Browsing by Author "Khan, Md. Abbas Ali"

Now showing 1 - 9 of 9

A Computer Vision System for Bangladeshi Local Mango Breed Detection using Convolutional Neural Network (CNN) Models
(Scopus, 2020) Haque, A.S. M. Farhan Al; Rahman, Md. Riazur; Marouf, Ahmed Al; Khan, Md. Abbas Ali
Magnifera Indica, traditionally known as mango, is a drupe found around the world in over 500 species. India has produced 19.5 million metric tons of mango in 2017. In Bangladesh, mango has been referred as the national tree and government has included endemic species of mango as geographical index (GI) of Bangladesh. Recognizing specific breeds has become a significant computer vision task. In this paper, we have proposed the convolutional neural network (CNN) based approach for detecting five mango species namely, Chosha, Fazli, Harivanga, Lengra and Rupali from 15000 different images. For better experimentation, we have applied three different models of CNN and analyzed the recognition rates with various criteria. For performance evaluation, we have utilized the classic metrics such as precision, recall, F1-score, ROC and accuracy. Among the experimented three models, the third model, outperformed in terms of accuracy with 92.80%.
A Novel Compound Feature Based Driver Identification
(Daffodil International University, 2022-01-20) Khan, Md. Abbas Ali; Ali, Mohammad Hanif; Haque, AKM Fazlul; Islam, Md. Iktidar; Islam, Mohammad Monirul
Abstract: In today's world, it is time to identify the driver through technology. At present, it is possible to find out the driving style of the drivers from every car through controller area network (CAN-BUS) sensor data which was not possible through the conventional car. Many researchers did their work and their main purpose was to find out the driver driving style from end-to-end analysis of CAN-BUS sensor data. So, it is potential to identify each driver individually based on the driver's driving style. We propose a novel compound feature-based driver identification to reduce the number of input attributes based on some mathematical operation. Now, the role of machine learning in the field of any type of data analysis is incomparable and significant. The state-of-the-art algorithms have been applied in different fields. Occasionally these are tested in a similar domain. As a result, we have used some prominent algorithms of machine learning, which show different results in the field of aspiration of the model. The other goal of this study is to compare the conspicuous classification algorithms in the index of performance metrics in driver behavior identification. Hence, we compare the performance of SVM, Naïve Bayes, Logistic Regression, k-NN, Random Forest, Decision tree, Gradient boosting.
Achieving Robust Global Bandwidth Along with Bypassing Geo-Restriction for Internet Users
(Scopus, 2019) Islam, Gazi Zahirul; Emran, Al- Nahian Bin; Juman, Aman Ullah; Khan, Md. Abbas Ali; Hossain, Md. Fokhray; Habib, Md. Tarek
Not all Internet Service Providers provide a sufficient amount of bandwidth to their users. Although the amount of local bandwidth is reasonable, global bandwidth is not satisfactory at all. Based on bandwidth allocation, location and price; service providers capped their users’ global bandwidth i.e., reducing global internet speed. As a consequence, we observe severe global bandwidth limitation among Internet users. In this article, we implement a flexible and pragmatic solution for Internet users to bypass global bandwidth restriction. To achieve robust global bandwidth, we utilize a combination of communication technologies and devices namely, Internet Exchange Point, Virtual Private Network, chain VPN technology etc. In this project, we show that internet speed of international route i.e., global bandwidth can enhance significantly if there are multiple ISPs use a common IXP and at least one of those ISPs provides pleasant global bandwidth. Usually, regional ISPs use a common IXP to route their local traffic using local bandwidth within the region without wasting global bandwidth. We show that using our proposed method global internet speed of a user can raise several times effectively utilizing assigned local bandwidth. In addition, we also implement a geo-restriction bypassing technique integrating an offshore ISP with local ISP using VPN. Thus, we enjoy tremendous Internet speed along with unrestricted access to the websites.
An App-Based IoT-NFC Controlled Remote Access Security Through Cryptographic Algorithm
(Scopus, 2021) Khan, Md. Abbas Ali; Ali, Mohammad Hanif; Haque, A. K. M. Fazlul; Debnath, Chandan; Jabiullah, Md. Ismail; Rahman, Md. Riazur
In the twenty-first century, a human being is passing through the world with generosity of technology and most of it’s the systems are being operated by automated or remote access control. However, sensor technology is already playing a vital role to control the smart home, smart office, etc. However, it is about to beyond a smart city. Remote access control is a part of the leading technology. An app-based innovative remote access control framework is adding an extra security to make this technology more convenient, secured and illustrate the usability of a person along with an authenticated system of the executive. NFC is used as a communication technology, and a microcontroller camera is also used for detection. An authentication process drives through a smartphone application over the IoT framework. A definitive objective of this paper is to ensure the security of remote access control, notification to the comer and admin, accessibility, usability and permissibility to enter the premises. In order to maintain the integrity and the confidentiality of data cryptographic, techniques like computational 512 bits hash functions are considered and encrypt the hashed data once AES-192 is used. The additional part of this paper is to measure the performance of an employee.
An Efficient and Optimized Tracking Framework through Optimizing Algorithm in a Deep Forest using NFC
(Indonesian Journal of Electrical Engineering and Computer Science, 2020) Khan, Md. Abbas Ali; Ali, Mohammad Hanif; Haque, A.K.M Fazlul; Debnath, Chandan; Bhowmik, Shohag Kumar
NFC is applying in various field of contemporary technology. Especially of convenience tag usability in any place. One of the facilities which can be added in the tracking system is the implementation of Near Field Communication in order to guide each tourist in the deep forest or any other location. In the deep forest, tracking or location detection activities need to be done efficiently, like desired path finding in a deep forest. At present, the tracking procedure in deep forest is working with the help of guides or local citizens. Currently, in any restricted area such as the “Sundarban” forest, no outside general people are allowed to travel in the jungle without any authorized guide which is not an efficient way to travel smoothly. The use of Near Field Communication can solve the problem related to lost the way, safety, and easily help the travelers to track the desired destination without the help of human resources or any guide. The NFC tags that hold mapping information of the area, in the point of tag setup all tags will be set up on several trees along with sequence.
An Investigative Design of Optimum Stochastic Language Model for Bangla Autocomplete
(Indonesian Journal of Electrical Engineering and Computer Science, 2019) Eyamin, Md.Iftakher Alam; Habib, Md. Tarek; Muhammad Ifte Khairul Islam; Rahman, Md. Sadekur; Khan, Md. Abbas Ali
Word completion and word prediction are two important phenomena in typing that have extreme effect on aiding disable people and students while using keyboard or other similar devices. Such autocomplete technique also helps students significantly during learning process through constructing proper keywords during web searching. A lot of works are conducted for English language, but for Bangla, it is still very inadequate as well as the metrics used for performance computation is not rigorous yet. Bangla is one of the mostly spoken languages (3.05% of world population) and ranked as seventh among all the languages in the world. In this paper, word prediction on Bangla sentence by using stochastic, i.e. N-gram based language models are proposed for autocomplete a sentence by predicting a set of words rather than a single word, which was done in previous work. A novel approach is proposed in order to find the optimum language model based on performance metric. In addition, for finding out better performance, a large Bangla corpus of different word types is used.
Deep Learning Based Sponge Gourd Diseases Recognition for Commercial Cultivation in Bangladesh
(International Conference on Artificial Intelligence & Industrial Applications, Springer, 2020-09-02) Mim, Tahmina Tashrif; Sheikh, Md. Helal; Chowdhury, Sadia; Akter, Roksana; Khan, Md. Abbas Ali; Habib, Md. Tarek
Sponge gourd, called as “ Open image in new window ” in Bangladesh, is scientifically known as Luffa cylindrical that belongs to Cucurbitaceae family. Sponge gourd or luffa gourd is one of the most easily found vegetable in Bangladesh. It is an edible wild vegetable that the plant can be seen anywhere around us during late summer till late autumn in Bangladesh. Cooked sponge gourd or the curry is a bit sweetish in taste. Even though sponge gourd is kind of a wild vegetable plant, in recent time a lot of people are cultivating it in the countryside thinking of profit and commercial production since there is a market demand for it. Despite of having every opportunity to commercial benefit most of the farmers neglect the issue of insects and diseases attack on the plant resulting on huge loss in the business. Also lack of proper knowledge of related diseases, advance technology and trustable source the farmers lag behind the diseases detection process to use pesticides or different methods of reducing diseases attack. If necessary steps can be taken to prevent the insects and diseases attack at the very beginning of cultivation, then the profits will increase as the crop yields increase. This research paper attempts to detect the leaf and flower diseases of sponge gourd using Convolutional Neural Network (CNN) and image processing techniques. CNN and image processing are one of the most recently introduced technologies using in the agriculture sector in Bangladesh ensuring highest accuracy. This system will take leaf images as input and after examining them healthy or detected diseases will be shown as output which has diffrent true and false values for different diseases and reached to the average accuracy of 81.52%.
IOT-NFC Controlled Remote Access Security and an Exploration through Machine Learning
(2020 18th International Conference on ICT and Knowledge Engineering (ICT&KE), IEEE, 2020-12-25) Khan, Md. Abbas Ali; Hanif Ali, Mohammad; Haque, A.K.M Fazlul; Sharmin, Farah; Jabiullah, Md. Ismail
Internet of Things (IOT) is a system that allows to connect the computing devices without the help of human-to-human or human-to-computer interaction. This paper proposes an app-based remote access control door lock security system (RACDLS) along with a short-range wireless communication naming Near Field Communication (NFC). The RACDLS system generates two sides' authentication systems rather than one side authentication like conventional systems. In the conventional system, users which are registered, can enter the premises only. In the proposed RACDLS system, users require permission from the room owner either they are registered or unregistered. Moreover, for maintaining the integrity and the confidentiality of data, a cryptographic technique (e.g.) we consider computational 512 bits hash function. In contrast, apply AES-192 for encrypting the hashed data. In addition, machine learning (ML) shows the performance of the employee activities including prediction with model accuracy. A definitive objective of this paper is to ensure the security of remote access control as well as allow notification of both ends, accessibility, usability, and permissibility of a personnel to enter the premises.
Machine Vision Based Potato Species Recognition
(5th International Conference on Intelligent Computing and Control Systems (ICICCS), IEEE, 2021) Nuruzzaman, Md.; Hossain, Md. Shahadat; Rahman, Md. Mostafijur; Shoumik, Ahete Shamul Haque Chowdhury; Khan, Md. Abbas Ali; Habib, Md. Tarek
Potato is one of the tasteful vegetable in the list of our daily delicious food. At present, there are 42000 kind of potatoes available in the world. In Bangladesh, we cultivate 82 species every year. Potatoes are used for other purposes besides eating. So it is produced by thinking of other purposes besides eating. Different varieties of potatoes are used for different purposes. But People usually do not know which variety of potato is suitable for which work. We took this research step to solve this problem. There are currently some conventional common detection methods that are not very convenient. Therefore, we have introduced Machine Vision Recognition (MVR) procedure to discover a suitable technique. As if, through this method people can easily identify the potato species. In this research paper, we want to show how to identify different varieties of potatoes in Bangladesh using machine vision approach. We have been collected total 1200 potato imagers from four fact for our experience. To reach our aim, several machine learning algorithms have been applied to the datasets, like Random Forest Classifier (RF), Linear Discriminant Analysis (LDA), Logistic Regression, Support Vector Machine (SVM), CART, NB, and KNN. Once we have been applied all the algorithms, different results have been shown by each algorithm. Logistic Regression shows the best result, which has an accuracy rate of 98%. In contrast, the lowest rate has been shown by the SVM. The accuracy rate of SVM is 33% which is not only a good fit for future research but also promising.