Department of Computer Science

Permanent URI for this communityhttps://repository.cuilahore.edu.pk/handle/123456789/16

Browse

Author: search.filters.author.Dr. Usama Ijaz BajwaSubject: search.filters.subject.Information technology

Search Results

Now showing 1 - 8 of 8
  • No Thumbnail Available
    Item
    GAME FOR DYSLEXICS
    (Library Information Services, COMSATS University Islamabad, Lahore Campus., 2020-11-20) Noor E Eman; FA16-BCS-286; Dr. Usama Ijaz Bajwa; LHR TP 6120
    The present project will be an empirical endeavour for the development of a reliable and valid diagnostic and rehabilitation tool for the children with reading disorder. It is seen that the activity of diagnosis and rehabilitation is boring for the children, consequently being less effective for them. Therefore, to devise a technique through which children learn more is the focus of this project. Children now a days show keen interest in computer games. Due to the proposed tool, which is a series of games, children will be able to differentiate between different sounds, words and letters. This would be done by different techniques and creating different scenarios and levels. This report presents our serious games created to screen and train different abilities which are usually impaired in children with dyslexia. We primarily test the cognitive abilities which are deeply related to the children’s future reading skills, i.e., visual search ability, rapid identification of visual inputs, visual sequential and auditory memory and the capability to associate visual and auditory stimuli. To this aim, we discuss here a series of serious games designed to train specific skills that have been proven to be effective against dyslexia according to Orton Gallium Study. This game would be a good platform for rehabilitation along with providing the results of performance at the end of game. The game would be developed using unity 3D engine, which provides variety of tools for game creation.
  • No Thumbnail Available
    Item
    Human Action Recognition in Low-Resolution Videos Based on Spatio-Temporal Features
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2023) Samia Akram; FA21-RCS-015; LHR TP 8678; Dr. Usama Ijaz Bajwa
    In recent years, the field of computer vision and pattern recognition has devoted a lot of research focus to the study of human action. Most of the current action recognition research focuses on high-quality videos with clearly apparent actions. Most actions are of limited quality and take place at a distance, making it difficult to identify them. Therefore, the issue of low video quality is still under-researched and difficult to solve in practical implementation. The goal of this study is to establish a deep learning based framework for recognizing human actions in low-quality video by utilizing spatial-temporal features. This research focuses on the use of annotated dataset TinyVIRAT-v2 of action recognition videos recorded in low quality in order to examine the applicability of deep learning architecture for action recognition from videos of low perceptual quality. First, in the proposed framework datasets will be pre-processed and spatial-temporal features will be retrieved using a CNN-based feature extractor. The extracted features will then be fed into a 3D custom model for classification, which is based on resnet50 as backbone network and c3d with a SoftMax layer for multiclass prediction, allowing for its implementation as a real-world practical application. Upon classifying, every testing video is allocated a predicted category, and the cumulative outcomes are evaluated using performance measure F1-Score which is 0.68.
  • No Thumbnail Available
    Item
    Multiclass Weeds and Crops Classification and Segmentation using Deep Learning
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2023) Seemab Ayub; FA21-RCS-009; LHR TP 8675; Dr. Usama Ijaz Bajwa
    The world's population has increased exponentially with the increase in food demand, so it's essential to focus on factors that affect the quality and quantity of food. Some basic natural factors are climate, soil, pests, and weeds. Weeds are unwanted plants in yields that compete with the crops and consume a lot of nutrients that affect the quality and growth of crops. The separation of these toxic weeds is a challenging task because of their color and quantity. Pakistan is an agricultural country, and almost a big part of its economy depends on agriculture. The growth of this department does not match the population growth; the reason behind this is the presence of weeds. Automated technological interventions represent the optimal approach for weed removal. However, notable challenges persist within the technological landscape, including the absence of comprehensive datasets and the need for well-automated systems specifically tailored for crop weed classification and segmentation. Focusing on these issues, most of the researchers are working on different techniques, but the inadequate size of the dataset is still a problem for researchers, so this study performed to generate an efficient large dataset by merging different small datasets on the base to pick only those classes of weed and crops that belong to Pakistan; also an effective system using UNet variations (UNet, Attention UNet, and Attention Residual UNet(ARUNet)) and ResNet where ARUNet has performed very well for segmentation process and especially the Attention mechanism boosted the segmentation process, on the other hand ResNet has performed best for classification process on our dataset (24- classes and 120000 images). The accuracy of 99% achieved for classification process and 95% for segmentation.
  • No Thumbnail Available
    Item
    A Robust Attention-based 3D CNN Model for Fire and Smoke Detection from Videos
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2023) Rimsha Shoukat; FA21-RCS-014; LHR TP 8677; Dr. Usama Ijaz Bajwa
    The unforeseen occurrence of a fire eruption has significant and adverse impacts on all aspects of human existence, including human life, land, industry, forests, and animal populations. The emergence of smoke before the apparent signs of fire makes smoke detection very advantageous in terms of promptly identifying fires. This early detection capability holds significant potential for preserving numerous lives and mitigating the occurrence of catastrophic events and associated losses. The problem of detecting smoke and fire using video-based methods presents significant challenges due to the considerable variability in their color, brightness, and shape. Most prior researchers conducted individual studies focused on the categorization of fire or smoke. Furthermore, video classification studies focused on a 2D approach, which was inadequate due to its inability to capture temporal information, so it was not able to learn the difference between frames. The convolutional neural network (CNN) has demonstrated exceptional performance in several domains, hence establishing itself as the leading approach. This study presents a robust classifier based on a three-dimensional convolutional neural network (3D-CNN) specifically developed to classify events into three distinct classes: fire, smoke, and neutral. The classifier is meant to effectively process video data. 3D models have the capability to acquire spatial and temporal characteristics from video sequences, rendering them appropriate for undertaking such tasks. The proposed 3D-CNN model had a notable test accuracy of 90% and showcased outstanding performance with an accuracy of 96% when subject to cross data [1] that have seven smoke videos. To enhance its capabilities, the model integrates the CBAM (Convolutional Block Attention Module) attention mechanism. The utilization of this approach allows the model to concentrate on significant regions of interest (ROI) throughout the entirety of the frame, hence enhancing its precision in categorizing complex attributes inside video frames. By integrating the attention mechanism, the model attained a remarkable gain of 6% in test accuracy with an accuracy rate of 96% and an F1-score of 98%.
  • No Thumbnail Available
    Item
    Deep Learning-Based Prediction of SARS-CoV 2 (COVID-19) and its Severity Classification using Multimodal Chest Radiography Images
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2021) Mughees Ahmad; SP20-RCS-019; LHR TP 7595; Dr. Usama Ijaz Bajwa
    The new COVID-19 emerged in a town in China named Wuhan in December 2019 and since then this deadly virus transmitted all over the world. This transmittable disease has infected 262 million people worldwide and 5.2 million deaths till November 2021. As we can see the rapidly spreading of this pandemic, different countries are facing limited resources such as medical test kits and ventilators because of the number of positive cases that have been increased out of control. In this devastating situation, it is very compulsory to develop an easily available, low-priced, and automatic deep learning model for COVID-19 prediction through chest radiography images (CRIs) such as X-rays, and CTs. The proposed study is using chest radiography images to detect chest infections such as bacterial, viral, and COVID-19 infection as X-rays and CTs give significant information about this deadly infection. The recent hybrid deep learning (DL) techniques can be used along with chest radiography images (CRIs) for the faster and accurate detection of different chest infections including COVID-19 and its severity levels (i.e., negative for pneumonia, atypical appearance, indeterminate appearance, and typical appearance). Therefore, a novel hybrid model named Lightweight Residual_Bi-GRU uses residual blocks and Bi-directional gated recurrent unit (Bi-GRU) for automatic and correct detection of non-COVID and COVID-19 infections by using preprocessed chest radiography images (CRIs). Lightweight Residual_Bi-GRU is used for the recognition of two-class classification (normal and COVID-19), three-class classification (normal vs COVID-19 vs viral pneumonia), four class classification (normal vs COVID-19 vs viral pneumonia vs bacterial pneumonia), and different COVID-19 severity types' classification. The presented model provides a classification accuracy of 99.5%, 98.4%, 90.2%, and 80.7% for 2 class, 3 class, 4 class, and COVID-19 severity levels classifications. The results prove that radiologists and medical officers can adopt this method for the screening of chest infections where test kits are limited.
  • No Thumbnail Available
    Item
    A Robust Multi-Camera Deep Person Re Identification Framework Using Spatiotemporal Context Modelling
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2021) Fatima Zulfiqar; SP20-RCS-005; LHR TP 7601; Dr. Usama Ijaz Bajwa
    In this thesis, research work is based on a robust automatic re-identification of a person from multiple non-overlapping cameras under variable and dynamic environmental conditions for an accurate re-identification and retrieval of targeted person identities. Person Re-Identification (ReID) aims at identifying query person of interest (POI) assigned with a unique identity label across multiple non-overlapping cameras. The query POI can be either an image or a video sequence. Person ReID has gained quite increasing attention among various research and developer communities in recent years. Several research challenges including occlusion, variable viewpoint, misalignment, unrestrained poses, background clutter, etc. are the major challenges in developing robust, lightweight, end-to-end trainable person ReID models. To address these issues, an attention mechanism that comprises local part/region aggregated feature representation learning is presented in this research by incorporating long-range local and global context modeling. The part-aware local attention blocks are aggregated into the widely used modified pre-trained ResNet50 CNN architecture as a backbone employing two attention blocks i.e. Spatio-Temporal Attention Module (STAM) and Channel Attention Module (CAM) thus improving both local and global feature representation learning. The spatial Attention block of STAM can learn contextual dependencies between different human body parts regions like head, upper body, lower body, and shoes from a single frame. On the other hand, the temporal attention modality is capable to learn temporal contextual dependencies of the same person’s body parts across all video frames. Lastly, the channel-based attention modality i.e. CAM can model semantic connections between the channels of feature maps. These STAM and CAM blocks are combined sequentially from a unified attention network named Spatio-Temporal Channel Attention Network (STCANet) that will be able to learn both short-range and long-range global feature maps respectively. Extensive experiments are carried out to study the effectiveness of STCANet on three images and two video-based benchmark datasets i.e. Market- x 1501, DukeMTMC-ReID, MSMT17, DukeMTC-VideoReID, and MARS. K reciprocal re-ranking of gallery set is also applied in which the proposed network showed significant improvement over these datasets in comparison to the state-of-the-art by achieving (mAP/Rank-1) score of (95.5/94.5), (90.7/92.3) and, (74.4/84.5) on Market-1501, DukeMTMC-ReID, and MSMT17 dataset respectively. In addition, the proposed modified STCANet also showed significant performance improvement in comparison to state-of-the-art methods by achieving (mAP/Rank-1) score of (96.6/97.1), (85.3.7/89.1) on DukeMTMC-VideoReID and MARS dataset respectively Lastly, to study the generalizability of STCANet on unseen test instances, cross-validation on external cohorts is also applied that showed the robustness of the proposed model. The proposed STCANet is lightweight, end-to-end trainable, and can be easily deployed to the real world for practical applications
  • No Thumbnail Available
    Item
    Detection of Fire and Smoke from Video Sequences
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2020) Ali Hussain; SP18-RCS-003; LHR TP 7280; Dr. Usama Ijaz Bajwa
    Since the introduction of deep neural networks in object detection, fire and smoke has been in the focus of many researchers. The recent state of the art Convolution Neural Network (CNN) based architectures provide more than 95% fire-smoke detection accuracy but in controlled environments, e.g., a fire in server rooms or production lines. Fire burns differently in certain environments; wildfires and domestic fire have distinctive characteristics and burning patterns. A fire detection system should be strong enough to better generalize different fire burning patterns. Training such a system needs a massive amount of annotated data describing unique fire patterns. In this research, a deep neural network-based fire and smoke detection system will be proposed. This network's primary focus will be consuming less training data for better generalization to achieve high accuracy with a low false-positive rate. The results have been reported using the ROC, accuracy, false positive rates
  • No Thumbnail Available
    Item
    Deception Detection using Facial Action Coding System in Videos
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2020) Hammad-ud-din Ahmed; SP17-RCS-015; LHR TP 5974; Dr. Usama Ijaz Bajwa
    Humans lie to each other on a regular basis in one form or another. To lie to someone is to share altered facts. Facts are important for decision making in every situation. Without unaltered facts, potentially harmful decisions can be made which can alter someone’s life in ways that may not want. This is why it is important to catch misinformation before any harm can be caused by them. Deception detection in videos has gained traction in recent times for its various real-life applications. When a person lies, they show facial expressions, also known as microexpressions so fast that they can be ignored by the untrained eye. Microexpressions can be used as a basis for creating a deception detection system. Facial Action Coding System is utilized as a way to encode and extract data from facial muscle movement during truthful and deceptive confessions. This data is used to train a deep learning model that utilizes long short-term memory (LSTM) to train and create the system. The real-life trial dataset is used to train and test the system provided one of the best facial only approaches to deception detection. Cross-data validation is also tested using the Real-life trial dataset, the Silesian Deception Database, and the Bag-of-lies Deception Database. Cross-data validation is something that has not yet been attempted by anyone else for a deception detection system which provides unique insight for the field. The results show that adding different datasets to train a neural network for the sake of creating a deception detection system worsens the accuracy of the system but the reasoning behind the poorer results actually encourages attempts at creating better datasets