Browsing by Author "Shamrat, F.M. Javed Mehedi"

Now showing 1 - 13 of 13

An Automated Smart Embedded System on Fire Detection and Prevention for Ensuring Safety
(Scopus, 2021) Shamrat, F.M. Javed Mehedi; Khan, Aliza Ahmed; Sultana, Zakia; Imran, M. M.; Abdulla, Abdulla; Khater, Ankit
One of the biggest issues for architects, planners, and landowners is house combustion. Singular sensors have been used in the case of a fire for a long time, but they cannot quantify the volume of fire to warn emergency service units. To resolve this problem, this research aims to develop an intelligent smart fire warning system that detects fires utilizing connected sensors and alerts property owners, emergency services. The current model is divided into three modules: Smoke Detection Module (SDM), which is responsible for detecting smoke to prevent unwanted incidents; Notification Send Module (NSM), which is responsible for creating an alert service to alert the closest support center and user; and Emergency Alarm Module (EAM), which is responsible for handling the emergency alarm schedule when a fire arises. The results prove that the device worked well, and it should be remembered that our proposal can be integrated into any kind of setting, such as a house, workplace, ship, or industry.
An Automation and Temperature Prediction on Smart AC System
(Daffodil International University, 2018-10-27) Shamrat, F.M. Javed Mehedi
A model of an automated temperature prediction on smart AC system for a room has been designed, developed and implemented with an embedded system. In a room, temperature of object (like human being) with the environment is detected, identified and analyzed, with an ideal temperature. Based on data, a mathematical formula can be derived and an algorithm has been formed by using the mathematical formula of the predicted temperature data and the values of the two sensors, where sensors are used for object temperature detection and the AC perform automatically turned on or turned off. Python programming language with its default library has been used to code for the successful implementation of the algorithm. This proposed embedded system can be implemented in any smart AC room where anyone can utilize the AC system automatically switched on/off with the predicted temperature. Exploit this embedded system in all over the places including for disabled peoples, personal room, conference room, hall room, classroom and transports, where manually control of Air conditioner is not feasible.
An Offline and Online-based Android Application ―travelhelp to Assist the Travelers Visually And Verbally for Outing
(International Journal of Scientific and Technology Research, 2020) Shamrat, F.M. Javed Mehedi; Rahman, A.K.M Sazzadur; Tasnim, Zarrin; Hossain, Syed Akhter
TravelHelp" is an android application thatwill be developed mainly to help tourists who are eager to visit Dhaka city. The application is mainly featured to be an offline app so that the user can use the app without any hassle to acquire data services. Both English and Bangla language are supported so that not only foreigners but also the local people can also use the app flexibly. The app is developed to be user-friendly and to support different screen sizes of various mobile and tablet devices. It has features like details about the point of attractions and map of the Dhaka city. Other features include a language translator in the app. This paper details the development process of the travel application. Thisthesis is intending to design & development the idea of a user-friendly android-based offline mobile and web-based application. The travel application will be developed to provide information about the point of attractions and map of the Dhaka city. The map will consist of the regional main attractions, traveling path and assumed expense. An extension of the work is including an instant voice translator in the app which might be very helpful for the foreign travelers who are traveling the Dhaka city. An automatic voice responder based on AI (artificial intelligence) is also planned to be included in the application. By using this tool, the foreigner can be easily able to communicate with the local people which will lead to a very convenient travel.
Comparative Analysis of Human Face Recognition Using SURF and Neural Network Methods
(Scopus, 2021) Shamrat, F.M. Javed Mehedi; Bhowmik, Shohag Kumar; Muntasim, Mst. Fahmida; Nibir, Tafsirul Islam; Chowdhury, Tahmid Rashik; Thapa, Sittal
In computer vision, facial recognition technology is used to recognize every person. This approach is a revolution that is perfect for analyzing a graphic image or a video frame specially or differently. There are, however, systemic methods where facial appreciation schemes initiative, typically, and the effort by comparing chosen facial features from a specific image with faces in a database. It's also known as a Biometric Artificial Intelligence-based app that can see a person in extraordinary detail by dissecting structures related to their facial exteriors and figures. The professional employs a variety of techniques in order to complete the mission. The SURF and Neural Network methods are two of these methods. The writers of this paper address the methods mentioned above and how they operate. The emphasis of the debate is on the methods' accuracy rates and determining which approach produces the most reliable outcome based on facial image results.
Comparative Analysis to identify the best Classifier for Parkinson Prediction
(Scopus, 2021) Shamrat, F.M. Javed Mehedi; Bhowmik, Shohag Kumar; Sultana, Zakia; Hossain, Ahbab; Amina, Mahdia; Thapa, Sittal
Parkinson disease has become one of the most common diseases among people over the age of 65. Neurodegenerative disease affects movement, speech and other cognitive abilities. Among patients, the symptoms vary at a different rate, for which diagnosis of the disease sometimes takes years by when treatment is no longer an option. However, using machine learning algorithms to classify the symptoms among patients, it is possible for early detection of the disease. İn this paper, the performance of machine learning algorithms are measures that can detect Parkinson disease. Three different datasets are used for the study. Each dataset goes through various feature selection techniques. Machine learning classifiers such as KNN, LDA, NB, LR, SVM, DT, RT, RF and ANN are implemented on the datasets and their performance is measured. It is observed that SVM has a high accuracy rate of prediction over all the feature selection techniques in all the datasets.
Face Mask Detection Using Convolutional Neural Network (CNN) to Reduce the Spread of Covid-19
(2021 5th International Conference Trends in Electronics and Informatics (ICOEI), IEEE, 2021-06-21) Shamrat, F.M. Javed Mehedi; Chakraborty, Sovon; Billah, Md. Masum; Jubair, Md. Al; Islam, Md Saidul; Ranjan, Rumesh
The COVID-19 coronavirus pandemic is wreaking havoc on the world's health. The healthcare sector is in a state of disaster. Many precautionary steps have been taken to prevent the spread of this disease, including the usage of a mask, which is strongly recommended by the World Health Organization (WHO). In this paper, we used three deep learning methods for face mask detection, including Max pooling, Average pooling, and MobileNetV2 architecture, and showed the methods detection accuracy. A dataset containing 1845 images from various sources and 120 co-author pictures taken with a webcam and a mobile phone camera is used to train a deep learning architecture. The Max pooling achieved 96.49% training accuracy and validation accuracy is 98.67%. Besides, the Average pooling achieved 95.190/0 training accuracy and validation accuracy is 96.23%. MobileNetV2 architecture gained the highest accuracy 99.72% for training and 99.82 % for validation.
Fruitseg30_segmentation Dataset & Mask Annotations: A Novel Dataset for Diverse Fruit Segmentation and Classification
(Elsevier, 2024-08-10) Shamrat, F.M. Javed Mehedi; Shakil, Rashiduzzaman; Idris, Mohd Yamani Idna; Akter, Bonna; Zhou, Xujuan
Fruits are mature ovaries of flowering plants that are integral to human diets, providing essential nutrients such as vitamins, minerals, fiber and antioxidants that are crucial for health and disease prevention. Accurate classification and segmentation of fruits are crucial in the agricultural sector for enhancing the efficiency of sorting and quality control processes, which significantly benefit automated systems by reducing labor costs and improving product consistency. This paper introduces the “FruitSeg30_Segmentation Dataset & Mask Annotations”, a novel dataset designed to advance the capability of deep learning models in fruit segmentation and classification. Comprising 1969 high-quality images across 30 distinct fruit classes, this dataset provides diverse visuals essential for a robust model. Utilizing a U-Net architecture, the model trained on this dataset achieved training accuracy of 94.72 %, validation accuracy of 92.57 %, precision of 94 %, recall of 91 %, f1-score of 92.5 %, IoU score of 86 %, and maximum dice score of 0.9472, demonstrating superior performance in segmentation tasks. The FruitSeg30 dataset fills a critical gap and sets new standards in dataset quality and diversity, enhancing agricultural technology and food industry applications.
Human Face Recognition Using Eigenface, SURF Method
(Springer, 2022-01-01) Shamrat, F.M. Javed Mehedi; Ghosh, Pronab; Tasnim, Zarrin; Khan, Aliza Ahmed; Uddin, Md. Shihab; Chowdhury, Tahmid Rashik
One such complicated and exciting problem in computer vision and pattern recognition is identification using face biometrics. One such application of biometrics, used in video inspection, biometric authentication, surveillance, and so on, is facial recognition. Many techniques for detecting facial biometrics have been studied in the past three years. However, considerations such as shifting lighting, landscape, the nose being farther from the camera, the background being farther from the camera creating blurring, and noise present renders the previous approaches bad. To solve these problems, numerous works with sufficient clarification on this research subject have been introduced in this paper. This paper analyzes the multiple methods researchers use in their numerous researches to solve different types of problems faced during facial recognition. A new technique is implemented to investigate the feature space to the abstract component subset. Principle component analysis (PCA) is used to analyze the features and uses speed up robust features (SURF) technique, eigenfaces, identification, and matching is done, respectively. Thus, we get improved accuracy and almost similar recognition rate from the acquired research results based on the facial image dataset, which has been taken from the ORL database.
Implementation of Deep Learning Methods to Identify Rotten Fruits
(2021 5th International Conference on Trends in Electronics and Informatics (ICOEI), IEEE, 2021-06-21) Chakraborty, Sovon; Shamrat, F.M. Javed Mehedi; Billah, Md. Masum; Jubair, Md. Al; Alauddin, Md.; Ranjan, Rumesh
Mostly in the agriculture sector, identifying rotten fruits has been critical. The classification of fresh and rotting fruits is typically carried out by humans, which is ineffective for fruit growers. Humans wear out by doing the same role many days, but robots do not. As a result, the study proposed a method for reducing human effort, lowering production costs, and shortening production time by detecting defects in agricultural fruits. If the defects are not detected, the contaminated fruits can contaminate the good fruits. As a result, we proposed a model to prevent the propagation of rottenness. From the input fruit images, the proposed model classifies the fresh and rotting fruits. We utilized three different varieties of fruits in this project: apple, banana, and oranges. The features from input fruit images are collected using a Convolutional Neural Network, and the images are categorized using Max pooling, Average pooling, and MobileNetV2 architecture. The proposed model's performance is tested on a Kaggle dataset, and it achieves the highest accuracy in training data is 99.46% and in the validation set is 99.61% by applying MobileNetV2. The Max pooling achieved 94.49% training accuracy and validation accuracy is 94.97%. Besides, the Average pooling achieved 93.06% training accuracy and validation accuracy is 93.72%. The findings revealed that the proposed CNN model is capable of distinguishing between fresh and rotting fruits.
Implementation of Machine Learning Algorithms to Detect the Prognosis Rate of Kidney Disease
(2020 IEEE International Conference for Innovation in Technology (INOCON) , IEEE, 2020-11) Shamrat, F.M. Javed Mehedi; Ghosh, Pronab; Sadek, Mahbubul Hasan; Kazi, Md. Aslam; Shultana, Shahana
The chronic kidney disease is the loss of kidney function. Often time, the symptoms of the disease is not noticeable and a significant amount of lives are lost annually due to the disease. Using machine learning algorithm for medical studies, the disease can be predicted with a high accuracy rate and a very short time. Using four of the supervised classification learning algorithms, i.e., logistic regression, Decision tree, Random Forest and KNN algorithms, the prediction of the disease can be done. In the paper, the performance of the predictions of the algorithms are analyzed using a pre-processed dataset. The performance analysis is done base on the accuracy of the results, prediction time, ROC and AUC Curve and error rate. The comparison of the algorithms will suggest which algorithm is best fit for predicting the chronic kidney disease.
Industrial Fault Detection Using Transfer Learning Models
(Scopus, 2021) Chakraborty, Sovon; Shamrat, F.M. Javed Mehedi; Afrin, , Saima; Saha, Shaikat; Ahmed, Ishtiak; Thapa, Sittal
Industry and equipment are critical factors in the advancement of human society in the era of the industrial revolution. Since factories are reliant on their machines, they must be maintained daily. However, if the machines are too large for us to observe, an automated process is required to monitor it. By diagnosing the signal data using the CNN algorithm, faults in the machines can be identified. This paper has proposed three transfer learning-based fault diagnosis models using AlexNet, InceptionV3, GoogLeNet with the pretrained weights of the ImageNet dataset. The results of the classification of the three models are compared for their performance. It is observed from the study that the proposed AlexNet architecture shows a very high performance by classifying faults in machines for the tested dataset compared to other models.
Multiple Cascading Algorithms to Evaluate Performance of Face Detection
(Springer, 2022-01-01) Shamrat, F.M. Javed Mehedi; Tasnim, Zarrin; Chowdhury, Tahmid Rashik; Shema, Rokeya; Uddin, Md. Shihab; Sultana, Zakia
This paper intends to evaluate the previous works done on different cascading classifiers for human face detection of image data. This paper includes the working process, efficiency, and performance comparison of different cascading methods. These methods are dynamic cascade, Haar cascade, SURF cascade, and Fea-Accu cascade. Each cascade classifier is described in this paper with their working procedure and mathematical induction as well. Each technique is backed with proper data and examples. The accuracy rate of the method is given with comparison with analyze the performance of the methods. In this literature, the human face detection process using cascading classifiers from image data is studied. From the study, the performance rate and comparison of different cascading techniques are highlighted. This study will also help to determine which methods are to be used for achieving an accurate accuracy depending on the data and circumstances.
The Impact of Software Fault Prediction in Real-World Application
(Scopus, 2020) Ahmed, Md. Razu; Ali, Md. Asraf; Ahmed, Nasim; Zamal, Md. Fahad Bin; Shamrat, F.M. Javed Mehedi
Software fault prediction and proneness has long been considered as a critical issue for the tech industry and software professionals. In the traditional techniques, it requires previous experience of faults or a faulty module while detecting the software faults inside an application. An automated software fault recovery models enable the software to significantly predict and recover software faults using machine learning techniques. Such ability of the feature makes the software to run more effectively and reduce the faults, time and cost. In this paper, we proposed a software defect predictive development models using machine learning techniques that can enable the software to continue its projected task. Moreover, we used different prominent evaluation benchmark to evaluate the model's performance such as ten-fold cross-validation techniques, precision, recall, specificity, f 1 measure, and accuracy. This study reports a significant classification performance of 98-100% using SVM on three defect datasets in terms of f1 measure. However, software practitioners and researchers can attain independent understanding from this study while selecting automated task for their intended application.