A5058861950- Handwritten Text Recognition Techniques
- Image Processing and 3D Reconstruction
- Vehicle License Plate Recognition
- Image Retrieval and Classification Techniques
- Natural Language Processing Techniques
- Topic Modeling
- Text and Document Classification Technologies
- Advanced Text Analysis Techniques
- Sentiment Analysis and Opinion Mining
- Advanced Image and Video Retrieval Techniques
- Remote-Sensing Image Classification
- Advanced Malware Detection Techniques
- Analytical Chemistry and Chromatography
- 3D Surveying and Cultural Heritage
- Currency Recognition and Detection
- Advanced Neural Network Applications
- Biometric Identification and Security
- Ethics and Social Impacts of AI
- Cultural Heritage Materials Analysis
- Network Security and Intrusion Detection
- Radiomics and Machine Learning in Medical Imaging
- Remote Sensing and LiDAR Applications
- Face recognition and analysis
- Brain Tumor Detection and Classification
- User Authentication and Security Systems
Lakehead University
2023-2025
Iqra University
2023-2024
NED University of Engineering and Technology
2018-2022
Western University
2022
University of Baghdad
2022
Chaitanya Bharathi Institute of Technology
2021
University of Technology Malaysia
2017-2020
King Saud bin Abdulaziz University for Health Sciences
2014-2020
Intel (Malaysia)
2019
Indus University
2019
Recurrent neural networks (RNN) have been successfully applied for recognition of cursive handwritten documents, both in English and Arabic scripts. Ability RNNs to model context sequence data like speech text makes them a suitable candidate develop OCR systems printed Nabataean scripts (including Nastaleeq which no system is available date). In this work, we presented the results applying RNN Urdu script. Bidirectional Long Short Term Memory (BLSTM) architecture with Connectionist Temporal...
The technological advancement and sophistication in cameras gadgets prompt researchers to have focus on image analysis text understanding. deep learning techniques demonstrated well assess the potential for classifying from natural scene images as reported recent years. There are variety of approaches that prospects detection recognition text, effectively images. In this work, we presented Arabic using Convolutional Neural Networks (ConvNets) a classifier. As data is slanted skewed, thus...
A myriad of the population has adapted to evolving technology, which includes text communication. Users advertently or inadvertently share emotions. As we know, emotions are one most critical aspects human life; they impact human's behavior, thinking, compelling action, and important, decision making. There many alleged known us, each having its significance. In this era modern it is hard find any unexplored area; applies emotion. People express their through a lot nowadays, led Emotion...
In recent years, significant progress has been achieved in understanding and processing tabular data. However, existing approaches often rely on task-specific features model architectures, posing challenges accurately extracting table structures amidst diverse layouts, styles, noise contamination. This study introduces a comprehensive deep learning methodology that is tailored for the precise identification extraction of rows columns from document images contain tables. The proposed employs...
Prior studies have shown that distinguishing text generated by large language models (LLMs) from human-written one is highly challenging, and often no better than random guessing. To verify the generalizability of this finding across languages domains, we perform an extensive case study to identify upper bound human detection accuracy. Across 16 datasets covering 9 19 annotators achieved average accuracy 87.6%, thus challenging previous conclusions. We find major gaps between machine lie in...
Recognition of Urdu cursive script is a challenging task due to the implicit complexities associated with it. The performance recognition system immensely dependent on extracted features. There are various features extraction approaches proposed in recent years. Among many, an approach based zoning proved be efficient and popular. Such represent significant information low complexity high speed. In this paper, we used for classification Nasta'liq text lines, combination 2-Dimensional Long...
The recognition of text in natural scene images is a practical yet challenging task due to the large variations backgrounds, textures, fonts, and illumination. English as secondary language extensively used Gulf countries along with Arabic script. Therefore, this paper introduces English-Arabic 42K image dataset. dataset includes appeared scripts while maintaining prime focus on can be employed for evaluation segmentation task. To provide an insight other researchers, experiments have been...
Malicious hackers breach security perimeters, cause infrastructure disruptions as well steal proprietary information, financial data, and violate consumers’ privacy. Protection of the whole organization by using firm's officers can be besieged with faulty warnings. Engineers must shift from console to put together investigative clues a result today's fragmented technologies that frustratingly sluggish investigations. Endpoint Detection Response (EDR) solutions adds an...
The similar nature of patterns may enhance the learning if experience they attained during training is utilized to achieve maximum accuracy. This paper presents a novel way exploit transfer on handwritten Urdu text analysis. MNIST pre-trained network employed by transferring it's Nastaliq Handwritten Dataset (UNHD) samples. convolutional neural used for feature extraction. experiments were performed using deep multidimensional long short term (MDLSTM) memory networks. obtained result shows...
Malware is any software that causes harm to the user information, computer systems or network. Modern computing and internet are facing increase in malware threats from internet. It observed different follows same patterns their structure with minimal alterations. The type of has evolved, file-based fileless malware, such kind also known as Advance Volatile Threat (AVT). Fileless complex evasive, exploiting pre-installed trusted programs infiltrate information its malicious intent. designed...
The use of machine learning in healthcare has the potential to revolutionize virtually every aspect industry. However, lack transparency AI applications may lead problem trustworthiness and reliability information provided by these applications. Medical practitioners rely on such systems for clinical decision making, but without adequate explanations, diagnosis made cannot be completely trusted. Explainability Artificial Intelligence (XAI) aims improve our understanding why a given output...
Hyperspectral Imaging (HSI) uses large portions of the electromagnetic spectrum to obtain information from images that would be very difficult otherwise. An important task in forensic analysis documents is extract signatures for authentication. Signatures often overlap with other parts a document such as typed text or stamps and hence it retrieve them. In this work we present novel algorithm recovering hyperspectral where text, seals, stamps, images. We used pure pixel index approaches...
This paper presents a comprehensive survey on Arabic cursive scene text recognition. The recent years’ publications in this field have witnessed the interest shift of document image analysis researchers from recognition optical characters to appearing natural images. Scene is challenging problem due having variations font styles, size, alignment, orientation, reflection, illumination change, blurriness and complex background. Among scripts, contemplated as more joined writing, same character...
There are a lot of intensive research on character and numeral recognition for some popular scripts such as Roman, Arabic, Persian/Farsi, Chinese Indian.Unfortunately, there is less work the Urdu language which used by ¼ population world.In this paper we present review Urdu, Arabic Farsi.Most published since 2003 recent publications in Farsi languages have been summarized survey paper.The Unicode system writing glyph digits analyzed its derivative languages.We also added future direction system.
Machine simulation of human reading has been a subject intensive research for almost four decades. Automatic Urdu character recognition remains challenging task due to its cursive nature despite the fact that latest improvements in methods and systems Latin script are very promising. This work introduces robust approach based on statistical models provide solution text Nasta'liq style. Contrary classical approaches which segment into words, ligatures or characters, we intend employ an...