Mohamed Uvaze Ahamed Ayoobkhan
@newuu.uz
Professor and Software Engineering
New Uzbekistan University
Mohamed Uvaze Ahamed Ayoobkhan currently working as Assistant Professor in Software Engineering Department, New Uzbekistan University, Tashkent, Uzbekistan. Area of research in Human-computer Interaction, Artificial Neural Network and Machine Learning.
EDUCATION
Bachelor of Engineering (CSE) - Anna University, India,
Master of Engineering (CSE) - Anna University, India,
Doctor of Philosophy (IT) - Multimedia University, Malaysia
RESEARCH INTERESTS
Machine Learning, Optimization, Medical Image Processing
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Scopus Publications
Sushma Kakkar, A. Jaya Lakshmi, Anand Deva Durai C, Wakeel Ahmad, Mohamed Uvaze Ahamed Ayoobkhan, P. Veeramanikandan, P. Nagasekhar Reddy, Vijay Kumar Dwivedi, and A. Rajaram
Informa UK Limited
Mohamed Uzave Ahmed A., Ulugbek Raximov, Mehriniso Sayidvalieva, and Aneesh Pradeep
IEEE
Electric vehicles are rapidly gaining popularity these days. This is due to the air pollution problem since 55.2% of greenhouse gas emissions stem from conventional cars. In contrast, only 5.76% of total air emissions produced emanate from electric cars. Nevertheless, electric cars are not the only environmentally friendly vehicles. Hydrogen vehicles have a fast and cheap recharge and are also as eco-friendly as electric cars, but they have failed to achieve the same success. In this paper, we will analyze the characteristics and compare them and find out how these two types of cars work. And most importantly, we will try to answer the question, "Can hydrogen cars replace electric ones?"
Ismoiljon Mirabdullayev, Mohamed Uvaze Ahamed Ayoobkhan, A.M. Jasmine Hashana, and Liayakath Ali Khan Subair Ali
IEEE
Drowsy driving poses a major threat to road safety, demanding effective solutions. To address this issue, this paper introduces a cutting-edge drowsy driving detection system that leverages advanced face detection algorithms to analyze real-time facial features. By closely monitoring eye movements and mouth patterns, the system is capable of accurately identifying early signs of fatigue. In response, timely warnings are promptly issued to the driver through various means, including audible alarms, visual cues, or even physical interventions. Although the system holds great potential in enhancing road safety, there are still challenges to overcome, particularly regarding accuracy and adaptability. However, ongoing research endeavors aim to refine the algorithms and integrate additional physiological and behavioral measures, with the goal of further improving the system's capabilities. By preventing accidents caused by drowsiness, this innovative system demonstrates a promising pathway towards bolstering road safety. By addressing the critical issue of drowsy driving, it contributes significantly to the overall safety of drivers and passengers alike, making our roads a safer place for everyone.
A.M. Jasmine Hashana, P. Brundha, Mohamed Uvaze Ahamed Ayoobkhan, and Fazila S
IEEE
Abstract-As a subset of machine learning, deep learning makes use of multiple-layer neural networks to learn with available data and make decisions or predictions. A large language model called ChatGPT is based on deep learning, specifically a type of neural network called a transformer. ChatGPT's transformer architecture uses attention mechanisms to focus on the most important parts of the input, allowing it to process and comprehend a large amount of text data. In order for the model to comprehend the context and meaning of natural language text, it is trained on a huge database of text, including articles and books. One of the main importance of using deep learning in ChatGPT is its intelligence to understand relationships and patterns from the input text and generate or predict new text that is homogeneous to the input/training data. Because of this, ChatGPT is able to respond to questions and prompts in a manner that is comparable to that of a human, making it useful for a wide scope of natural language processing missions like translating languages, summarizing texts, and responding to questions. It's worth noting that, while deep learning has been highly effective in ChatGPT, it is not without its limitations. To train, deep learning models can be very complex and require a lot of data and computing power.
S. Navaneetha Krishnan, D. Yuvaraj, Kakoli Banerjee, P Joel Josephson, T CH Anil Kumar, and Mohamed Uvaze Ahamed Ayoobkhan
Elsevier BV
A. Jayanthiladevi, Mohamed Uvaze Ahamed Ayoobkhan, R. ThamaraiSelvi, Laishram Jimmy, Piyush Mishra, and Nismon Rio Robert
Inderscience Publishers
R.G. Vidhya, T.S. Sasikala, Ayoobkhan Mohamed Uvaze Ahamed, Subair Ali Liayakath Ali Khan, Kamlesh Singh, and M. Saratha
IEEE
Breast cancer develops from breast tissue and leads to abnormally growing cells in the chest. Doctors usually look for tumors on a mammogram, and some mammograms contain abnormal macrocalcifications and microcalcifications when the image quality is very poor. The presence of these abnormal amounts of calcium deposits in the breast is a sign of early breast cancer and should never be ignored. The image quality should be of the highest quality for an accurate interpretation of this mammographic deposit. Proposed research work is ongoing, exploring other screening methods and the stages of breast cancer. Improved Adaptive Fuzzy C-Means (IAFCM), Ant Colony Algorithm (ACA), and TNM (The size of the breast tumour (T), adjacent lymph nodes and Metastasized methods are used which builds the proposed medical image processing systems into an efficient way. Modified Poisson Inverse Gradient, Metastasized classifier (MPIG) has been used for classification. More than 500 image modalities are involved in all of the approaches. Clinical practitioners who make decisions based on photographs are predicted to benefit from the findings of this study.
Sathana Venkadasubbiah, D. Yuvaraj, Subair Ali, and Mohamed Uvaze Ahamed Ayoobkhan
Springer International Publishing
Minhajur Rahman, Saimunur Rahman, and Mohamed Uvaze Ahamed Ayoobkhan
IEEE
Classification of vehicles plays an important role in the intelligent transport system. In this paper, we propose a framework for the classification of Bangladeshi vehicle images based on fine-tuned convolutional neural networks (CNNs). Using the proposed framework and an unified experimental setting, we fine-tuned fifteen popular CNN architectures, namely, AlexNet, Inception-V3, VGG-II, VGG-13, VGG-16, VGG-19, ResNet18, ResNet-34, ResNet-50, ResNet-101, ResNet-152, DenseNet121, DenseNet-161, DenseNet-169 and DenseNet-201 pretrained on ImageNet dataset on a public Bangladeshi vehicle image dataset. We conduct a systematic and comprehensive analysis on the performance of the fine-tuned CNNs which leads to new insights. Our experimental results show that ResNet-152 and DenseNet-201 fine-tuned with our proposed strategy provide excellent performance for the classification of Bangladeshi vehicle images. Our implementation is available at https://github.con MinhajurRFahad/bd-vehicle-cnn-benchmark.
Minhajur Rahman, Saimunur Rahman, and Mohamed Uvaze Ahamed Ayoobkhan
IEEE
Food image classification has received significant attention from researchers in recent years. A large number of methods have been shown to be working well on generic image food classification and dealing with problems like scale and appearance. However, their applicability to the classification of food images from a specific country and culture is still unknown. In this paper, we focus on Bangladeshi meat based curry image classification. We experiment with multiple transfer learning strategies to perform curry image classification with popular pretrained CNNs. Our proposed transferlearning strategies can be useful for achieving better classification performance when full fine-tuning and training from scratch of a CNN is not possible. We also proposed a new dataset for the task and performed a relevant ablation study. Our source code and dataset are available at https://github.com/MinhajurRFahad/bd-curry-image.
D. Yuvaraj, S. Shanmuga Priya, M. Braveen, S. Navaneetha Krishnan, S. Nachiyappan, Abolfazl Mehbodniya, A. Mohamed Uvaze Ahamed, and M. Sivaram
Computers, Materials and Continua (Tech Science Press)
Suganthi ST, Mohamed Uvaze Ahamed Ayoobkhan, Krishna Kumar V, Nebojsa Bacanin, Venkatachalam K, Hubálovský Štěpán, and Trojovský Pavel
PeerJ
Deep Learning is an effective technique and used in various fields of natural language processing, computer vision, image processing and machine vision. Deep fakes uses deep learning technique to synthesis and manipulate image of a person in which human beings cannot distinguish the fake one. By using generative adversarial neural networks (GAN) deep fakes are generated which may threaten the public. Detecting deep fake image content plays a vital role. Many research works have been done in detection of deep fakes in image manipulation. The main issues in the existing techniques are inaccurate, consumption time is high. In this work we implement detecting of deep fake face image analysis using deep learning technique of fisherface using Local Binary Pattern Histogram (FF-LBPH). Fisherface algorithm is used to recognize the face by reduction of the dimension in the face space using LBPH. Then apply DBN with RBM for deep fake detection classifier. The public data sets used in this work are FFHQ, 100K-Faces DFFD, CASIA-WebFace.
Elbek Normurodov and Mohamed Uvaze Ahamed Ayoobkhan
IEEE
Since a large number of heterogeneous tasks are used in service-oriented cloud computing frameworks. The problem of random hardware failures of the processors is turning out to be a progressively noticeable task. In Uzbekistan, the method of managing these tasks is mostly done using a “pen and paper” approach. Business owners from other countries have already implemented digital management systems. It is high time to introduce a national task and time management system, which is cheaper, suits better for national business operations as well as better designed for business needs based on the analysis of other foreign competitor applications. In this work, we advise a fast task assignment system to fulfill unwavering quality prerequisite for a parallel application in a heterogeneous service-oriented cloud computing framework. This is the platform, where workers can enter their action/activity during a specific timeframe. This permits employers to have a clear view of the time required by the task to be completed by the employee.
Laith R. Flaih, Shaimaa Awadh Baha al_Deen, and Mohamed Uvaze Ahamed Ayoobkhan
IOP Publishing
Abstract This investigation provides a methodology for surface quality measurement. In machine based vision, an optical inspection is validated to identify defects over materials. As well, normalization approach is used to process homogeneous thickness. With compensation procedures flaws are identified and analyzed. However, after defect identification, decision rules are defected for appropriate classification which offers optimal performance and diminishes tuning complexity. The anticipated approach is effectual and fulfils inspection requirements. Experimental outcomes may validate performance of anticipated approach to recognition rate and inspection speed.
D. Yuvaraj, M. Sivaram, A. Mohamed Uvaze Ahamed, and S. Nageswari
Springer International Publishing
M. Deva Brinda, Mohamed Uvaze Ahamed. A, R. Sundar, M. Padmapriya, V. Porkodi, and Sangeetha Krishnan
IEEE
The environment changes like increase in moisture level, humidity, fall in temperature and everyday changes in the voltage levels leads to the generation of fault current. This high value of fault current has to be monitored continuously. The need for a monitoring and alert system is to protect the electronic device. A proper grounding for equipment is must if not it leads to the induction of fault current. This fault current has a larger probability in providing a proper zero resistance path this leakage current. also provides safety to both humans as well as machines. This device is specifically designed for ATM system across the country, is capable of measuring the earth-neutral voltage up to 10 volts. When there occurs any isolation in the voltage levels due to the above mentioned changes, this microcontroller based alert system sends information to the authorised person through SMS and E- MAIL.As this device is not connected with the main system any surges in the ATM will not affect this device and has the ability to disconnect the device from fault in emergency cases.
M. Sivaram, Mohamed Uvaze Ahamed A, D. Yuvaraj, G. Megala, V. Porkodi, and Manivel Kandasamy
IEEE
Biometrics manages the computerized acknowledgment of people dependent on natural and social attributes. The example acknowledgment framework perceives an individual by deciding the credibility of a particular conduct normal for person. The primary rule of biometric framework is recognizable proof and check. A biometric confirmation framework use fingerprints, face, hand geometry, iris, and voice, mark, and keystroke elements of a person to recognize an individual or to check a guaranteed character. Biometrics authentication is a form of identification and access control process which identify individuals in packs that are under reconnaissance. Biometric security system increase in the overall security and individuals no longer have to deal with lost ID Cards or forgotten passwords. It helps much organization to see everyone is at a certain time when something might have happened that needs reviewed. The current issues in biometric system with individuals and many organization facing are personal privacy, expensive, data’s may be stolen.
R. Sathya Bama Krishna, B. Bharathi, Mohamed Uvaze Ahamed. A, and B. Ankayarkanni
IEEE
Moving object in a video could be explored using hybrid methodologies as one among the enticing field of vision in computers. It is extensively applied in video surveillances and target identification system. Extracting reliable information accurately is a rigorous task in a challenging environment. This paper investigates the problem of detecting an object in dynamic scenes. We suggest two method 1) feature extraction using FBF 2) Image matching using ISURF. The ISURF (Improved Speeded up Robust Feature detection) is the improvised method of original SURF algorithm. In this the matching duration is reduced by limiting the total number of features to be compared. The FBF (Fast Bilateral Filtering) algorithm is suggested for feature extraction and denoising the captured key frames. Thus this paper proposes a hybrid method for moving object exploration in a dynamic scene with reduced time.
Bilal Hikmat Rasheed and
The World Academy of Research in Science and Engineering
Attacks over the internet have become an increasing menace in recent time which tries to hack or illegally tamper with the data available over the networks. On the other hand, there has been an increase in volume in research contributions to effectively counter attack these attacks and implement a strong defence mechanism. There have been numerous algorithms and frameworks implemented in recent times which are intelligent and soft computing based. These evolution based algorithms play a vital role in self adapting the system under attack towards increasing and new types of attacks which are increasing day by day. One such area of soft computing algorithms investigated in this chapter is the artificial neural network or popularly known as ANNs. They work analogous to the biological neurons in the human body. The chapter is organized in a systematic manner to give an insight in to ANN based network models to counter attack DDoS attacks which has been the primary focus of this thesis, architecture and implementation of ANNs, the experimental investigations and findings which help in drawing an inference of ANN based defence models.
Yuvaraj D and
The World Academy of Research in Science and Engineering
Among the various attacks found extensively in the literature distributed denial of service attack is a special form of attack which poses to be a great menace and if not properly dealt with has the capability of bringing the power of computing systems to a halt with severe financial losses. Of the several defence mechanisms found in the literature, the most prevalent and prominently used ones are the intelligent and soft computing based evolutionary algorithms. Three such algorithms have been taken, investigated and experimented in this thesis for defence against DDoS attacks. This paper investigates the last algorithm namely ant colony optimization (ACO) which is yet another nature inspired algorithm for providing optimality in the DDoS defence system implemented. The last part of this chapter provides a comparative analysis of all the three implementations with respect to certain network critical parameters and inferences drawn based on the research findings.