NUST Institutions Library Catalogue Search for 'an:"119689"'

Smartphones as 3D Imagers: A Novel Hardware Configuration with Structure from Motion Techniques /

By SAEED, KAINAT . . 64p. ; 30cm..

Place Hold on Smartphones as 3D Imagers: A Novel Hardware Configuration with Structure from Motion Techniques /

Diagnosing and localizing Covid-19 in High resolution CT(HRCT) scans using Deep learning /

By Munir, Zonaira. . 62p. ; 30cm..

Place Hold on Diagnosing and localizing Covid-19 in High resolution CT(HRCT) scans using Deep learning /

Focus and Engagement Level Detection Using Computer Vision and Machine Learning in a Classroom Environment /

By Poonja, Hasnain Ali . . 64p. , Due to Covid 19, the global education system has changed toward online learning, which has a high dropout rate. Therefore, it is vital that students maintain their level of interest. Therefore, detection of engagement level alone is insufficient for analyzing and improving learning and teaching techniques. To promote student engagement in STEM and online learning environments, technologies such as AR/VR and Haptics should be implemented. Utilizing facial emotion, body pose, and head rotation, a web-based computer vision system is developed and implemented to identify student involvement levels using webcams during tasks such as online classrooms, haptic interaction, and augmented reality. In addition, an AR and Haptics-based World Map is being designed and developed. To evaluate and compare three types of learning scenarios, namely (1) Traditional, (2) Augmented Reality-based, and (3) Haptics-based, two methods are employed: (1) Trained Computer Vision models are tested for 3 scenarios, and (2) A user study is conducted using the Positive and Negative Affect Schedule (PANAS) Questionnaire and NASA-Task Load Index, from which conclusions are drawn. The results of a comparison of Traditional, Augmented reality, and Haptics-based learning indicate that Haptics and Augmented Reality-based learning are the most immersive and increase levels of engagement during online learning and STEM training, whereas Traditional learning methods are the least effective during online classes. User studies and computer vision models are utilized to validate the results. 30cm.

Place Hold on Focus and Engagement Level Detection Using Computer Vision and Machine Learning in a Classroom Environment /

Lung damage estimation through ground glass opacity detection from 3D reconstructed HRCT scans /

By Naeem, Abdul Hanan . . 55p. , High-resolution computed tomography (HRCT) scans have become an essential tool for the diagnosis of lung diseases, especially during the COVID-19 pandemic. However, the manual analysis of these scans by clinicians can be time-consuming and error-prone, leading to delayed diagnosis and treatment. In this thesis, we present a deep learning-based system for the automated estimation of lung damage through the detection of ground-glass opacities (GGOs) using 3D reconstructed HRCT scans. The system utilizes a MobileNetV3 backbone combined with a Lite Reduced Atrous Spatial Pyramid Pooling (LR-ASPP) segmentation head to accurately segment GGO regions in the lung. The 3D reconstruction of the scans helps to provide clinicians with a more comprehensive view of the lungs, allowing for better identification and analysis of GGOs. To train and evaluate our system, we utilized a custom dataset of HRCT scans. The results demonstrate that our system achieved high accuracy in detecting and segmenting GGO regions in the lungs, with an overall IOU of 0.62. Additionally, our system was able to provide clinicians with a more efficient method for analyzing HRCT scans, reducing the time required for diagnosis and allowing for earlier detection of lung diseases. In conclusion, our deep learning-based system provides a promising approach for the automated estimation of lung damage through GGO detection using 3D reconstructed HRCT scans. By combining state-of-the-art techniques in deep learning and medical imaging, our system can provide clinicians with an accurate and efficient method for analyzing HRCT scans, potentially leading to improved patient outcomes and reducing the burden on healthcare systems. 30cm.

Place Hold on Lung damage estimation through ground glass opacity detection from 3D reconstructed HRCT scans /

Optimal Channel Selection for Improved Classification Accuracy in fNIRS BCI /

By Mehboob ,Aakif. . 63p. ; 30cm..

Place Hold on Optimal Channel Selection for Improved Classification Accuracy in fNIRS BCI /

EEG-fNIRS based hybrid Image reconstruction and classification for BCI /

By Ehsan Mughal ,Nabeeha . . 71P. ; 30cm..

Place Hold on EEG-fNIRS based hybrid Image reconstruction and classification for BCI /

Enhanced Accuracy for Six Commands BCI System Using Hybrid ERPs/SSVEP Based Paradigm /

By NAEEM ,MEHREEN. . 67P. ; 30cm..

Place Hold on Enhanced Accuracy for Six Commands BCI System Using Hybrid ERPs/SSVEP Based Paradigm /

Enhanced Accuracy for Six Commands BCI System Using Hybrid ERPs/SSVEP Based Paradigm /

By NAEEM ,MEHREEN. . 67P. ; 30cm..

Place Hold on Enhanced Accuracy for Six Commands BCI System Using Hybrid ERPs/SSVEP Based Paradigm /

Genetic Algorithm based Optimal Feature Selection for Hybrid EEG-EOG, EEG-EMG and EEG-fNIRS for BCI /

By MATEEN ,NIDA . . 132p. ; 30cm..

Place Hold on Genetic Algorithm based Optimal Feature Selection for Hybrid EEG-EOG, EEG-EMG and EEG-fNIRS for BCI /

Pakistani Traffic-sign detection using Deep Learning /

By Nadeem, Zain . . 56p. , Traffic-sign Detection is one of the major aspects of the working of a modern car, more so in the case of a self-driving car. They need to be detected and recognized up to a certain degree of accuracy. This research revolved around the detection of Pakistani Traffic-sigs. The research was conducted in 3 phases; firstly, a fixed camera was used to collect video feed from real-world car rides. These videos were then extracted and manually annotated using pertinent software tools. This helped create a dataset of images of traffic signs which was important as the model being used is Deep Learning-based, which require colossal amounts of data to function properly. Secondly, this data is used to train a Deep Learning model to detect and classify the type of traffic sign. The trained model produced a mean average precision (mAP) of 75.636% on the training dataset and 49.699% on the validation dataset while the mAP stood at 43.453% for the test dataset. All these results are state-of-the-art and strong enough for implementation as real-world models. The model was cross-validated and regularized to help improve the model’s working. The final model was tested in real-world scenarios and tweaked according to requirements. cl 30cm.

Place Hold on Pakistani Traffic-sign detection using Deep Learning /

Food Quality Assessment Based on Deep Learning Models /

By Sandhu, Maryum. . 61p. , Objective. In this paper, a novel dataset has been collected in accordance with Pakistani needs and is used to develop an architecture for the quality assessment of fruits and vegetables. Approach. The dataset contains images captured under uncontrolled conditions with respect to illumination, temperature, humidity, image resolution, image aspect ratio, angle of capturing images and background. Images captured contain items individually as well as in groups. To the best of the knowledge gathered, this is the first of its kind dataset. This dataset is then preprocessed. Among usual preprocessing techniques, an aspect ratio adjustment algorithm has been introduced. After preprocessing, the data is used to train multiple models (AlexNet, VGG-16, ResNet-50, Fruits-360 Model and a proposed model with relatively lesser depth). This performs recognition of fruits and vegetables and endorse the validity of the dataset. Going further, the dataset is then prepared for quality assessment with three quality labels for each fruit/vegetable: Eatable, Partially Rotten and Rotten. Quality assessment is then performed using pre-trained VGG16 through transfer learning, adding a fully connected network and fine-tuning the model. Main Results. The highest recognition accuracy on the validation set is 98.9% and the highest validation accuracy for quality assessment is 92.9%. Significance. Outcomes of this research demonstrate that dataset collected under an uncontrolled environment can be used for recognition of fruits/vegetables with remarkable accuracies. Moreover, quality assessment of fruits/vegetables is performed accurately with the same dataset using deep learning and three quality labels. 30cm.

Place Hold on Food Quality Assessment Based on Deep Learning Models /

Enhanced Drone Control Using Reinforcement Learning /

By Moin, Hassan . . 101p. , Quadcopters have already proven their effectiveness in both civilian and military applications. Their control, however, is a difficult task due to their under-actuated, highly nonlinear, and coupled dynamics. Most quadcopter autopilot systems utilize cascaded control schemes, where the outer loop handles mission-level objectives in 3D Euclidean space, and the inner loop is responsible for stability and control. Such complex systems are generally operated using PID controllers, which have demonstrated exceptional performance in multiple scenarios, such as obstacle avoidance, trajectory tracking and path planning. However, tuning their gains for nonlinear systems using heuristics or rulebased methods is a tedious, time-consuming and difficult task. Rapid advances in the field of computational engineering, on the other hand, have paved way for intelligent flight control systems, which have become an important area of study addressing the limits of PID control, most recently through the application of reinforcement learning (RL). In this dissertation, an optimal gain auto-tuning strategy is implemented for altitude, attitude, and position controllers of a 6 DoF nonlinear drone system using a deep actor-critic RL algorithm having continuous observation and action spaces. The state equations are derived using Lagrange’s (energy-based) method, while the drone’s aerodynamic coefficients are estimated numerically using blade element momentum theory. Furthermore, the cascaded closed loop system’s asymptotic stability is studied using the theory of Lyapunov. Finally, the proposed strategy is validated by simulation results, where the gains learned by RL agents allow the quadcopter to track a given trajectory accurately. Moreover, these optimal gains satisfy the conditions obtained through Lyapunov’s stability analysis, indicating that the RL algorithm is an extremely powerful tool which can assess uncertainties existing within any complex nonlinear system 30cm.

Place Hold on Enhanced Drone Control Using Reinforcement Learning /

Random Filter-Switching-based Defense Against Decision-based Adversarial Attacks on Machine Learning /

By Khalid, Rashad . . 69p. , In the AI and machine learning research field, adversarial machine learning(AML), a technique that tries to deceive models using erroneous data, is becoming a major concern. By exploiting the inherent vulnerability of ML models’ data reliance, AML can be used to generate adversarial attacks. Researches have shown that a small perturbation in input image can create disastrous results for an autonomous car system e.g. miscalssifying stop sign as speed limit sign near school. To counter these adversarial attacks, several defense mechanisms have been proposed. Some of the most prominent defenses are adversarial training, pre-processing-based defenses, Generative Adversarial Networkbased defenses. However, most of these defenses are either computationally expensive or become in-effective under the white-box threat model or against the decision-based attacks (Adversarial attacks that exploit the final decision of the attack under black-box settings). Therefore, there is a dire need to develop efficient defense mechanisms that can effectively counter the attacks while maintaining the classification accuracy. In this thesis, we propose to develop a computationally efficient and effective defense mechanism that effectively counters the score-based and decision-based adversarial attack under black-box settings while maintaining the classification accuracy on clean images. 30cm.

Place Hold on Random Filter-Switching-based Defense Against Decision-based Adversarial Attacks on Machine Learning /

An autonomous car crash prevention system based on behavioral assessment of driver /

By Arif, Mahad . . 63p. , Many road-side accidents occur due to the driver being not in the emotional state of driving. i.e., the driver is fatigued or is not alert. Computer Vision is one of the widely used fields in the world right now. The amount of work being carried out in this field is enormous and very helpful as well. One of such works is detecting human mood at any given time by analyzing the facial expressions of that person. The mood can be of these types. i.e., Alert, Fatigued, Happy, Sad, Drowsy etc. The "OpenCV" open-source Computer Vision library makes it possible to analyze facial expressions. In this thesis, different behavioral assessments are made on a car driver’s video recordings to detect drowsiness. These behavioral assessments include Eye Blinks detection, Yawning Detection, Percentage Eye Closure (PERCLOS) and Pose Estimation. All these are ensembled together to give a more accurate prediction of a driver being drowsy. It was concluded that the number of false positives increase during night-time and thus the accuracy of the system goes down when the lighting conditions are low. Also, camera for driver’s video recording should be placed just behind the left of steering wheel for maximum number of true detections. The system also works in realtime thus making it more useful. 30cm.

Place Hold on An autonomous car crash prevention system based on behavioral assessment of driver /

Adaptive Hemodynamic Signal Estimation Using Kalman Estimator /

By Quddusi, Basiq Warrad . . 80p. , Functional Near Infrared Spectroscopy (fNIRS) is a technology that measures changes in the oxygenation level of blood present in brain whenever an activity is performed by human being. It is a non-invasive technique and uses near infrared light to detect changes in the concentration of two chromophores i.e., oxygenated and deoxygenated haemoglobin. During the recording, information related to neural activity in fNIRS signals gets compromised. This is due to the interference of noises from the environment outside as well as inside the human body. External noises can be light in the room and powerline noise. Internal noises are physiological noises such as cardiac, respiratory and mayer waves. Therefore, during analysis, it is required to remove these noises first and then extract main activity signal i.e., hemodynamic signal. Many techniques and methods have been proposed and practiced up to this date. Among them the most popular technique is General Linear Modelling (GLM). GLM models the signal by breaking it down into sum of all components present in the signal along with an error term. Previous studies and research that have used GLM for the reconstruction of activity signal used single frequency value for each noise but in reality, the frequency for each noise varies with the level of activity performed by the subject. This can lead to less accurate reconstruction of activity signal. In this study, this problem is kept under consideration and a method is developed to keep account for all the values of frequency that can corrupt fNIRS signal. Ranges of frequencies are considered instead of single values. These frequency ranges are first extracted using Continuous Wavelet Transform (CWT) and their possible magnitudes are estimated using Kalman filter. Similarly, activity signal is extracted from fNIRS signal using Discrete Wavelet Transform (DWT) and then its magnitude is estimated using Kalman filter. Output of these two steps is fed to GLM for reconstruction of possible hemodynamic signal. Results from this method are compared with the results of conventional GLM and significant improvement is observed both visually and statistically. 30cm,.

Place Hold on Adaptive Hemodynamic Signal Estimation Using Kalman Estimator /

Disturbance Rejection and Roll Over Estimation for Control of Non-Linear Robotic System/

By Malik, Kamal Mazhar. . 216,p; 30,cm..

Place Hold on Disturbance Rejection and Roll Over Estimation for Control of Non-Linear Robotic System/

A Fabric-Based Soft Robotic Glove for Enhancing Hand Function /

By Saleem, Muhammad Anas . . 87p. , The goal of this research project is to show how to design, build, and test a soft robotic glove that can be used in real life to help people with hand injuries. The glove uses both hybrid control mechanisms and pneumatic actuation. The device is meant to help patients who can't move their hands very well, such those who are recovering from a stroke, by giving them both passive and active movement support. The main part of the gadget is five custom-made pneumatic bladder actuators. These actuators are made of cow leather, elastic sleeves, and big latex balloons. When you fill them up, they move like fingers do naturally. The electronic glove has two modes: Auto Mode, which lets the user run programmable treatment cycles, and EMG Mode, which lets the user control the glove based on realtime surface electromyography (sEMG) data. One of the most important new aspects of the system is the addition of an industrial-grade pressure sensor that can measure between 4 and 20 milliamperes. The HW-685 current-to-voltage converter connects to this pressure sensor. This makes sure that the pressure feedback is clear and free of noise. This method solves a typical challenge when it comes to academic prototypes. You can utilize a touchscreen interface with visuals made just for the glove to control it. You can see live EMG charts, get real-time status updates, and make several changes to the settings with this interface. There are other safety measures in place, such as the software-based emergency stop logic and controlled depressurization operations. The full experimental validation was done with people who were in good health. A calibrated WIKA master gauge was used for the pressure accuracy test, and the findings showed that the pressure stayed the same up to 600 kPa. The error was less than three percent, and the overshoot was not very big. We used a Jamar digital dynamometer to do the testing, and the results showed that the peak grip forces could reach a maximum of 9.3 kilograms. Compared to the established benchmark values for functional grab xvi support, our results were either like or better than those values. We used a goniometer to measure the range of motion, and the results showed that the angles of finger flexion can reach up to fifty degrees, which is very close to how the hand typically moves. The results of this study show that it is possible to create a soft robotic system that is both flexible and cheap, and that may be used in hospitals. Because it has reliable sensing, adaptive control, and an easy-to-use interface, it would be great for use in clinical or home rehabilitation settings. A translation like this would be very helpful for both patients and therapists in many ways. 30cm.

Place Hold on A Fabric-Based Soft Robotic Glove for Enhancing Hand Function /