Research Title: Plant Disease Detection Using Deep Learning

Author: Mahmoud Mohammed Mahmoud Hussein, Published Year: 2023

Journal of Intelligent Systems and Applications, 15

Faculty: Information Technology

Abstract: Agricultural development is a critical strategy for promoting prosperity and addressing the challenge of feeding nearly 10 billion people by 2050. Plant diseases can significantly impact food production, reducing both quantity and diversity. Therefore, early detection of plant diseases through automatic detection methods based on deep learning can improve food production quality and reduce economic losses. While previous models have been implemented for a single type of plant to ensure high accuracy, they require high-quality images for proper classification and are not effective with low-resolution images. To address these limitations, this paper proposes the use of pre-trained model based on convolutional neural networks (CNN) for plant disease detection. The focus is on fine-tuning the hyperparameters of popular pre-trained model such as EfficientNetV2S, to achieve higher accuracy in detecting plant diseases in lower resolution images, crowded and misleading backgrounds, shadows on leaves, different textures, and changes in brightness. The study utilized the Plant Diseases Dataset, which includes infected and uninfected crop leaves comprising 38 classes. In pursuit of improving the adaptability and robustness of our neural networks, we intentionally exposed them to a deliberately noisy training dataset. This strategic move followed the modification of the Plant Diseases Dataset, tailored to better suit the demands of our training process. Our objective was to enhance the network's ability to generalize effectively and perform robustly in real-world scenarios. This approach represents a critical step in our study's overarching goal of advancing plant disease detection, especially in challenging conditions, and underscores the importance of dataset optimization in deep learning applications.

Keywords: CNN, Deep Learning, EfficientNetV2S, Classification, Plant Diseases

Research Title: A Framework for Predicting Breast Cancer Recurrence

Author: Mahmoud Mohammed Mahmoud Hussein, Published Year: 2024

Expert Systems with Applications, 240

Faculty: Information Technology

Abstract: Breast cancer is one of the serious diseases that threaten the life of many women worldwide. The seriousness of this disease is that it is often discovered in late stages after a period of its occurrence. This causes a wide spread of the disease and difficulty in its treatment. Another important characteristic of this disease is that it can return again after a period of its treatment. Therefore, predicting the occurrence or recurrence of such disease early is the best solution to have a high cure rate. The main objective of this paper is to improve the prediction performance of the breast cancer recurrence. Many methods have been proposed to predict breast cancer recurrence. However, these methods did not achieve the desired results on one of the most famous datasets in the field of breast cancer recurrence’s prediction (i.e., Wisconsin Prognosis Breast Cancer (WPBC) dataset). The highest accuracy achieved using the previous methods is 89.89%. Therefore, this paper provides a framework for improving the prediction of breast cancer recurrence. The proposed framework has the ability to overcome many of the challenges in the existing dataset such as: (a) the problem of imbalance between classes using a data over-sampling technique; and (b) the large number of data dimensions using Principal Component Analysis (PCA), and a wrapper dimensionality reduction technique based on Genetic Algorithm (GA). It also uses the neural network algorithm to fuse the results of two individual classifiers (i.e., Random Forest (RF) and Support Vector Machine (SVM)). Our proposed framework evaluation showed a significant improvement in the predication performance. It achieved an accuracy of 98.3%, area under the curve of 99%, and precision, recall, and f1-measure of 98%.

Keywords: Breast cance, prediction, imbalance data

Research Title: Secure Context-Aware Traffic Light Scheduling System: Integrity of Vehicles’ Identities

Author: Said Ahmad Ammar Ghoul, Published Year: 2025

world electric vehicle Journal, 16

Faculty: Information Technology

Abstract: Autonomous vehicles and intelligent traffic transportation are widely investigated for road networks. Context-aware traffic light scheduling algorithms determine signal phases by analyzing the real-time characteristics and contextual information of competing traffic flows. The context of traffic flows mainly considers the existence of regular, emergency, or heavy vehicles. This is an important factor in setting the phases of the traffic light schedule and assigning a high priority for emergency vehicles to pass through the signalized intersection first. VANET technology, through its communication capabilities and the exchange of data packets among moving vehicles, is utilized to collect real-time traffic information for the analyzed road scenarios. This introduces an attractive environment for hackers, intruders, and criminals to deceive drivers and intelligent infrastructure by manipulating the transmitted packets. This consequently leads to the deployment of less efficient traffic light scheduling algorithms. Therefore, ensuring secure communications between traveling vehicles and verifying the integrity of transmitted data are crucial. In this work, we investigate the possible attacks on the integrity of transferred messages and vehicles’ identities and their effects on the traffic light schedules. Then, a new secure context-aware traffic light scheduling system is proposed that guarantees the integrity of transmitted messages and verifies the vehicles’ identities. Finally, a comprehensive series of experiments were performed to assess the proposed secure system in comparison to the absence of security mechanisms within a simulated road intersection. We can infer from the experimental study that attacks on the integrity of vehicles have different effects on the efficiency of the scheduling algorithm. The throughput of the signalized intersection and the waiting delay time of traveling vehicles are highly affected parameters

Keywords: intelligent transport system; context-aware algorithm; traffic light; security; integrity; vehicle identity

Research Title: Optimized National Grid Harmonics Using Low Switching PSO Algorithm of Multilevel Inverter

Author: Akram Mohammed Al-mahrouk, Published Year: 2025

International Conference on Engineering and Computing Technologies (EngiTek 2025), Irbid/Jordan

Faculty: Engineering and Technology

Abstract: This paper presents an optimized technique for mitigating low-order harmonics in the national grid (NG) using an on-grid DC–AC seven-level (7L) multilevel inverter (MLI) operated at low switching frequency. The switching sequences are determined by particle swarm optimization (PSO) based on the NG harmonic spectrum, with 315 sampling points per period of the 50 Hz waveform. Case studies are conducted, including pure NG waveform and targeted harmonic reduction. The proposed technique effectively reduces selected harmonics of the NG, such as the fifth, seventh, and ninth orders, to approximately 1%. The proposed design is based on a lookup table to reduce processing time, making it suitable for practical implementation while enhancing harmonic reduction performance.

Keywords: Multilevel Inverter, Total Harmonics Distortion, Practical SWARM Objective, Smart Grid.

Research Title: A Comprehensive Framework for Improving Remote Sensing Image Classification: Combining Augmentation and Missing Pixel Imputation

Author: Amgad Monir Mohaned Elsayed, Published Year: 2024

International Journal of Computers and Information, 11

Faculty: Information Technology

Abstract: Remote sensing image classification is crucial in various domains including agriculture, urban planning, and environmental monitoring. However, limited labeled data and missing pixels pose challenges to achieving accurate classification. In this study, we propose a comprehensive framework that integrates data augmentation using a latent diffusion model and reinforcement learning-based missing pixel imputation to enhance deep learning models' classification performance. The framework consists of three layers: data augmentation, missing pixel imputation, and classification using a modified VGG16 architecture. Extensive experiments on benchmark datasets demonstrate the significant impact of our framework, surpassing state-of-the-art techniques by significantly improving classification accuracy and robustness. The results highlight the effectiveness of our augmentation and imputation techniques, achieving remarkable Dice Score, Accuracy, and Recall metrics of 97.56%, 97.34%, and 97.34%, respectively. Our proposed framework provides a valuable solution for accurate remote sensing image classification, addressing the challenges of limited data and missing pixels, and has broad applications in various domains.

Keywords: VGG 16; convolution neural network; diffusion model; remote sensing; satellite image.

Research Title: Vertical and Horizontal Data Partitioning for Classifier Ensemble Learning

Author: Amgad Monir Mohaned Elsayed, Published Year: 2019

Progress in Computer Recognition Systems (CORES 2019), Polanica Zdroj, Poland

Faculty: Information Technology

Abstract: Multiple classifier systems have proven superiority over individual ones to solve classification tasks. One of the main issues in those solution relies in data size, when the amount of data to be analyzed becomes huge. In this paper, the performance of ensemble system to succeed by using only portions of the available data is analyzed. For this, extensive experimentation with homogeneous ensemble systems trained with 50% of data and 50% of features is performed, using bagging sampling schema. Simple and weighted majority voting schemes are implemented to combine the classifier outputs. Experimental results including 25 datasets show the benefit of using multiple classifiers trained on limited data. The ensemble size and the accuracy obtained with individual model trained over the entire dataset is compared.

Keywords: Classifier ensemble ; Pattern classification ; Data complexity

Research Title: Advanced solar radiation prediction using combined satellite imagery and tabular data processing

Author: Amgad Monir Mohaned Elsayed, Published Year: 2025

Scientific Reports, 15

Faculty: Information Technology

Abstract: Accurate solar radiation prediction is crucial for optimizing solar energy systems. There are two types of data that can be used to predict solar radiation, such as satellite images and tabular satellite data. This research focuses on enhancing solar radiation prediction by integrating data from two distinct sources: satellite imagery and ground-based measurements. By combining these datasets, the study improves the accuracy of solar radiation forecasts, which is crucial for renewable energy applications. This research presents a hybrid methodology to predict the solar radiation from both satellite images and satellite data. The methodology basis on two datasets; the first data set contains tabular data, and the second dataset contains satellite images. The framework divides into two paths; the first path take the input as the satellite images; this stages contains three steps; the first step is removing noise using latent diffusion model, the second step is about pixel imputation using a modified RF + Identity GAN (this model contains two modification the first modification is adding the identity block to solve mode collapse problem in the GANs and the second modification is to add the 8-connected pixel to generate a value of missing pixel near to the real missed pixel. The third step in the first path is about using the self-organizing map to identify the special informative in the satellite image. The second path take the input as tabular data and use the diffusion model to impute the missing data in the tabulated data. Finally, we merge the two path and use feature selection to be as input for the LSTM for solar radiation predictions. The experiments done prove the efficiency of the used stage such as missing pixel imputation, removing noise, missing data imputation and prediction using LSTM when compared with other available techniques. The experiments also prove the enhancement of all prediction model after adding two paths before the prediction step.

Keywords: Solar radiation; LSTM; GANs; Identity block; Latent diffusion model

Research Title: A hybrid deep learning framework for solar irradiation prediction based on regional satellite images and data

Author: Amgad Monir Mohaned Elsayed, Published Year: 2025

Neural Computing and Applications, 37

Faculty: Information Technology

Abstract: Accurate prediction of solar irradiation, particularly Diffuse Horizontal Irradiance (DHI), is essential for optimizing solar energy systems. This paper presents a hybrid framework that integrates satellite imagery and satellite-derived data to improve the precision of DHI forecasts. Two key datasets were employed: ground-based Solar Irradiation Measurement (SRM) and satellite-based Solar Irradiation (SSR). The proposed methodology utilizes machine learning models along two paths. The first path processes satellite imagery using advanced techniques, including pixel imputation with a modified Random Forest (RF) and Generative Adversarial Network (GAN), noisy region detection with Self-Organizing Maps (SOM), and noise removal through a Latent Diffusion Model (LDM). The second path handles tabular data through a diffusion probabilistic model designed for missing data imputation. These features are then merged and fed into a Long Short-Term Memory (LSTM) network, enhanced to capture seasonality and trends for accurate DHI predictions. The paper makes significant contributions by introducing a robust hybrid framework that leverages both satellite imagery and tabular data, incorporating novel preprocessing methods such as GAN-based pixel imputation and diffusion-based noise removal. The LSTM model is further adapted to handle seasonal and trend components, resulting in enhanced DHI forecasting accuracy. The proposed model demonstrates superior performance, achieving a Mean Squared Error (MSE) of 8.170 W/m2, Root Mean Squared Error (RMSE) of 1.749 W/m2, Mean Absolute Error (MAE) of 0.693 W/m2, and an R-squared value of 2.574, significantly outperforming conventional methods such as Artificial Neural Networks (ANN), Support Vector Regression (SVR), and traditional LSTM models. Additionally, the model records a low Mean Absolute Percentage Error (MAPE) of 0.8183% and a computation time of 35.4 s, highlighting its efficiency. These results contribute to more accurate and reliable solar energy resource management by offering enhanced DHI forecasts across various spatial and temporal scales.

Keywords: Solar irradiation; LSTM; GANs; Identity Block; Latent diffusion model

Research Title: A hybrid deep learning approach for accurate water body segmentation in satellite imagery

Author: Amgad Monir Mohaned Elsayed, Published Year: 2025

Earth Science Informatics , 18

Faculty: Information Technology

Abstract: Precise water body segmentation in satellite imagery plays a vital role in environmental monitoring, water resource management, and disaster prevention. This study introduces a high-performance segmentation framework leveraging Sentinel-2 imagery, integrating advanced methodologies to enhance data integrity and segmentation accuracy. To increase dataset diversity and model generalization, StyleGAN3-based augmentation was implemented, yielding a 5?curacy improvement over conventional methods. An Attention-Guided Denoising Autoencoder with Skip Connection (AG-DAES) was utilized for noise reduction, effectively preserving spatial details and strengthening segmentation robustness. To address missing and corrupted pixels, Bi-ConvRNN was employed for pixel restoration, significantly boosting performance. Additionally, the Particle Swarm Dandelion Optimization (PSDO) algorithm was used for hyperparameter tuning, contributing to an additional 3–5?curacy gain. Feature extraction was refined through the Multiscale Strip Convolution Module (MSSCM), enhancing spatial-spectral representation and leading to a 6–8?curacy increase. The segmentation process was executed using the Map U-Net model, which, after integrating all proposed improvements, achieved state-of-the-art accuracy exceeding 99%. A comparative study demonstrated that the proposed framework outperforms existing methods, particularly in complex scenarios involving vegetation interference, occlusions, and mixed land–water transitions. This adaptable and scalable approach sets a new standard for water body segmentation in satellite image analysis, offering a powerful tool for future research in the field.

Keywords: Water body segmentation; Satellite imagery; Deep learning; Noise reduction; Pixel imputation

Research Title: Dynamic Classification Ensembles for Handling Imbalanced Multiclass Drifted Data Streams

Author: Amgad Monir Mohaned Elsayed, Published Year: 2024

Information Sciences, 670

Faculty: Information Technology

Abstract: Machine learning models often encounter significant difficulties when dealing with multiclass imbalanced data streams in nonstationary environments. These challenges can lead to biased and unreliable predictions, which ultimately impact the overall performance of the models. To address these issues, we propose an innovative approach that integrates dynamic ensemble selection, an adaptive technique for managing imbalanced multiclass data streams, with a concept drift detector for recognizing stream changes and the K-nearest neighbor (KNN) algorithm to tackle issues related to class overlap. The primary objective was to improve the classification of imbalanced multiclass drifted data streams. The adaptive oversampling method generates synthetic samples to mitigate the issues associated with imbalanced data streams. This method utilizes KNN to ensure that the generated samples do not overlap. To handle incoming data streams, a drift detector assists in deciding whether to retain the existing classifiers or create a new one. Dynamic Ensemble Selection (DES) was utilized to select the most appropriate classifier for incoming data, aiming to optimize the performance of the classification task. The proposed method offers an effective solution for achieving an accurate and resilient classification in the context of imbalanced multiclass drifted data streams. To evaluate the effectiveness of our proposal, we conducted experiments on a variety of datasets, including benchmark datasets, real application stream datasets, and synthetic data streams. The experimental results demonstrate the superiority of our contribution in addressing the challenges posed by imbalanced multiclass drifted data streams.

Keywords: Dynamic ensemble selection; Concept drift detection; Imbalanced data classification