) • KAZE • AKAZE (Accelerated-KAZE) • Dense SIFT • DAISY (Dense Adaptive Scale- Invariant Descriptor) • BRISK (Binary Robust Invariant Scalable Key- points) • FREAK (Fast Retina Key point).

Author: Aman Kumar

Abstract: Real-time object detection plays a pivotal role in the functionality of autonomous systems, including self-driving cars, drones, and industrial robotics. The ability to accurately and swiftly identify and localize objects in dynamic and often unpredictable environments is crucial for ensuring operational efficiency and safety. Traditional object detection techniques, limited by their computational complexity and lack of adaptability to real-world conditions, have been largely replaced by deep learning-based methods. These methods leverage advanced neural network architectures to achieve remarkable performance in terms of speed and accuracy.

This paper delves into the evolution and application of deep learning models for real-time object detection, highlighting key advancements in model architectures such as YOLO (You Only Look Once), SSD (Single Shot Detector), and Faster R-CNN. Modern approaches like EfficientDet and Vision Transformers (ViT) are also discussed for their potential to optimize performance in resource-constrained environments. A detailed examination of the challenges faced in implementing these models, including computational complexity, latency, robustness to environmental variations, and data scarcity, underscores the need for innovative solutions.

The paper explores various optimization techniques such as model compression, edge computing, and hardware acceleration, which enhance the viability of deep learning models for real-time applications. Furthermore, real-world applications across diverse domains—ranging from autonomous vehicles and drones to industrial robotics and surveillance systems—are analyzed to illustrate the transformative impact of these technologies. Emerging trends like multi-task learning, synthetic data generation, and federated learning are discussed as promising avenues for further advancements.

By addressing current limitations and identifying future directions, this paper aims to provide a comprehensive understanding of how deep learning models are revolutionizing real-time object detection and contributing to the growth of autonomous systems. The insights offered serve as a roadmap for researchers and practitioners seeking to harness the full potential of these technologies in real-world applications.

Keywords: Deep learning, real-time object detection, autonomous systems, convolutional neural networks, YOLO, SSD, R-CNN, self-driving cars, robotics, drones, performance optimization.

Authors: Meera Verma, Jatin Saxena, Kavya Saini

Abstract: Cloud computing has revolutionized how businesses and individuals store, manage, and process data, but it has also introduced new vulnerabilities, making robust cyber security frameworks critical. Block chain, known for its decentralization and immutable features, has emerged as a promising technology to enhance cloud security. This paper explores the integration of block chain into cloud computing security frameworks, examining the benefits, challenges, and potential implementations. We present a comprehensive analysis of block chain’s role in enhancing data integrity, authentication, access control, and overall system security in cloud environments. Additionally, we propose a novel block chain-integrated cyber security model tailored to address specific cloud vulnerabilities.

Keywords: Block chain, Cloud Computing, Cyber security, Data Integrity, Access Control, Decentralized Security, Cloud Vulnerabilities, Block chain Frameworks

Authors: Ankit Rathi, Kavya Patel

Abstract: This paper delves into the optimization of edge computing in the context of Internet of Things (IoT) networks. Edge computing provides a solution to the challenges posed by the vast amounts of data generated by IoT devices by processing data closer to the source. This minimizes latency and reduces bandwidth usage, which is critical for time-sensitive IoT applications. The paper identifies key challenges in edge computing for IoT, including resource constraints, security concerns, and network reliability. Furthermore, it discusses several innovative solutions aimed at overcoming these challenges, including efficient resource management, distributed architectures, and advanced security mechanisms. Finally, the paper highlights future directions and research areas for optimizing edge computing in IoT networks.

Keywords: Edge Computing, IoT Networks, Optimization, Resource Management, Security, Latency, Distributed Architectures, Data Processing

Authors: Arvind Mehta, Rahul Deshmukh, Mayank Aggarwal

Abstract: As quantum computing rapidly evolves, its potential to disrupt existing cryptographic protocols has raised significant concerns. This paper explores how quantum algorithms can be integrated into cryptography to enhance the security of future systems. We review the challenges posed by quantum computing, particularly with respect to classical encryption methods, and examine how quantum-resistant algorithms such as lattice-based cryptography, hash-based signatures, and quantum key distribution (QKD) offer promising alternatives. We also discuss the implications for data privacy and the future of secure communications in the age of quantum technology. The paper concludes by emphasizing the importance of developing quantum-resistant protocols and preparing for the advent of quantum computing.

Keywords: Quantum Computing, Cryptography, Quantum Algorithms, Quantum Key Distribution, Lattice-Based Cryptography, Data Privacy, Quantum-Resistant Algorithms

Authors: Dr. Meena Desai, Parched Saxena

Abstract: The rise of social media platforms has drastically transformed the way individuals communicate, share opinions, and interact with content. These platforms produce vast amounts of unstructured data that, if analyzed correctly, can provide valuable insights into public sentiment. This paper explores the application of Natural Language Processing (NLP) techniques in sentiment analysis of social media data. We discuss various methods such as tokenization, lemmatization, and machine learning algorithms for classification. Sentiment analysis can be used for a wide range of applications, including brand management, political discourse analysis, and customer feedback. The paper reviews key approaches in the field, evaluates challenges faced in the analysis of noisy and complex social media text, and offers potential solutions.

Keywords: Natural Language Processing, Sentiment Analysis, Social Media, Machine Learning, Text Mining, Tokenization, Lemmatization, Deep Learning, Opinion Mining

Authors: Kiran Patel, Sonia Gupta

Abstract: Big Data Analytics (BDA) has emerged as a critical tool for businesses, enabling data-driven decision-making and providing valuable insights. This paper explores the impact of BDA on business decision-making processes, focusing on how it enhances operational efficiency, customer understanding, and strategic planning. The study examines various BDA techniques, such as predictive analytics, machine learning, and data mining, and their applications in different business domains. Additionally, the paper discusses the challenges and limitations of implementing BDA, including data quality issues, the need for skilled professionals, and ethical considerations. The findings highlight the transformative potential of BDA in driving business success and competitiveness.

Keywords: Big Data Analytics, Business Decision-Making, Predictive Analytics, Machine Learning, Data Quality

Authors: Radhika Sinha, Nitin Kapoor, Sneha Rao

Abstract: Blockchain technology has the potential to revolutionize supply chain management by enhancing transparency, traceability, and efficiency. This paper examines the application of blockchain technology in supply chains, focusing on its ability to provide a secure and immutable ledger for recording transactions. The study explores the benefits of blockchain, such as reducing fraud, improving product traceability, and streamlining operations. Additionally, the paper discusses the challenges and limitations of blockchain adoption in supply chain management, including scalability issues, regulatory concerns, and the need for industry collaboration. The findings suggest that blockchain technology can significantly improve supply chain processes, but its successful implementation requires addressing technical and organizational challenges.

Keywords: Blockchain Technology, Supply Chain Management, Transparency, Traceability, Scalability

Authors: Priyanka Sharma

Abstract: The proliferation of the Internet of Things (IoT) has introduced new cybersecurity challenges, as the increasing number of connected devices expands the attack surface for cyber threats. This paper explores the cybersecurity threats associated with IoT and examines potential solutions to mitigate these risks. The study discusses the vulnerabilities of IoT devices, such as weak authentication, insufficient encryption, and lack of regular updates. It also highlights the importance of robust security frameworks, including network segmentation, anomaly detection, and secure communication protocols. The findings emphasize the need for a comprehensive approach to IoT cybersecurity that involves collaboration among manufacturers, developers, and users.

Keywords: Cybersecurity, Internet of Things (IoT), Vulnerabilities, Security Frameworks, Anomaly Detection

Authors: Radhika Menon

Abstract: Cloud computing has emerged as a transformative technology in the field of education, offering numerous benefits for both learners and educators. This paper explores the impact of cloud computing on educational practices, focusing on how it enhances learning and teaching experiences. The study examines the deployment of cloud-based tools and services in educational institutions, such as virtual classrooms, online collaboration platforms, and resource management systems. Additionally, the paper discusses the challenges and limitations associated with the adoption of cloud computing in education, including data security concerns, the digital divide, and the need for technical infrastructure. The findings highlight the potential of cloud computing to revolutionize education by providing flexible, scalable, and accessible learning environments.

Keywords: Cloud Computing, Education, Virtual Classrooms, Data Security, Digital Divide

Authors: Shalini Reddy

Abstract: Artificial Intelligence (AI) is revolutionizing medical imaging by enhancing the accuracy and efficiency of diagnostic procedures. This paper delves into the advancements AI has made in medical imaging, focusing on image recognition, segmentation, and interpretation. It discusses the integration of AI algorithms with imaging modalities like MRI, CT scans, and X-rays. The paper also addresses the challenges associated with AI in medical imaging, including data privacy concerns, the need for large datasets, and the interpretability of AI models. By examining these aspects, the paper aims to provide a comprehensive overview of the current state and future potential of AI in medical imaging.

Keywords: Artificial Intelligence, Medical Imaging, Image Recognition, Data Privacy, AI Algorithms

Author: Mansi Negi

Abstract: As cloud usage expands across sectors such as finance, stock markets, and industry, ensuring its security has become paramount. This paper addresses the growing demand for cloud services by focusing on the implementation of the Rijndael encryption algorithm. Additionally, it integrates a compression technique, Open-SSL, to tackle storage and management challenges. The proposed system features a user-friendly interface for file uploads to the cloud. Upon upload, the system employs compression followed by encryption to securely store the file. Access to the encrypted file is restricted to authorized personnel only.

Keywords: Cloud Security, Security Issues, Hybrid secure storage cloud, Rijndael, Open SSL, Base64 and SHA-256 Algorithm

Authors: Gayatri Kashyap, Hardik Verma

Abstract: Quantum computing has emerged as a revolutionary field, promising to solve complex problems exponentially faster than classical computers. Central to this promise are quantum algorithms, among which Shor's and Grover's algorithms stand out. This paper delves into the theoretical foundations, operational mechanisms, and potential applications of these groundbreaking algorithms, shedding light on their transformative potential in various fields.

Keywords: Quantum computing, Quantum algorithms, Shor's algorithm, Grover's algorithm, Cryptography, Optimization, Quantum hardware, Error mitigation, Algorithmic optimization.

Authors: Hemangni Mehta, Jeet Solanki, Nensi Panchal, Hemanshu Patel, Trupti Dilhiwal

Abstract: This paper presents a comprehensive analysis of the usage of traditional AI models and new generation advanced AI models in modern applications and tools development. The study explores the advantages and disadvantages of each model and provides insights into selecting the most suitable model based on specific application requirements. Furthermore, the paper investigates the feasibility and implications of using both traditional and advanced AI models simultaneously.

Keywords: Artificial Intelligence, Traditional AI Models, New Generation

Authors:  Mrs. Vidhya Mali, Mrudula M. Ligade, Samruddhi S. Bhosale, Suhani M.Patil, Manasi A. Shendure

Abstract:  In this today’s world there is huge expansion in the technical field in today’s world. In early world there is only computer system which performs some specific task but now the technology has grab this world of computer there are many technologies like artificial intelligences, machine learning and many more. One of the claim of artificial intelligence is Natural processing Language. The NPL helps humans to comminute with the computer system in their own language foe example voice assistant. This paper ambition to create an electronic mail system that will support even visually challenged people to use these service areas to communicate without earlier exercise. By which machine will no longer permit the client utilize the console alternatively will work best on mouse operation and speech conversion to text. Even, the present device could be utilized by anyone too as an instance the one like the person who can't even study. The device is totally primarily reached from cooperative voice feedback to be able to make it clear and efficient to apply.

Keywords: Speech recognition, Text To speech Voice email, visually challenges people, Natural processing language, python.

Authors: Md. Rakibul Hasan, Md. Kamrul Hassan_, Saikat Barua

Abstract: Mobile phones have become an essential part of our daily lives, enabling us to communicate, access information, and perform various tasks anytime and anywhere. However, the development of mobile phones and their network technologies was not a straightforward process. It involved many challenges, innovations, and collaborations among different stakeholders. In this paper, we provide a brief overview of the history of mobile phones and their network generations, from the early experiments with radio communication to the current deployment of 5G and beyond. We highlight the main features, advantages, and limitations of each generation, as well as the social and economic impacts of mobile phones on society.

 Keywords: Mobile phone Network Generations (1G, 2G, 3G, 4G, 5G, 6G), Wireless communication, Social and economic impacts, Challenges and opportunities, History and evolution, Spectrum and bandwidth, Data rates and quality, Coverage and connectivity, Roaming and interoperability

Authors:-Md. Rakibul Hasan, Saikat Barua

Abstract:-Modeling tools play a crucial role in both educational and industrial settings, enabling engineers to simulate power systems under various operating conditions, including normal and faulty scenarios. This research focuses on the design of a relay system capable of addressing both over and under frequency conditions. Digital relays exhibit distinct advantages over traditional electromechanical relays, notably in terms of accuracy and response speed.

The significance of frequency control cannot be overstated, as significant fluctuations can potentially lead to complete power system blackouts. Historical incidents have demonstrated the severe consequences of frequency instability, often stemming from supply-demand imbalances and unforeseen contingencies. With the rise of distributed generation and the inherent challenges of islanding in modern power systems, the attention of both industrialists and researchers has once again turned to frequency relaying solutions.

 This study aims to evaluate the performance of the proposed digital frequency relay under diverse system dynamics using simulation tools such as MATLAB/Simulink and microcontroller-based implementations. By conducting rigorous testing and analysis, this research endeavors to contribute to the enhancement of grid stability and the prevention of power system disruptions caused by frequency deviations.

Keywords:-Digital Frequency Relays, Grid Synchronization, Power Grid Maintenance, Frequency Protection, Electrical Grid, Relay Technology, Grid Stability, Grid Resilience, Frequency Deviations

Authors:Maryam Al-Washahi,Yassine Jamoussi ,Youcef Baghdadi

Abstract:Some studies have examined various model-based testing approaches and have presented an overview of these methods employing diverse techniques. Similarly, this paper reviews the state of the art of model-based testing using a different framework called faceted classification framework. This framework consists of four perspectives or views which are what, why, how, and which. Therefore, it has become easier to explore and compare the different approaches by using this framework as well as identify the gaps and weaknesses

Keywords:Model-Based Testing, Classification Framework ,UML semantics ,Design Pattern 

Authors:Niharika Sharma, Rashmi Jain

Abstract:Natural Language Processing (NLP) has made remarkable strides in recent years, revolutionizing how we interact with technology and one another. However, the rapid growth of NLP applications has brought to light pressing ethical concerns regarding bias and fairness. This paper explores the multifaceted landscape of ethics and bias in NLP, examining the potential societal impact of biased language models. It delves into the challenges of identifying, understanding, and mitigating biases in NLP systems while highlighting the importance of responsible development and deployment.

Keywords-:CMOSEthics, Bias, Natural Language Processing, Fairness, Inclusivity, Machine Learning, Societal Impact, Responsible AI, Regulation, Industry Initiatives, Bias Mitigation, Ethical Guidelines, Transparency.

Authors:N. Salmasulthana, Srikanth Veesam

Abstract:In this paper we propose a novel design of a low power static random access memory (SRAM) cell for high speed operations. The model adopts the voltage mode method for reducing the voltage swing during the write operation switching activity. Dynamic power dissipation increases when the operating frequency of the SRAM cell increases. In the proposed design we use two voltage sources connected with the Bit line and Bit bar line for reducing the voltage swing during the write “0” or write “1” operation. We use 90 nm CMOS technology with 1 volt of power supply. Simulation is done in Microwind 3.1 by using BSim4 model. Dynamic power for different frequencies is calculated. We compare it with conventional 6-T SRAM cell. The simulation results show that the power dissipation is almost constant even the frequency of the proposed SRAM model increases. This justifies the reduction of the dynamic power dissipation for high frequency CMOS VLSI design.

Keywords-:CMOS; Dynamic power; SRAM; Voltage Mode; Voltage Swing

Authors:  Shraddha Pandey, Dr. Sudhakar Singh, Santosh Bahe

Abstract : For biological processes to occur, noncovalent interactions are essential. Everything from molecular recognition to molecular assembly to enzyme catalysis to the dynamic behaviour of biomolecules is aided by noncovalent interactions. Applications of noncovalent interactions have been increasingly important in recent years and are on the verge of becoming much more so in issues relating to a variety of fields of chemistry, biology, and physics. Due to their prevalence and high binding energy, noncovalent interactions that are dominated by electrostatic contributions have drawn the most attention. Electrostatic interactions have a wide spectrum and are directed. However, the directional components of electrostatic dominated noncovalent interactions have been completely ignored in the literature up to this point due to the fundamentally long range impactful nature of electrostatic contributions. To fully comprehend the electrostatic contributions, this thesis focuses on the significance and directional nature of long-range electrostatic interactions. In this chapter, we have provided a quick introduction of noncovalent bonds and their function in chemistry. In order to demonstrate how understanding of the significance of long-range electrostatic interactions has changed over time, we also went over a brief history of the scientific literature.

Keywords- Computer, Assessment, Science, Related to Materials, Bio-Materials

Authors:  Riya Mishra,  Dr. Amit Chouksey

Abstract: The underutilised section of the wireless spectrum needs to be used effectively due to the anticipated exponential growth in traffic volume in 5G-based networks. Due to the popularity of numerous smartphone applications, cellular networks have seen a huge increase in data traffic in recent years. Therefore, expanding the network's capacity is essential to support new applications and services. However, when cooperative relays are used, multihop D2D communication necessitates the use of extra nodes for data forwarding. We must address the issues of transmission scheduling, routing, and channel allocation in order to establish effective multi-hop D2D cooperative communication. In order to deal with network dynamics, the suggested system architecture additionally emphasises online algorithms.

Keywords-Cognitive Radio Network, Device-to-Device Communication, Multihop-Based Protocol.

Authors: Riya Mishra, Dr.Amit

Abstract: The underutilized portion of the wireless spectrum will need to be better utilized due to the expected exponential growth in traffic volume in 5G-based networks. Apps for smartphones have caused an increase in data traffic on cell phone networks recently. As a result, expanding the network's capacity to accommodate new applications and services is critical. D2D communication with multiple hops requires more nodes for data transmission, especially when cooperatively-operated relays are used. Long-Term Evolution (LTE) is the most recent and most technologically advanced cell phone technology that is about to be introduced to the market. LTE and its advanced version appear to be an appealing solution for many businesses since they offer exceptional peak data speeds in both the uplink and downlink directions. Public safety communications is currently one of the fastest-growing fields in the world. In accordance with two homogeneous Poisson Point Processes, beacon-enabled and simple LTE terminals are dispersed in the vicinity of a significant event. This research looks into direct-to-device (D2D) communications. In this paper, we explore the likelihood of LTE mobile terminals forming in D2D networks using a stochastic geometry technique, and we then build an unique D2D protocol.

Keywords-LTE, Machine Learning, Device to Device Communication, Public Safety Applications.

Authors: Omkar Javali, Varsha Desai, Snehal Kadolkar, Sapana Yakkundi

Abstract: Traffic law violations are a major cause of concern in our society. People’s irresponsible and careless attitudes towards driving laws has the potential to weaken the moral fiber of our society. While some progress has been made to update traffic laws, the human factor in our existing system remains a hurdle, resulting in dissatisfactory outcomes that could have been prevented. Legislation has mandated the use of safety belts and helmets by drivers, but with millions of drivers on the roads every day, it is difficult for police officers to monitor compliance. To ensure compliance with traffic regulations and improve road safety, an effective and efficient system for automatic traffic rule violation detection and fine collection is crucial in modern cities. Fortunately, the emergence of technological advancements has enabled Machine Learning techniques to provide solutions to this problem. By leveraging image processing and Machine Learning, our proposed system is able to analyze CCTV footage to detect traffic violations, such as the failure to wear seat belts or helmets. Machine Learning algorithms such as Convolutional Neural Network (CNN) and Automatic Number Plate Recognition (ANRP) can be used to increase the accuracy and efficiency of detecting traffic rule violations and thus improve compliance with the law and road safety.

Keywords:  Machine Learning, Violation Detection, ANRP, CNN