http://dspace.dtu.ac.in:8080/jspui/handle/123456789/100 2026-10-17T19:52:50Z 2026-10-17T19:52:50Z RAY, SANTOSH KUMAR Susan, Seba (SUPERVISOR) http://dspace.dtu.ac.in:8080/jspui/handle/repository/23383 2026-10-16T06:34:06Z 2026-09-01T00:00:00Z

Title: DESIGN AND ANALYSIS OF CLASSIFICATION FRAMEWORK FOR STREAMING DATA Authors: RAY, SANTOSH KUMAR; Susan, Seba (SUPERVISOR) Abstract: The rapid growth of digital technologies, sensor networks, and internet services generates massive amounts of streaming data that arrive sequentially and must be processed in real time under strict constraints. Its statistical properties often change over time due to environmental factors, user behavior, or system dynamics, causing concept drift. This challenges traditional machine learning models, which assume stationary data, resulting in degraded predictive performance on evolving data streams. Therefore, adaptive, scalable online machine learning frameworks are needed to learn incrementally, detect changes in patterns, and maintain performance in dynamic environments. This thesis explores robust classification for drift detection and adaptation, as well as anomaly-detection frameworks for streaming data. The research combines ensemble learning, drift detection, and online methods to create adaptive systems for evolving data. It explores various online frameworks in real-world streaming, aiding machine learning systems that process large data streams in real time while staying accurate, adaptable, and efficient. It begins with an adaptive ensemble classification framework to tackle concept drift, which can impair model performance. To address this, an error-weighted drift-adaptive ensemble framework improves classification with evolving data streams by combining multiple online classifiers, each built with adaptive random forests based on Hoeffding trees for incremental learning. It integrates multiple drift-detection techniques that monitor individual prediction errors to detect data shifts. When detected, the framework adapts by updating the ensemble and adjusting the learning rate. A dynamic error weighting strategy assigns weights based on real-time errors, giving more influence to better models. Experiments on real and synthetic data show that this approach significantly improves accuracy, precision, recall, and F1-score compared with other methods, demonstrating its effectiveness in handling both gradual and abrupt concept drift. vi Building on the adaptive classification framework, this work explores anomaly detection in streaming environments, where identifying cyber intrusions, fraud, or hazards in real time is crucial. Detecting anomalies is challenging due to unlabeled data, shifting patterns, and ongoing learning. A heterogeneous, multi framework stacked ensemble approach combines River and PySAD, offering various unsupervised algorithms for incremental learning. The ensemble uses stacking: base models analyze the data and generate anomaly scores, which are then fed into meta learners such as logistic regression and AdaBoost to produce final predictions. By integrating tree-based, density-based, isolation-based, and distance-based methods, detection improves across different anomalies. Experiments on benchmark datasets show the ensemble outperforms individual models and latest techniques, confirming that combining diverse frameworks enhances accuracy and robustness. This study assesses Delhi's air quality data using online anomaly detection algorithms to identify pollution events. Sensor data monitor changes in pollutant levels and deviations in AQI to detect abnormal conditions. Multiple models are tested; some perform well across different stations and pollutants. Results indicate that online machine learning improves real-time monitoring and detection of dangerous pollution episodes. The study emphasizes the potential of streaming anomaly-detection algorithms to enhance the responsiveness and reliability of environmental management systems. This research develops classification and anomaly-detection frameworks and evaluates online machine learning tools for streaming data. With the rise of real time analytics, several tools support incremental learning and streaming data, but their effectiveness varies in terms of design, efficiency, and compatibility. A systematic evaluation compares their performance on classification and anomaly detection tasks using accuracy, precision, recall, F1 Score, and ROC AUC. Results show that some frameworks better support streaming data through optimized incremental learning, aiding in tool selection. Additional experiments evaluate unsupervised anomaly detection algorithms in IoT environments with continuous sensor data, revealing their strengths and limitations. Ensemble and hybrid methods generally outperform individual models in complex, evolving data. This study explores adaptive machine learning for streaming analytics, combining ensemble methods, heterogeneous frameworks, and performance assessments. The approaches improve prediction, anomaly detection, and adaptability, advancing data stream mining and online learning. Applications include network security, finance, environmental sensing, healthcare, and smart cities. This work enables effective processing of continuous data, supporting next generation real-time machine learning systems.

2026-09-01T00:00:00Z SRIVASTAVA, MEENAKSHI Sisaudia, Varsha (SUPERVISOR) Meena, Jasraj (C0-SUPERVISOR) http://dspace.dtu.ac.in:8080/jspui/handle/repository/23345 2026-10-16T06:29:54Z 2026-04-01T00:00:00Z

Title: DESIGN AND DEVELOPMENT OF AI-BASED FRAMEWORK FOR PLANT DISEASE CLASSIFICATION Authors: SRIVASTAVA, MEENAKSHI; Sisaudia, Varsha (SUPERVISOR); Meena, Jasraj (C0-SUPERVISOR) Abstract: Agriculture plays a pivotal role in economic growth and food security, with over 60% of India’s population relying on it for livelihood. However, crop production faces ongoing obstacles, including pest invasions, environmental variability, and plant leaf diseases, which can severely impact output and quality. Manual disease diagnosis takes time and money and is frequently out of reach for farmers in remote areas. This requires the development of intelligent, automated illness-detection technologies that are both timely and accurate. This thesis provides a complete framework that combines Artificial Intelligence (AI), Deep Learning (DL), and the Internet of Things (IoT) to identify and classify plant diseases. The work begins with a thorough literature review and a taxonomy of AI and IoT-based tech niques to identify key research needs. It then explores transfer learning-based models and the Deep Transfer Learning and Extreme Learning Machine (VESTACK) framework, fol lowed by attention-driven deep learning models such as Agriculture Transfer Learning Vision Transformer (AgriTL-ViT) and Asymmetric and Dilated Convolution ResNet (ADResNet) with Attention, which improve feature extraction and classification efficiency. Furthermore, a Federated Learning (FL) technique based on Convolutional Neural Networks (CNNs) and Transformer encoders is created to enable decentralized, privacy-preserving plant disease diagnostics in real-world agricultural environments. The suggested methodologies show consistent performance across diverse plant leaf datasets, demonstrating the promise of AI-driven solutions for real-time, scalable, and smart agricul ture. This thesis advances precision farming by providing a pathway to early disease iden tification, reduced crop losses, and enhanced agricultural sustainability. Future directions include expanding the framework for multi-disease identification, integrating cross-device FL, and deploying real-world IoT systems to improve smart agriculture operations.

2026-04-01T00:00:00Z NARULA, MANU Vishwakarma, Dinesh Kumar (SUPERVISOR) Meena, Jasraj (CO-SUPERVISOR) http://dspace.dtu.ac.in:8080/jspui/handle/repository/22744 2026-06-08T05:32:09Z 2026-02-01T00:00:00Z

Title: DESIGN AND DEVELOPMEN T OF SECURE FEDERATED LEARNING FRAMEWORK FOR DATA-SENSITIVE APPLICATIONS Authors: NARULA, MANU; Vishwakarma, Dinesh Kumar (SUPERVISOR); Meena, Jasraj (CO-SUPERVISOR) Abstract: In an era where security breaches and cyber fraud have become increasingly common, con- cerns about data privacy have also risen exponentially. This presents a substantial hur- dle for training Artificial Intelligence (AI) applications that handle sensitive data, such as healthcare and finance. Federated Learning (FL) provides a solution for Data-Sensitive Applications (DSA), enabling collaboration between competing parties while guaranteeing complete control over data privacy. Over time, the attacks on FL have evolved, and the native technology alone is insufficient to maintain the expected privacy levels when han- dling such data. Hence, efforts are made to integrate other technologies with FL to enhance certain aspects, such as resource management, security, privacy, and efficiency, with varying results and pros and cons. This thesis examines the integration of security mechanisms with FL, focusing on networks with resource scarcity, which include AdHoc networks, Internet of Things, and Internet of Medical Things (IoMT), commonly found in the field of healthcare and other DSA, addressing critical issues such as data privacy, security, access control, and scalability. It proposes solutions to enhance data security and privacy in healthcare systems by examining various existing studies and frameworks. FL is adopted for its ability to facilitate the integration of AI in domains where conven- tional centralized training approaches are infeasible due to data distribution constraints, privacy concerns, or competitive restrictions. Furthermore, FL enables remote or decen- tralized systems to achieve performance comparable to state-of-the-art centralized methods while incurring minimal resource overhead. By eliminating the need for centralized data aggregation, FL significantly mitigates the risks associated with large-scale data breaches and unauthorized access. This thesis makes a substantial contribution in summarizing the existing state of FL, its various types, supporting technologies, and challenges, designing hybrid security frameworks for FL aimed at DSA networks, and performing extensive evaluation of the state-of-the-art FL techniques. Our contributions are as follows: • An extensive systematic literature review has been conducted to investigate the cur- x rent state of FL implementations in DSA fields, such as healthcare and finance. This analysis highlights the complex challenges of implementing FL in mainstream health- care applications, including concerns about transmission costs, data security, privacy, and data/system heterogeneity. Additionally, we provide a detailed taxonomy for the existing literature, focusing on DSA. Alongside identifying the gaps, the review also highlights the fundamental challenges that FL may pose in real-world scenarios. • As organizations and developers explore FL solutions for various applications, the multitude of FL tools and frameworks can feel overwhelming for a beginner. We studied various tools and frameworks available in both open and proprietary forms to provide a concise view of their advantages, limitations, and utility, thereby filling a gap in the proposed literature. • It contributes to improving the security in resource-constrained DSA networks such as IoMT by employing quantization. Unlike prior techniques, we incorporate a workload- aware client selection scheme to overcome the quantization loss, optimize bandwidth, and provide stability to the training network by minimizing the straggler nodes with negligible performance overhead. Thus, providing a reliable security solution to resource-constrained DSA. The experimental results show a 77% to 95% decrease in straggler nodes and an 8% to 0% decrease in accuracy compared to standard FL, depending on the dataset’s complexity. • To ensure security in small edge networks, IoMT networks, and AdHoc networks, a dynamic, lightweight cryptographic FL framework is proposed. Conventional cryp- tographic techniques are computationally expensive for small networks. Thus, we propose a block cipher with an ever-changing key that takes multiple client and net- work attributes into account and is unique to all participants. The overhead in- duced by the cipher and key generation is significantly lower than that of state-of- the-art solutions available. The experimental evaluation indicates that the proposed scheme’s relative performance is comparable to that of native FL. The results reveal an increase of approximately 4.5% to 8% in per-round computation time compared to standard FL. However, this marginal computational overhead is justified by the significantly enhanced security guarantees provided by the proposed approach, par- ticularly in resource-constrained network environments, where conventional security mechanisms with similar computational budgets offer comparatively lower protection • To ensure privacy in networks with sufficient resources, we propose a hybrid FL xi framework that utilizes differential privacy in conjunction with homomorphic encryp- tion. To counter the loss incurred by the differential privacy scheme, we employ the workload-aware client selection. The simulation results demonstrate minimal compu- tational overhead and comparable performance to other state-of-the-art techniques. The proposed Fed-HDVE imposes a minimal time overhead of 2.8 to 3.5 seconds, depending on the model size. The performance evaluation, analysis, and experimental results indicate that the proposed solutions offer a viable and effective security solution. Moreover, the comparative study demonstrates that the suggested approaches perform on par with existing solutions, with lower computational overhead and resource consumption.

2026-02-01T00:00:00Z MOOL, AKSHAY Panda, Jeebananda (SUPERVISOR) Sharma, Kapil (CO-SUPERVISOR) http://dspace.dtu.ac.in:8080/jspui/handle/repository/22696 2026-03-12T05:10:21Z 2024-10-01T00:00:00Z

Title: FACE DETECTION AND TRACKING Authors: MOOL, AKSHAY; Panda, Jeebananda (SUPERVISOR); Sharma, Kapil (CO-SUPERVISOR) Abstract: With the use of various technological advancements and devices in today’s routine life, detection and tracking of human faces and facial features become very essential areas of focus. They become important so that more techniques could be developed for increasing their working efficiency. The field of Computer Vision uses these intermediary processes of face detection and tracking to track and analyze the input of visual information about humans, their faces and/or body movements, and correspondingly proceed to the desired application. The present thesis work has been taken up for the development of (i) an optimiz- able face detection and tracking model based on facial landmark localisation and feature tracking, for better and efficient processing of faces in high quality video streams, and (ii) a Non-Neighbourhood Background Elimination component us- ing the built model with mathematical and statistical modelling, for reducing the processing time and computations required for finding the target face in a frame when it has already been detected. Many algorithms have been developed to facilitate Face Detection and Track- ing applications. Viola and Jones were able to develop an algorithm in 2004, that achieves real-time performance with decent accuracy in detecting faces. It was one of the first algorithms to achieve such efficient performance, that is why it is still used as a standard algorithm to compare against other upcoming algorithms, and therefore has been specifically discussed in this thesis. vii There is a lot of visual information that is generated in various fields, rang- ing from daily routine to specialized applications. Processing all these types of information efficiently is a valid concern and need focused research. Most face detection algorithms have to deal with low quality data in videos, since they’re mainly focused on surveillance applications, whose information capturing devices capture less information per frame. Consequently, this thesis reviews some state- of-the-art face detection algorithms and compares their processing efficiency on low and high quality videos. The comparative analysis reveals that these recent and modern algorithms do not work as effectively on high quality videos as they do on lower quality videos. Therefore, there is an increasing need to focus re- search on analysis of high quality information in videos in an efficient manner, so as to keep up the pace of their analysis with the information that is generated. High quality videos (data generated by current applications like social media, Multimedia content, etc) mostly exist in offline mode, that could be used for post processing by the Computer Vision applications. To address this need, an effort has been made to focus on developing such an algorithm that gives faster results on high quality videos, at par with the algorithms working on live low quality video feeds. The proposed algorithm uses Convolutional-MTCNN as base algo- rithm, and speeds it up for high definition videos. The proposed model speeds up the face detection process really fast, up to 19+ FPS, while still maintaining above 90% accuracy. This paper also presents a novel solution to the problem of occlusion and detecting partial or fully hidden faces in the videos. This is achieved by using statistical and probabilistic approaches, given that the face has been identified in first few frames, to give the algorithm an estimate of where the face should be in the occluded region. Since the focus of our research is to efficiently process high quality data, some viii commercially used face detection algorithms in open literature have also been considered in our research. Models like FaceNet, HOG, YuNet, alongwith Viola- Jones algorithm and MTCNN, have been discussed and analysed in our thesis. The research done is compared against these models, in an effort to improve their performance in commercial settings. Further analysis lead to the conclusion that modern face detection algorithms fail to provide optimal results when they have to deal with larger amounts of data per frame while processing higher quality videos. This thesis discusses an- other proposed work that tackles discussed problem and offers a solution to deploy commercially used state-of-the-art face detection algorithms to process only the regions of interest in a frame, and discard the rest to decrease the data to be processed. The model maintains the accuracy of the base algorithm while de- creasing the processing time per frame, thereby increasing the overall efficiency. The selection of region of interest is dependent on the detection of facial window in the previous frame. Therefore, the choice of base algorithm plays an important role in determining the speed of the model. The model achieves increased pro- cessing speeds of about 69–76% more than the standalone usage of the detection algorithms for analyzed frame rates.

2024-10-01T00:00:00Z