Department of Computer Science

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

Browse

Author: search.filters.author.Dr. Zeeshan GillaniSubject: search.filters.subject.Information technology

Search Results

Now showing 1 - 10 of 15
  • No Thumbnail Available
    Item
    Wildfire Detection using Satellite Imagery
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2025) Faisal Saqib; FA22-RCS-007; Dr. Zeeshan Gillani; LHR TP 9494
    This thesis presents an innovative approach to wildfire detection using satellite imagery, aimed at enhancing detection accuracy and robustness through the fine-tuning of pre trained state-of-the-art deep learning models. Utilizing a dedicated wildfire satellite imagery dataset, our research focused on optimizing these models to better generalize across diverse environmental conditions encountered in satellite data. Key evaluation metrics employed in this study included Dice Loss, Intersection over Union (IoU), and Accuracy, which facilitated comprehensive performance assessments. Our method involved adjusting existing model architectures to improve their efficacy in segmenting and identifying wildfire-affected areas accurately. The results of our experiments indicate significant improvements in model performance, with increased precision in detecting and delineating wildfire boundaries. These findings not only demonstrate the potential of advanced computational techniques in the field of remote sensing for wildfire detection but also provide valuable insights for future applications in environmental monitoring and disaster management. This work underscores the effectiveness of tailored deep learning solutions in enhancing the capabilities of satellite imagery analysis for critical real-world challenges.
  • No Thumbnail Available
    Item
    Vehicle Tracking System
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2021) Rao Mubashar Ali , Ali Suhaib Shahid , Ammar Yasee; FA17-BCS-124 , FA17-BCS-082 , FA17-BCS-033; Dr. Zeeshan Gillani; LHR TP 7519
    The increase in population in metropolitan cities has given rise to severe traffic management problems and security issues. The rise in CCTV (Closed-circuit television) based solution has enabled us to monitor traffic but the sheer number of the cameras has given rise to another challenge of monitoring these systems and analyse this rich information. The advancement in computer vision techniques and GPU (Graphical processing unit) has now enable techniques than can automatically task which was only possible by humans in the past. CNN (Convolutional neural networks) based models can enable us to use real time videos of traffic data to aid in analysing the traffic conditions which in turn helps us in effective decision making regarding our cities to make them intelligent and safe. We will employ futuristic algorithms on the NVIDIA AI city datasets to evaluate the traffic detection and tracking systems.
  • No Thumbnail Available
    Item
    Artistify – Artist Identification of an Artwork using CNN
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2021) Usama Anwar Butt; SP17-BCS-128; Dr. Zeeshan Gillani; LHR TP 7169
    Identifying the artist of the painting is a particularly difficult problem if there are no other metadata concerning the accessible image. Historians and Librarians in the past used to catalogue of notable artworks and used to debate over possible painters that might have originally painted the piece. This is a topic of active research in the computer vision field because of the challenging nature of the problem. We will apply a class of deep neural networks which is a Convolution Neural Network to analysing visual imagery. Using a dataset that contains artwork from the most creative artists of all time like Andy Warhol and Vincent Van Gogh, my network will learn the pattern and characteristics of the art and classify the given artwork to best predict the likely artist. Our goal is to create a model that learns to identify the artist analyzing new pictures.
  • No Thumbnail Available
    Item
    Plant Disease Phenotype Identification and Classification via Deep Learning
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2021) Haider Ali Khichi , Muhammad Abdullah Aqib; SP17-BCS-038 , SP17-BCS-062; Dr. Zeeshan Gillani; LHR TP 7163
    Humans face global food shortages in the upcoming years, and we must maximize the yield of common crops to feed the growing population. Plant diseases are a major threat to small and large farm owners alike. These diseases reduce the potential yield of the crops, and in some severe but frequent cases, up to 100%. The identification of these diseases remains a challenge despite government efforts to educate farmers. There is a need for identification of these diseases early on so that they can be treated quickly. The pervasiveness of smartphones among farmers around the world offers the potential of adopting recent technological developments in computer science to develop a solution that can help in plant disease phenotype identification. This rich ecosystem of diverse communities can be a great advantage to crop heath and consequently the human race. To utilize the potential of these technologies we use a dataset curated by PlantVillage that contains over 50,000 images of 14 different crops that can be utilized to perform disease identification using modern deep learning algorithms. The models can then be employed by smartphones to identify the diseases and recommend possible treatments
  • No Thumbnail Available
    Item
    Deep learning approach for mango varieties identification using UAV imagery
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2021) ANISH HASSAN; SP19-BCS-009; Dr. Zeeshan Gillani
    In this study, we propose a deep learning approach for the detection, variety identification, and health assessment of mango trees using UAV imagery. We used a DJI P4 Multispectral camera to collect images of mango orchards which were then used to train YOLOv5, YOLOv7, and Detectron2, three state-of-the-art object detection models. Our models were able to accurately detect trees in an image and classify the variety of the mango tree, as well as identify whether the tree is healthy or not. This approach can be useful for monitoring and managing mango orchards, as it allows for the efficient and accurate identification of tree variety and health status. This can be helpful for farmers as it can assist in early identification of diseased or unproductive trees, enabling timely action to be taken, such as providing targeted treatment or replacing the unproductive trees. Additionally, identifying the variety of the tree can help farmers in making better decision regarding harvesting, pruning and other orchard management tasks. The use of DJI P4 Multispectral camera enabled us to acquire both RGB and NDVI data, which helped us to differentiate the healthy trees from the unhealthy ones. The results of our study showed that the YOLOv5 model performed the best, achieving an overall accuracy of more than 80% in tree detection, variety identification and health assessment, demonstrating the potential of UAV imagery, multispectral sensor and deep learning in fruit orchard management. The proposed approach can be easily scaled and applied to other crop fields as well, providing farmers with a reliable and efficient tool for monitoring and management of their orchards.
  • No Thumbnail Available
    Item
    CNN based techniques for Detecting Anomalies in the Crop
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2021) Rimsha Urooj , Shifa Allah Baksh , Fatima Batool; FA17-BSE-154 , FA17-BSE-164 , FA17-BSE-017; Dr. Zeeshan Gillani; LHR TP 7027
    The population across the globe is increasing at an alarming rate across the globe. This is giving rise to many challenges and one of the primary challenges is to feed an ever-increasing population with the same resources availed. This will lead to a food security and food crisis if we are not able to adapt our farming methods to modern technologies. With the advancement in computer vision techniques by using advanced CNN-based methods we can now analyse crops to detect anomalies in the crop. These applications could provide a foundation for the development of expertise aid or automatic screening tools. Such tools could contribute to further sustainable agricultural traditions and greater food production safety.
  • No Thumbnail Available
    Item
    Terrific Traffic
    (Library Information Services, COMSATS University Islamabad, Lahore Campus., 2020-11-20) Sarmad Khan; FA16-BCS-099; Dr. Zeeshan Gillani; LHR TP 6158
    The proposed is a system to detect traffic violations through anomalies detection using Machine Learning techniques from within a live traffic video stream. An anomaly is considered an irregularity or an event which would not occur in normal circumstances. Keeping traffic violations in account the circumstances will be such as Accidents, car crashes and lane violation. The proposed system will be able to differentiate not only between a normal and an abnormal event based on the events in the video stream but will also classify the anomaly accordingly. The purpose of this system is to automate and facilitate the traffic surveillance tasks in daily life. This result is achieved through implementing Image Processing and Deep learning techniques to predict the anomalous section of a video.
  • No Thumbnail Available
    Item
    Enhancing Soil Moisture Estimation through Integration of Remote Sensing and Machine Learning Techniques
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2023) Mehwish Mukhtar; SP22-RCS-013; LHR TP 8683; Dr. Zeeshan Gillani
    Soil moisture (SM) dynamics are integral to effective irrigation management, influencing both crop yield and the conservation of water resources within agricultural practices. Soil Organic matter is a pivotal factor in detecting the soil moisture. The capability to accurately detect irrigation events is essential for water use optimization, a critical concern in regions facing water scarcity where judicious irrigation can bolster crop health and advance sustainable farming practices. Traditional monitoring techniques, reliant on manual field surveys and intermittent data collection, are impeded by their limited scale and lack of continuous temporal coverage. Prior research has predominantly leaned on discrete point-based measurements or episodic remote sensing data, often inadequate for the detailed, persistent monitoring that precision irrigation scheduling demands. Addressing these limitations, our study introduces a cohesive framework for enhanced irrigation detection by using the organic matter of the bare soil, over extensive corn fields by integrating high-frequency satellite imagery with advanced machine learning techniques. Our comprehensive methodology encompasses the acquisition of Sentinel-1 satellite, SMAP and SMOS data, validation with ground-truth references, and sophisticated preprocessing tactics to navigate the common temporal and spatial discrepancies encountered in remote sensing analysis. We assessed various machine learning models, notably. Linear Regression, Random Forest, SVR, and KNN, with Liner Regression demonstrating the most accurate performance in predicting irrigation events and along with other models Naïve Bayes model were used for the Bare Soil classification and Soil Organic matter Detection. Mean Square Error (MSE) and Mean Absolute Error (MAE) are the evaluation parameters for the irrigation detection and F1-Score, Precision, Recall, Support and Accuracy are the evaluation parameters for the Soil Moisture estimation. However, satellite data dependency may introduce atmospheric inconsistencies, and model adaptability across different terrains or crop types may necessitate recalibration. Future directions broader environmental parameters, exploring deep learning potentials, and model validation across diverse agricultural settings. Ultimately, our vision is to develop a universally adaptable platform that delivers real time, actionable insights for stakeholders, revolutionizing irrigation practices on a global scale Soil moisture (SM) dynamics are integral to effective irrigation management, influencing both crop yield and the conservation of water resources within agricultural practices. Soil Organic matter is a pivotal factor in detecting the soil moisture. The capability to accurately detect irrigation events is essential for water use optimization, a critical concern in regions facing water scarcity where judicious irrigation can bolster crop health and advance sustainable farming practices. Traditional monitoring techniques, reliant on manual field surveys and intermittent data collection, are impeded by their limited scale and lack of continuous temporal coverage. Prior research has predominantly leaned on discrete point-based measurements or episodic remote sensing data, often inadequate for the detailed, persistent monitoring that precision irrigation scheduling demands. Addressing these limitations, our study introduces a cohesive framework for enhanced irrigation detection by using the organic matter of the bare soil, over extensive corn fields by integrating high-frequency satellite imagery with advanced machine learning techniques. Our comprehensive methodology encompasses the acquisition of Sentinel-1 satellite, SMAP and SMOS data, validation with ground-truth references, and sophisticated preprocessing tactics to navigate the common temporal and spatial discrepancies encountered in remote sensing analysis. We assessed various machine learning models, notably. Linear Regression, Random Forest, SVR, and KNN, with Liner Regression demonstrating the most accurate performance in predicting irrigation events and along with other models Naïve Bayes model were used for the Bare Soil classification and Soil Organic matter Detection. Mean Square Error (MSE) and Mean Absolute Error (MAE) are the evaluation parameters for the irrigation detection and F1-Score, Precision, Recall, Support and Accuracy are the evaluation parameters for the Soil Moisture estimation. However, satellite data dependency may introduce atmospheric inconsistencies, and model adaptability across different terrains or crop types may necessitate recalibration. Future directions broader environmental parameters, exploring deep learning potentials, and model validation across diverse agricultural settings. Ultimately, our vision is to develop a universally adaptable platform that delivers real time, actionable insights for stakeholders, revolutionizing irrigation practices on a global scale
  • No Thumbnail Available
    Item
    Integrating Remote Sensing and Deep Learning Techniques for Improved Crop Yield Estimation
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2023) Umer Usman; SP22-RCS-023; LHR TP 8684; Dr. Zeeshan Gillani
    Accurate yield estimation is crucial for making knowledgeable choices about agriculture and food safety around the world. However, many current studies in this place face issues inclusive of neglecting soil properties as predictors, limited generalization throughout special regions, and confined use of climate variables. This study targets to fill those gaps by using the machine and deep learning and techniques to predict wheat yield in 36 districts of Punjab, Pakistan. Using a dataset created with Google Earth Engine, the dataset consists of detailed statistics on soil residences, climate and other relevant agricultural variables or flora indices. The study included an in-intensity evaluation of the dataset, followed by using training and assessment of various machine learning and deep learning techniques, which includes Artificial Neural Networks, RandomForest, GradientBoosting, and XGBoost.The findings revealed that the ANN model encountered problems together with overfitting because of limited data, while GradientBoosting and XGBoost excel in accuracy and robustness. Specifically, GradientBoosting gave an RMSE of 362.4946 and a MAE of one 101.2008 tons/district, whilst XGBoost gave an RMSE of 280.4313 and an MAE of 102.5128 tons/district. Both algorithms confirmed high R² values, indicating their effectiveness for predicting district-level productiveness. The performance of the models was evaluated with the aid of comparing their predictions with the ground reality information provided by using CRS of Punjab, with error metrics given in tons at district level. The insights gained from this study are very essential to farmers, agronomists and policy makers, allowing them to make informed decisions and promote sustainable agricultural practices. This research makes a big contribution to the sphere of yield estimation and highlights factors for future studies, such as expanding the dataset and exploring other modelling techniques including LSTM, to enhance protection of food and sustainable agriculture
  • No Thumbnail Available
    Item
    City Scale Multi Camera Vehicle Tracking Using Deep Learning
    (Library Information Services, COMSATS University Islamabad, Lahore Campus, 2023) Muhammad Sameed Khan; SP20-RCS-013; LHR TP 8465; Dr. Zeeshan Gillani
    Intelligent city and traffic management depend on city-scale multi-camera vehicle tracking; however, this work has various difficulties. On the different viewing angles, problems arise that include the variation on the large scale, frequent occlusion and appearance variation. In this study, we use the cross-camera tracking technique and multi-camera tracking system that considers aggregation loss. To address the challenges of multi-camera vehicle tracking, the suggested system has four key parts. First, we extract the tracks with the help of a single camera view by identifying the object and multi-object tracking modules. These modules combine their detection capabilities to provide efficient tracking between frames. After obtaining the tracklets, we use a multi camera re-identification module to match them. The tracklets acquired by several cameras are connected by this module using re-identification techniques. We used OsNetX1_0 and retinaNet50 for reidentification. In the final stage, we deal with the isolated trackless and tracking of the synchronize ids that rely on the outcomes of the re-identification. We improve the computational performance with the help of less parameter models and cohesion of the tracking system by removing isolated tracklets and assuring consistent tracking IDs. With practical and effective alternatives, the research advances multi-camera vehicle tracking on the city scale. However, this system shows the significance of the fast-multi-target cross-camera tracking approaches and the loss of aggregation in the study challenges. The AI city challenges' success reveals our system's potency and viability.