TIET Digital Repository Thapar Institute of Engineering & Technology
Not a member yet
6889 research outputs found
Sort by
An Ultra-Lightweight Visual Privacy Protection System for Deep Optical Light Field Imaging Applications
No full textLight field imaging captures both spatial and angular information from a scene, thus
enabling advanced computational photography applications such as post-capture refocusing,
depth estimation and 3D reconstruction. Such capabilities make light field data
extremely valuable for applications in areas such as healthcare, intelligent surveillance,
virtual/augmented reality and robotics. However, 4D light field data is rich and very
high dimensional, and it poses enormous threats to visual privacy, especially in sensitive
personal or medical information scenarios. Classic encryption schemes are generally
unable to handle the unique structure and bulk of the data streams that light fields
generate in an efficient manner. In order to address this novel and unreleased threat, this
thesis proposes PRESHMAC-256, a hybrid encryption scheme that combines the ultralightweight
Present block cipher with the cryptographic robustness of HMAC-SHA256.
The suggested method is able to encrypt the sub-aperture images extracted from light
field captures with 64-bit blocks and with a 128-bit symmetric key derived securely
from SHA-256 hashing. The encryption is designed to be lightweight and reversible,
claiming a robust level of security with minimum computational and power overhead-an
essential requirement for real-time and resource-constrained applications like embedded
or mobile devices. The experimental validation carried out using the EPFL Light
Field dataset to prove that the presented approach indeed works. Evaluation metrics
considered are histogram analysis, information entropy, pixel correlation coefficients,
PSNR, SSIM, key sensitivity analysis, occlusion attack resilience and avalanche effect
measurements. The findings emerge favoring PRESHMAC-256 as not only ensuring
good levels of image security and fidelity, but also decreasing model complexity and
processing latency as compared to the traditional encryption schemes. Certainly, this
work thus opens possibilities for such an application on the grounds of a fairly practical,
highly scalable and secure environment in which data privacy and computation
efficiency would both matter significantly
Parkinson’s Disease Early Detection using Hand-drawn Data with Deep Learning, Ensemble Learning, and Explainable AI
No full textParkinson's Disease (PD) impairs the brain's ability to control movements. Early diagnosis remains
challenging due to limitations in conventional diagnostic methods, which are expensive and
technically demanding, and gait and voice analysis, which are non-invasive but may require
specialized equipment or controlled environments. This research aims to develop and evaluate a cost-
effective, non-invasive, and accurate method for early PD diagnosis using hand-drawn images using
deep learning models, the Ensemble method, and Explainable AI. A dataset of 3,264 images (1,632
healthy and 1,632 PD) was normalized, augmented (using flipping, zooming, and rotation), and split
into training, validation, and test sets. Pre-trained models like VGG16, VGG19, ResNet50, and
DenseNet121 were used for feature extraction, followed by custom classification layers for final
prediction. Among individual models, DenseNet121 achieved the best results, with 98\% accuracy,
0.97 sensitivity, and 0.99 specificity. The soft-voting ensemble (VGG16, ResNet50, and
DenseNet121) outperformed it, attaining 99\% accuracy, 0.98 sensitivity, and 1.00 specificity.
Employed XAI techniques such as Grad-CAM, LIME, and SHAP to enhance interpretability, with
Grad-CAM providing the most effective visual explanations. Although this approach requires
moderate computational resources, it establishes a foundation for future multimodal diagnostic
systems integrating EEG, MRI, and voice data to improve diagnostic confidence further
High Performance Ternary Logic Design With CNTFETS for Future Nanoelectronics
No full textA viable substitute for conventional binary systems is ternary logic which provides higher data density,
simpler circuitry, and possibly lower power consumption. In this work, we investigate the use of
Carbon Nanotube Field-Effect Transistors (CNTFETs) for the implementation of ternary logic.
Because of their superior electrical properties—such as high carrier mobility, excellent scalability, and
low power dissipation—CNTFETs have become a promising candidate for next-generation Nano
electronic devices. In this work ternary logic circuits of NAND and NOR and standard ternary inverter
are develop and simulated while considering the CNTFET as a device. By choosing the right threshold
voltages for the various states (0, 1, 2), the use of CNTFETs in ternary logic design provides improved
tenability in addition to making circuits smaller and more energy-efficient. The simulation results
further demonstrate the higher energy efficient.
This work proposed new circuits of standard ternary logic gates, including the ternary inverter, ternary
NAND, and ternary NOR, using CNTFET device. The proposed and existing circuits of these gates
are simulated with 32nm CNTFET technology node using Synopsys HSPICE. The simulation results
confirm the correct ternary behavior, where the ternary inverter follows Vout = 2 - Vin, the ternary
NAND produces a high output unless both inputs are high, and the ternary NOR generates a low output
when either input is high. Compared to existing designs, the proposed CNTFET-based ternary circuits
demonstrate improved performance in terms of reduced power consumption, lower transistor count,
and enhanced switching characteristics. These results validate the feasibility of CNTFET-based ternary
circuits, paving the way for efficient ternary arithmetic circuit design for low-power VLSI applications
Decision Support System for Early Detection of Cardiac Health Status
No full textCardiovascular diseases have surpassed cancer as the leading cause of death on the planet today. Numerous decision-making systems with computer-assisted support have been developed to assist cardiologists to detect heart disease, and thus, lowering the mortality rate. The purpose of this research is to classify audio signals received from the heart as normal or abnormal. The PhysioNet Computing in Cardiology (CinC) 2016 benchmark dataset, popularly known as PhysioNet 2016, has been used to validate the proposed methodology presented here. PhysioNet 2016 contains a total of 3,200 PCG recordings divided into sub-datasets A-F. In this work, researchers have proposed three different techniques with respect to decision support system. In first technique, textural features such as Linear Binary Pattern (LBP), Adaptive-LBP, and Ring-LBP have been extracted from the existing spectrogram and combined with the features extracted from the chromagram. It has been observed that the combination of features extracted from both the image variants has resulted in a greater accuracy as compared to the scenario where researchers were using only the spectrogram. The experiment yielded the mean accuracy, precision, and F1-score as 94.87, 93.11, and 95.273, respectively. In second technique, authors suggest a unique methodology for the detection of important events in an audio signal using a biologically-inspired depiction of the audio stream through a picture known as Gammatonegram which correlates to the processing of audio in the cochlea membrane of the inner human auditory system. In this study, texture-related features which include Linear Ternary Pattern (LTP), Local Directional Pattern (LDP), Geometric Local Textural Pattern (GLTP), and Local Phase Quantization (LPQ) have been extracted from a visual representation of PCG signal such as Spectrogram, Scalogram, Mel-spectrogram, and Gammatonegram. As compared to the case when researchers were employing other images, it has been noticed that the fusion of the features retrieved from the Gammatonegram has led to an increased overall classification performance metrics. The experiment resulted an overall accuracy of 94.00 % with precision and F1 scores of 91.77 and 93.61 respectively. In third technique, conversion of PCG signals into 2-D Time-Frequency images, viz. Tempogram, Chromagram, and Spectrogram has been performed. Further, data augmentation iv methods have been used to improve the imbalanced dataset. The present study utilizes a U-Net architecture-based Convolutional Neural Network (CNN), incorporating CNN, Residual Neural Network (ResNet), Visual Geometry Group (VGG), and Inception V3 blocks as encoders and decoders, to make a comparative evaluation of PCG signal classification models. The model under consideration achieved a validation accuracy of 95.89% with F1-score as 95.90%
Interdependence Between Segregation and Flow Properties of Bulk Solids
No full textThis thesis presents the results of an ongoing investigation into the characteristics of segregation,
which is critical for ensuring product homogeneity and process reliability. Additionally, moisture
content certainly affects the segregation behaviour of powders, which may degrade the product
quality and cause lot rejection. Therefore, understanding the segregation characteristics affected by
changes in moisture content plays a vital role in achieving the desired goals in powder handling
industries. In this research, a deeper understanding of powder segregation and flow behaviour was
achieved by addressing a critical gap in the literature for Geldart Group A to B borderline powders,
providing practical insights and predictive tools applicable to real-world scenarios. A key
contribution is the introduction of a novel dimensionless parameter, the modified dimensionless
cohesion number, which improves the prediction of segregation in complex powder fluidising
systems. High-accuracy models were developed by incorporating important factors such as moisture,
cohesion, flow function, and particle properties. Notably, the research also established clear empirical
relationships between moisture levels and segregation behaviour, an area with minimal prior
modelling despite its industrial significance. By highlighting the role of moisture management in
maintaining product uniformity and preventing lot rejection, this work supports higher safety and
quality standards. These models were validated across a diverse set of powders, including fly ash,
pharmaceuticals, detergents, and semolina, demonstrating strong generalisability. The study further
explores segregation challenges in pneumatic conveying systems. A 75:25 blend ratio of coarse-to
fine ash was found optimal for dense phase flow. Criteria based on the bulk powder Froude number
(<10) and coarse-to-fine ratio (>10) were proposed for reliable conveying. By highlighting the role
of moisture management in maintaining product uniformity and preventing lot rejection, this work
supports higher safety and quality standards. Based on the comprehension testing that includes the sifting segregation test, fluidisation test, particle and flow properties testing on various powders such as 6 different fly ash, sand, three different brands of detergent and semolina, the model for sifting and fluidisation segregation index has developed based on particle, flow and powder bed properties. The developed models can be used
as a design tool based on particle, flow, and powder bed properties, which may assist in controlling
the segregation issues by considering the required changes in designing and operating parameters. A
study which aims at modelling the sifting and fluidisation segregation index for Geldart group A to B
borderline materials, for which very little research has been carried out till date. In a separate set of experiments, physical and flow property tests were carried out on 8 pharmaceutical powders. The results obtained from heap analysis and shear cell testing have been compared. A relationship to represent the static angle of repose has been developed using a bulk powder-based Froude number and fine size. The experimental results and predictions for a range of bulk properties have shown that the values of static angle of repose decrease with an increase in the value of bulk powder Froude number, increase in particle shape factor and decrease in the median to fine ratio. Out of all the dimensionless parameter groupings, the particle shape factor strongly influences the static angle of repose, as indicated by its larger absolute value of the exponent in the power function format relationship. Sifting and fluidisation segregation characteristics were determined for 6 different fly ash samples (particle size ‘d(50)’ ranging from 68 to 141 µm) using standard testers. The results have shown that the coarser particles have a greater tendency to sifting segregation, and the finer powders respond more to fluidisation segregation. The angle of repose for the fine ash and coarse ash were 55° and
38˚, respectively, which indicated poor to good flowability conditions. The flow function test shows
that all the samples were in an easy-flowing to a free-flowing zone. The angle of repose and material
flow function have provided a good correlation with the sifting segregation index. In contrast,
cohesion between particles, the ratio of free terminal velocities and diameters for coarse to fine
particles have shown a good fit with fluidisation segregation indices. For both sifting and fluidisation
segregation, the model correlation values are 0.91 and 0.94, indicating the predicted results are a
good fit to the experimental data. Experiments were carried out using sand, three different brands of detergent and semolina powders. Additionally, data from six fly ash samples were taken for the purpose of modelling. While comparing the powder characteristics in the first and last samples in sifting segregation and top and
bottom samples in fluidisation segregation, considerable differences in the bulk properties were
found, indicating the occurrence of segregation in the case of Geldart group A to B borderline
powders. A new model has been developed for the sifting segregation index using flow function,
course-to-fine ratio, and shape factor, which resulted in 97% prediction accuracy. A novel
dimensionless cohesion number has been developed as a ratio of inter-particle cohesion to the
dynamic pressure of air along; this number has been used with minimum fluidisation velocity to
model the fluidisation segregation index. The model has shown an 86 % fit to the experimental data. Another study aims to model the sifting and fluidisation segregation index for powders from Geldart
group A to B borderline while considering the change in moisture content. Five different powders,
such as sand, three different brands of detergent, and semolina powders, have been utilised for
physical and flow properties and segregation tests at three different moisture content levels. The
results revealed that with an increase in moisture content, there is a decrease in the value of the
sifting and fluidisation segregation index. A new model has been developed for the change in sifting
segregation index w.r.t change in moisture content based on flow function, course-to-fine ratio, and
shape factor, providing a good fit of 91 % accuracy with the experimental data. By using the
minimum fluidisation velocity of the particle and a novel dimensionless cohesion term, a model was
developed for the change in fluidisation segregation index that showed an 85 % fit with the
experimental data. In the initial study, the model presented in Chapter 4 was best described by the
parameter "Median to Fine ratio (d50/d10)." To encompass the entire range of a powder, the
parameter "Coarse to Fine ratio (d90/d10)" was found to be more relevant for the models discussed
in Chapters 5 to 8. Finally, the difficulty of handling segregated powder samples having narrower particle size
distribution compared to wider size distribution has been illustrated for a pneumatic conveying
system. Based on a pilot plant study of conveying 10 blends of ash, 75% coarse ash and 25% fine ash
provided the optimal blend for dense phase conveying. A new bulk powder Froude number term
(based on loose poured bulk density) and coarse-to-fine ratio have been used to represent reliable
conveying criteria. For reliable dense-phase conveying, the ash mixture should have a bulk powder
Froude number < 10 and a coarse-to-fine ratio > 10. The significance of this study improves scientific understanding and predictive ability for segregation characteristics of Geldart group A and B borderline powders by integrating particle properties, flow characteristics, and moisture effects into highly accurate models. The introduction of novel parameters, including the modified dimensionless cohesion number and bulk powder Froude number, provides robust, generalisable tools applicable across diverse industrial contexts such as fly
ash, pharmaceuticals, detergents, construction, and food powders. The validated models deliver
practical design and operational criteria to reduce segregation, maintain product uniformity, reduce
lot rejection, and ensure reliable pneumatic conveying, thereby improving process reliability and
overall quality standards
Computational Fluid Dynamics (CFD) and Thermal Analysis for Kettle Reboiler
No full textKettle reboilers play a vital role in refinery processes, where the operation has a direct impact on overall
system performance. To enhance reliability and minimize operational issues such as liquid carryover, a
comprehensive approach involving mechanical optimization and continuous monitoring is essential. This
optimization process requires careful evaluation of several critical design and operational parameters,
including the entrainment ratio to ensure proper vapor-liquid separation, optimal positioning of inlet and outlet
nozzles to maintain flow efficiency, appropriate shell sizing to facilitate effective heat transfer, and assessment
of current operating conditions to identify potential improvements. By systematically addressing these factors,
refineries can significantly reduce liquid carryover, improve thermal efficiency, and extend the operational
lifespan of kettle reboilers, ultimately leading to more stable and cost-effective refinery operations.
ANSYS Fluent software is used for geometric modeling, meshing, thermal simulation, and post-processing.
Simulations have been done by varying the shell side diameters of the Kettle Reboiler, with the given boundary
conditions as per the problem. The physics of the problem employs a steady-state approach and utilizes a
viscous model, specifically the Realizable k-ε turbulence model, and the Eulerian phase change model.
Three simulation trials have been conducted in ANSYS Fluent to analyze the effect of reducing the kettle
diameter. The baseline model has a diameter of 1790 mm. And second trial was done with a diameter of 1750
mm, and the third trial was done with a diameter of 1690 mm. The results from the first two trials were
approximately the same, both achieving a vapor quality (dryness fraction) of about 0.997. This indicates that
a reduction of 40 mm is feasible without compromising performance. This size reduction would save
approximately 0.412 feet of sheet metal used in the kettle's fabrication. However, the third trial was conducted
on a 1690 mm diameter, resulting in a lower vapor quality of 0.97, which is not acceptable for the application.
Therefore, we conclude that the diameter cannot be reduced further than 1750 mm.
Keywords: CFD; Heat Transfer; Kettle Reboiler; Shell and Tube Heat Exchanger; Entrainment Rati
Streamlining Design Verification and Protocol Validation
No full textDesign verification is a critical phase in the development of Intellectual Property (IP) cores to ensure
they function correctly and meet the intended specifications. As IP cores are increasingly reused across
multiple systems, the need for a robust and efficient verification process has become more crucial. This
paper explores various methodologies and tools used for IP verification, including simulation-based
techniques, formal verification, and hardware-assisted approaches like FPGA prototyping. We highlight
the importance of creating a comprehensive testbench, using coverage metrics, and adopting universal
verification methodologies (UVM) to improve verification efficiency and quality. The challenges posed
by the growing complexity of IP cores and the need for scalability in verification are also discussed.
The ultimate goal is to achieve a bugfree design before integration into larger systems, reducing timeto-market and ensuring reliable performance in end products.
By addressing the evolving challenges of IP verification, this paper provides insights into best practices
and emerging trends in the field, supporting more efficient design cycles and higher
quality IP cores