Browsing by Author "Hossain, Shahriar"

Now showing 1 - 12 of 12

Core Networking System and NOC of ICX at VOICETEL Ltd.
(East West University, 2018-12-19) Hossain, Shahriar
This report depends on the core system activities of ICX department of VOICETEL Ltd. It is an Interconnection Exchange (ICX) operator. They provide services on routing/switching inter operator domestic voice calls, routing/ switching international calls between ANS and IGW, ENUM, IMEI and number portability services, special code number, emergency number, call center number etc. They have three branches, main office is in Dhaka, others are in Chittagong and Khulna. They are performing their network management successfully. I have worked in their core part of NOC department. For the greater part of the designing understudy it is vital to deal with a professional workplace. I have adapted such a significant number of things, similar to how to deal with the function weight despite a modern domain. In this report I have clarified my work encounters. I have figured out how they deal with their function routine and how they check the association. At the point when association gets flopped then they have shown me how to deal with the circumstance and changed the association with other dynamic system.
Design of a solar charge controller set point and lead acid battery capacity tester
(University of Dhaka, 2016-05-08) Rayhan, Tanvir; Hossain, Shahriar
This aim of this thesis is the designing of testing instrument for SHS. SHS is a growing sector in our country. In every SHS batteries and a DC charge controller are used. But capacities of those batteries or the performance of the DC charge controllers are not always verified before use. Due to that the system performance may become a problem for the consumers. These instruments can measure the capacity of a battery and sense the different set-points of a DC charge controller. From the measured data the capacity of the battery can be measured and can be decided whether it’s good enough to use in a SHS or a battery which is already being used should to be replaced according to standard specification. By sensing the set-points of a DC Charge Controller we can compare with the standard set points and decide if the Charge Controller is good enough to use in a SHS. The purpose of this project is to determine the quality of the batteries and the DC charge controllers can be measured before using it on a SHS, so the consumer can be secured from a faulty system. We have used “Arduino IDE”, open source software is used for the programming and communication interface between the computer and the data acquiring circuits. The developed systems have been in the laboratory of IE. It is found that the developed instrument works properly.
Development of Home Automation System Using Voice Command
(Daffodil International University, 23-02-12) Hossain, Shahriar; Rahman, Asifur
Home automation system has gained popularity nowadays. Considering this, we have developed a home automation system based on voice control. There are many advanced voice-controlled home automation systems like “Alexa” from Amazon, “Cortana” from Microsoft, and “Google Home” from Google available in the market but there is no machine learning approach based on machine learning which works on the CNN algorithm. This can greatly improve the convenience and efficiency of managing a home, as it allows users to control various devices and systems remotely and automate certain tasks. The accuracy of CNN is very high so we have worked on this algorithm to recognize voice commands. The system uses natural language processing techniques to interpret user inputs and perform corresponding actions on connected devices. Many algorithms work on speech recognition like VAD, SBN, PLP features, Deep neural networks, discrimination training, and WFST framework but we thought that CNN with ML will deliver more accurate results in this field. We have collected our voice dataset from the internet & processed using Python & ML. After everything, we got 81% accuracy of that dataset but we think that we can get more optimal results in the future if we can do proper use of this algorithm & ML.
Diabetic retinopathy detection and classification by using deep learning
(BRAC University, 2022-01) Hossain, Shahriar; Evan, Md. Nurusshafi; Farhin, Fariya Zakir; Nabil, Mashrur Karim; Sadman, Sameen; Chakrabarty, Amitabha
Eyes are the most sensitive part of a human being and it is one of the most challenging tasks for a computer-aided system to classify its diseases. Many visionthreatening diseases such as, Glaucoma and Diabetic Retinopathy are treated using digital fundus imaging and retinal images by the specialist at a primary level. However, a computer-aided system that can classify if the eye has a disease or not could be a handy tool for the specialists and a challenging task for computer aided system developers. A branch of machine learning which is deep learning is making a revolutionary impact on medical diagnosis using image processing and pattern recognition. Therefore, we aim to make use of some Convolutional Neural Network (CNN) architectures such as ResNet50, Inception V3, Xception, DenseNet-169 and MobileNetV3 Large to extract the features and classify if the eye has a disease or not using digital fundus photography and retinal image. For our research, we used a competition dataset available from Kaggle [1] and another dataset from IDRiD [2]. Our final dataset contained a total of 2,517 images with each stage having around 500 images in them. Upon training and testing the selected architectures, we have found that Inception V3 has an accuracy of 86.31% and 87.7% (with a lowered learning rate). Similarly for Xception, we attained 86.9% accuracy with default learning rate and 87.9% accuracy with lowered learning rate. ResNet50 gave an accuracy of 46.83%, MobileNetV3 Large gave the lowest accuracy standing at 23.81%. DenseNet-169 gave us the highest accuracy among all other models, soaring at 88.29% accuracy.
Effect of Supportive Treatment along with Antibiotic in the Treatment of Community Acquired Pneumonia(CAP) of Children Admitted in a Tertiary Level Hospital in Dhaka
(East West University, 6/9/2011) Hossain, Shahriar
The Community-acquired Pneumonia one of the most common illness and can affect people of all age. The aim of the study was to establish the supportive treatment along with antibiotic which will help doctor to improve disease condition and patient relief from Pnewnonia. This is a antibiotic study between the Department of Pharmacy, East West University and Institute of child health and Shisue Shasthy Foundation (lCH & SSF) which carried out from August 2007 to december 2010. This is a descriptive study in which 70 children suffering from Community-acquired Pneumonia (2 month-5 years). Among 70 children, 66% & 34% were having community Acquired Pneumonia (CAP). Among the 70 children 51 % children received oxygen 76% received Nebulization, 45% NG Fluid, 25% received Bronchodilator and 15% recieved IV Fluid along with antibiotic treatment in case of suffering from Community-Acquired Pneumonia (CAP).1his study shows that Among the 70 children 90% children received BCG children received DPT, 99% children received Polio, 24% children received Measles, 34% children received MMR, 8% children received Hepatitis and 3% children received other vaccines in case of Community Acquired Pnewnonia (CAP).
MediTrack: A Web Based Application for Pharmacy Management System
(Daffodil International University, 2025-09-17) Hossain, Shahriar
This project introduces Meditrack, a comprehensive web-based application designed to revolutionize business in pharmacies by integrating inventory management, billing, customer records, and an online buy-sell system. A huge number of pharmacies in Bangladesh are run manually, which leads to errors in stock calculations, time delays in obtaining sales information, and customer credit management problems. Meditrack addresses these issues by providing a single solution through which pharmacies can manage medicine, note purchases, generate invoices, maintain customer credits, and allow customers to purchase medicines online. The application is built using React.js for the frontend, Node.js and Express for the backend, and MongoDB as the database, with inherent provisions for VAT, discount, and unit calculations. Testing with sample data found Meditrack to be consistent, accurate, and user-friendly. It reduces manual labor, enhances efficiency, and makes pharmacy services available to online clients, with potential scope for advanced reporting, analytics, and role-based access in the future.
Prospect of Doripenem in Bangladesh Pharma Market
(East West University, 12/23/2009) Hossain, Shahriar
Doripenem is a novel carbapenem drug. It is broad spectrum antibiotic. Effort was given to find about the characteristics of the drug and whether it is suitable to launch in the Bangladesh market. Information was gathered from website, journal and books. For market fisibility study IMS, MIMS and QUIMP was observed. It was found that doripenem is non inferior to meropenem, imipenem and ertapenem. It can be used in the complicated urinary tract infections and complicated intra abdominal infections. Doripenem is not in Bangladesh yet. If it is launched doripenem market size would be 10,00,000 taka and its growth would be 10% first year, 25% second year, 15% in third year. From the information from market fisibilty study, swot analysis and product positioning it may be profitable to launch doripenem in Bangladesh.
RansomListener: Ransom call sound investigation using LSTM and CNN Architectures
(BRAC University, 2020-12) Rahman, Rafeed; Rahman, Mehfuz A; Hossain, Shahriar; Hossain, Sajid; Milon, Md.Iqbal Hossain; Akhond, Mostafijur Rahman
Getting calls for ransoms are common phenomena in kidnapping and abduction related incidents where the life of the victim remains extremely vulnerable. These phone calls are often analyzed in real-time by law enforcement authorities to quickly identify the suspects and get crucial information for quick action. However, it is often difficult to manually analyze those phone calls due to the quality of sounds and the presence of several background noises. Even with much high-end software in their inventory, it is futile to accurately refine the incoming calls as it takes a huge amount of time to declutter the different layers of noises in the call. This paper proposes a model based on deep convolutional neural network and signal processing for automatic classification of crucial sounds in ransom related phone calls. We have proposed LSTM and 2D CNN customized models and compared their outputs with VGG16 and AlexNet. Moreover, this paper also presents a unique dataset of different sounds in terms of voices like male or female and the environmental sounds where the victim might be in which can be a probable clue for investigation purposes consisting of 17650 audio clips collected from verified online sources. Finally, the models produced very high classification accuracy with the accuracy of LSTM reaching around 93.4%.
Study on Customer Service Quality and Customer Satisfaction at Credit Card in the Context of Prime Bank Limited in Bangladesh
(East West University, 12/21/2008) Hossain, Shahriar
Credit Card is an electronic based plastic card bearing an account number assigned to a cardholder with a credit limit that can be used to purchase goods and pay for services with a credit facility and without cash transaction. TIle Prime Bank Ltd. (PBL) has started its credit card business in 1999 through Master-card and VISA card. The report titled "Study on Customer Service Quality and Customer Satisfactiop ilt credit card. in the context of Prime Bank limited in Bangladesh is prepared with a view to provide the bank some valuable after sale infOrrn$on in this business. The broad objective of the study is to draw an overall view of the satisfaction level of the credit card user of the Prime Bank Ltd. and then identify how to improve the level, The specific focuses of the study are analysis of the satisfaction level of the card users of The Prime Bank Ltd" comparing their product and service offerings with other issuers, judgitlg cardholders' level of satisfaction, and explorip.g the opportunities and competitive advantages that can be exploited by PBL. Both primary and secondary sources of data have been used to gather the necessary information for the analyses of the studY. Primary data sources are the cardholders (convenient sample size of 77), merchant (sample size of 20) and 4 issuers namely Standard Chartered Bank (Sea), Prime Bank Limited (PBL), National Bank through questionnaire survey. In the market share analysis, at present, SCB has the highest market share of 69% with around 200,000 cardholders (as on 15th April, 05). Limited (NBL) and Dhaka Bank Limited (DBL), AU the primary data sources are reached.
Traffic Management System Traffic management system with vehicle detection and counting vehicle detection and Counting
(Daffodil International University, 2021-10-25) Hossain, Shahriar; Azam, Md. Taskin Mostofa; Siddika, Fahima
"Traffic Management System with vehicle detection and counting” is a research-based initiative with the primary purpose of detecting, tracking, and classifying automobiles, but it can also be applied to driver behavior detection, lane recognition, and other applications. This framework can be used in a variety of domains, including public safety, accident detection, vehicle detection, theft detection, parking lots, and human identification. It can also be used to locate criminals on the road and traffic rule violators so that traffic controllers can take swift action. People are expanding in number, and vehicles are increasing in number as well. Due to a growth in the number of automobiles, highways and roadways are becoming overcrowded. As a result, the frequency of accidents and violations of traffic laws has skyrocketed. For traffic managers, vehicle detection and counting become essential. As a result, we suggested a traffic management system framework. Our work is mostly based on a video-based technique for vehicle recognition and counting that employs the Python programming language OpenCV. Visual Studio Code was used to create and implement the framework for this article. To achieve real-time automatic vehicle detection and counting, software was combined with Intel's OpenCV video stream processing system. This framework can quickly recognize and track automobiles, as well as assist in the counting of objects
Traffic Management System with Vehicle Detection and Counting
(Daffodil International University, 2021-10-30) Hossain, Shahriar; Azam, Md. Taskin Mostofa; Siddika, Fahima
"Traffic Management System with vehicle detection and counting” is a research-based initiative with the primary purpose of detecting, tracking, and classifying automobiles, but it can also be applied to driver behavior detection, lane recognition, and other applications. This framework can be used in a variety of domains, including public safety, accident detection, vehicle detection, theft detection, parking lots, and human identification. It can also be used to locate criminals on the road and traffic rule violators so that traffic controllers can take swift action. People are expanding in number, and vehicles are increasing in number as well. Due to a growth in the number of automobiles, highways and roadways are becoming overcrowded. As a result, the frequency of accidents and violations of traffic laws has skyrocketed. For traffic managers, vehicle detection and counting become essential. As a result, we suggested a traffic management system framework. Our work is mostly based on a video-based technique for vehicle recognition and counting that employs the Python programming language OpenCV. Visual Studio Code was used to create and implement the framework for this article. To achieve real-time automatic vehicle detection and counting, software was combined with Intel's OpenCV video stream processing system. This framework can quickly recognize and track automobiles, as well as assist in the counting of objects.
Transforming Bangladesh agriculture: AI for precision crop disease management
(BRAC University, 2025-06) Hossain, Shahriar; Nahin, Al-Zaber; Hassan, Tasnuva; Haque, Zarif Ayman; Farin, Nusrat Jahan; Hossain, Muhammad Iqbal
The agricultural sector encompasses a large chunk of the economy of Bangladesh as it has the necessary preconditions and factors to be suitable for agriculture. Agriculture is wholly at the whims of the environment and associated natural factors. Innovations from the time man has mastered the art of farming have allowed us to have in control some factors to ensure the desired output however there remains room for improvement and innovation especially in regards to disease detection. Currently even with a large agricultural sector, the methods for disease detection and risk management are lacking due to the inefficiencies in the system which can be very costly. To mitigate this technological innovations such as machine learning and image processing can be used to combat visible signs of disease and achieve early detection. In this paper we have explored the current options available and what can be done to make it suitable to our conditions, which ones are the best for our problem and finally we have proposed a solution we deem feasible. In our reviewed past works we have come across three models, namely Xception, VGG19 and ResNet50 which perform the best for our use cases, giving us the best results for leaf disease detection. These models have been implemented with a transfer learning approach to achieve the best results. Finally we have created a hybrid model approach combining Xception and a Vision Transformer to get the advantage of both a CNN and a Transformer to achieve the best result for our purpose.