NUST Institutions Library Catalogue Search for 'an:"119487"'

Detection of Lung Diseases through Medical Imaging using Deep Learning /

By Ahmed, Mudassar . . 35p. 30cm.

Place Hold on Detection of Lung Diseases through Medical Imaging using Deep Learning /

Enhancing Tumor Segmentation in 3D MRI Scans with YOLOv8 And Volumetric segmentation: A Computational Efficient Approach /

By Abdullah. . 71p. ; 30cm..

Place Hold on Enhancing Tumor Segmentation in 3D MRI Scans with YOLOv8 And Volumetric segmentation: A Computational Efficient Approach /

Early Detection & Stage Classification of Parkinson’s Disease using Deep Learning /

By Zeeshan, Muhammad Muzzamil . . 38p. ; 30cm..

Place Hold on Early Detection & Stage Classification of Parkinson’s Disease using Deep Learning /

Plug and play: nnUNet for the Auto-Segmentation of Head & Neck Tumors and Lymph Nodes on 3D FDG PET/CT Scans /

By Basharat, Sonia . . 35p. 30cm.

Place Hold on Plug and play: nnUNet for the Auto-Segmentation of Head & Neck Tumors and Lymph Nodes on 3D FDG PET/CT Scans /

Brain Tumor Segmentation using CT Hybrid

By Siddiqah, Mariyam . . 30p. 30cm.

Place Hold on Brain Tumor Segmentation using CT Hybrid

Multimodal 3D-MRI Brain Tumor Segmentation via Modified 3D U-Net with Transformer /

By Khan, Hussain Nasir . . 61p. 30cm.

Place Hold on Multimodal 3D-MRI Brain Tumor Segmentation via Modified 3D U-Net with Transformer /

Apple Leaf Diseases Detection and Classification using Deep Learning /

By Ullah, Wasi. . 42p. ; 30cm..

Place Hold on Apple Leaf Diseases Detection and Classification using Deep Learning /

Design and Development of an Indigenous Mobile Robot to Navigate in Cluttered Environment /

By Ali Shah, Muhammad Soleman . . 64p. , The field of smart autonomous systems has experienced significant growth in recent years, with the development of robots aimed at assisting humans in various tasks. In particular, autonomous manipulators have been designed for disaster management and other situations where humans are inaccessible. This master's thesis presents the design and development of a mobile manipulator that can autonomously move in cluttered environments and perform pick and place tasks using 2D SLAM on ROS and 3D camera-based object detection. The proposed solution addresses the SLAM problem by utilizing the gmapping SLAM algorithm, which allows the robot to simultaneously locate itself and map its surroundings. The robot is equipped with a custom-made rover and a 6-DOF robotic arm assembled from ready-made links with joint servos. The arm is used to perform the pick and place tasks, and the 3D camera is used to estimate the coordinates of the targeted object, which is then used to control the robotic arm using inverse kinematics. The localization of the robot is done through 2D pose estimation using Kalman filter, and the destination position is set via RVIZ. The robot is designed to operate in indoor environments and can navigate autonomously using the 2D SLAM technique. The project demonstrates that the robot is capable of detecting the target object's 3D pose, estimating its coordinates, and accurately moving the robotic arm to achieve the desired pick and place task. Real experiments and demonstrations of the mobile manipulator's capabilities were performed using two Arduinos, one controlling the rover's motor and the other controlling the robotic arm's servos. The results of the experiments confirm the robot's ability to move autonomously and perform pick and place tasks accurately and efficiently. Overall, the mobile manipulator designed in this thesis provides a reliable solution for assisting humans in disaster management scenarios and other inaccessible environments. The use of 2D SLAM, 3D camera-based object detection, and inverse kinematics control for the robotic arm ensures efficient and accurate navigation and pick and place operations. The project can be extended to more challenging environments, such as outdoor and unstructured environments, with the integration of advanced sensors and algorithms. 30cm.

Place Hold on Design and Development of an Indigenous Mobile Robot to Navigate in Cluttered Environment /

Automatic Detection and Recognition of Citrus Fruits Diseases Using Deep Learning Model /

By Faisal, Shah . . 59p. , In a country's economy, agriculture plays a very vital role. Agriculture's yield and production are reduced by plant diseases, resulting in significant economic losses and instability in the food market. In plants, the citrus fruit crop is one of the most important agricultural products in the world, produced and grown in around 140 countries. It has a lot of nutrients, such as vitamin C. However, due to pests and diseases, citrus cultivation is widely affected and suffers significant losses in both yield and quality. The majority of plant diseases exhibit visible symptoms, and the accepted method used today is for a skilled plant pathologist to detect the diseases by examining affected plant leaves under a microscope, which is a costly and time-consuming method. During the last decade, computer vision and machine learning have been widely adopted to detect and classify plant diseases, providing opportunities for early disease detection and bringing improvements to agricultural production. The early detection and accurate diagnosis of plant diseases are essential for reducing their spread and damage to crops. In this work, we presented an automatic system for early detection and recognition of citrus plant diseases based on a deep learning (DL) model to improve accuracy and reduce computational complexity. The most recent transfer learning-based models were applied to our dataset in order to increase classification accuracy. In this work, we successfully proposed a CNN-based pre-trained model (EfficientNetB3, ResNet50, MobiNetV2, (InceptionV3) for the identification and classification of citrus plant diseases using transfer learning. In order to assess the performance of the model, we found that the transfer of an EfficientNetb3 model led to the highest training, validating, and testing accuracies, which were 99.43%, 99.48%, and 99.58%, respectively. The proposed CNN model exceeds other cutting-edge CNN network architectures developed in earlier literature in the identification and categorization of citrus plant diseases. 30cm.

Place Hold on Automatic Detection and Recognition of Citrus Fruits Diseases Using Deep Learning Model /

Enhanced Accuracy for Motor Imagery Detection Using Deep Learning for BCI /

By Sarwar ,Ayesha . . 49P. ; 30cm..

Place Hold on Enhanced Accuracy for Motor Imagery Detection Using Deep Learning for BCI /

Integration of Continuous Wavelet Transform and Convolutional Neural Network for multiclass EEG dataset classification /

By NASEEM ,SIDRA . . 58p. ; 30cm..

Place Hold on Integration of Continuous Wavelet Transform and Convolutional Neural Network for multiclass EEG dataset classification /

Design and Control of Flapping Wing Unmanned Arial Vehicle Mimicking Falcon /

By ALI ,HAIDER . . 49p. ; 30cm..

Place Hold on Design and Control of Flapping Wing Unmanned Arial Vehicle Mimicking Falcon /

Traffic Detection for Advanced Driver Assistance System /

By Nadeem, Hamza . . 44p. , The Advanced Driver Assistance System (ADAS) is not a new phenomenon. To minimize road accidents and other related issues, the current vehicles can be improved for a better driving experience through an automated system that assists the driver. Some of the basic elements that such ADAS systems utilize include, but are not limited to, sensing the environment, traffic signs, pedestrians, and other vehicles. The need for traffic to be detected and recognized up to a certain degree of accuracy arises due to our objective i.e., to ensure that the car and the passengers in it are safe. Traditional Image Processing techniques have previously been used which are way slower. Recently, CNNs have been deployed heavily in Traffic detection and identification. However, CNNs do require a huge number of input images to work efficiently, and no such traffic recognition dataset exists in Pakistan. In this research, we deployed a YOLOv7 based architecture trained on a self-collected and manually annotated Pakistani Traffic Type and Sign Recognition Dataset (PTSD) to detect and classify the types of traffic. The Deep Learning model was trained and tested to produce a mean average precision (mAP) of 87.20%. These results are state-of-the-art and strong enough for implementation as real-world models. The model was further tuned to help improve the model’s working, and then tested in real-world scenarios. The final model was used to develop an ADAS Unit— which works on a priority-based decision system, providing specified instructions for the detected conditions. 30cm.

Place Hold on Traffic Detection for Advanced Driver Assistance System /

Development of High-Resolution Imaging System for Blood Activity Monitoring /

By Qasim, Syed Muhammad Ali . . 46p. , fNIRS is a non-invasive, portable & easy to use brain imaging modality. It can estimate the hemodynamic response of the brain by measuring the absorption of IR light with respect to time. The standard channel separation between source & detector in fNIRS is 3cm but this distance has a disadvantage of higher channel noise. To resolve the issue, we have presented a new fNIRS design with small channel separation to minimize the channel noise. In this research, we have designed the fNIRS device using 2 sources & 14 detectors in a circular configuration. The detectors are placed in two circles each circle having 7 LEDs with a radius of 1.5cm & 2.25cm respectively. After the software design, we implemented it on a hardware & tested the device using occlusion. Once the device got tested through occlusion, we acquired the brain signals by placing it on the left frontal cortex. After placing the device on the frontal cortex, we reverse counted for 200sec with rest & count intervals. The whole experimental design is described in the sections below. We applied Modified Beer Lambert Law (MBLL) to deduce results. The results proved to be promising as the channel noise reduced & we got better signals. The results can be improved further by using the high-power IR LEDs having better penetrating ability 30cm.

Place Hold on Development of High-Resolution Imaging System for Blood Activity Monitoring /