Browsing by Author "Asaduzzaman"

Now showing 1 - 10 of 10

Automatic Exam Mark Input for DIU Faculty Members
(Daffodil International University, 2019-11-10) Sohel, Amir; Juwal, Almabud; Mahbub Al Hasan, Md.; Asaduzzaman
Automatic exam mark input for DIU faculty members” is a Mobile Application for teachers of DIU. This mobile application will help teachers to entry exam marks of students into DIU portal on two click only. They need not input marks manually by typing ID anymore. It’s only two click away to update exam marks to portal. It will save a lot of time of teachers. To use this app teachers must login first with their portal ID and Password given by DIU. After that it will provide a mobile interface where teacher will just scan the cover page of exam paper and upload it to the system and system will automatically detect the Student ID, Course Code, Semester Name and Marks obtained then system will return processed data to another interface and teacher can be recheck it whether information are correct or not. If there is something mistake then teacher can be edit information. By clicking upload button system will update the marks of students automatically to DIU portal. We processed image using Microsoft Vision API that return text from image. Then system will send student marks and information to portal using API given by DIU. It’s totally secure and simple to entry marks.
Compression of Large-Scale Image Dataset using Principal Component Analysis and K-means Clustering
(Faculty of Electrical and Computer Engineering, CUET, 7-Feb-2019) Rayan, Rushrukh; Hossain, Md. Sabir; Asaduzzaman
Digital images, being on the verge of its utmost
Design an efficient MAC Protocol for Cognitive Radio Systems using Game Theory
(Department of Computer Science and Engineering, CUET, 21-Nov-2013) Saha, Debbrata Kumar; Asaduzzaman; Rahman, Mohammad Obaidur
In this paper we consider power distribution between cognitive users which is the most crucial problems in cognitive radio systems. To ensure perfect distribution game theory is applied. Conflicting users who are trying to access the same channel are detected first. Targeted signal to noise ratio (SNR) is achieved by iterative fashion which determines the signal strength. Power is calculated for each user by convergence theorem. Power is distributed in such way that it increases system utility. At last Nash equilibrium point is formulated for ensuring the avoidance of selfish behavior of cognitive users. Experimental results show that the proposed strategy achieves the ability to reduce the power consumption in order to increase the system utility.
Detection of Parkinson’s disease from Neuro-imagery using deep neural network with transfer learning
(BRAC University, 2020-04) Asaduzzaman; Sakib, A.F.M. Nazmus; Shusmita, Sanjida Ali; Kabir, S. M. Ashraf; Parvez, Mohammad Zavid; Reza, Md. Tanzim
Parkinson’s disease is a neurological condition that is dynamic and steadily influences the movement of the human body. It causes issues within the brain and slowly increments time by time. Tremor is the major side effect of PD where the entire body begins shaking. Besides, a person’s muscle may end up rigid or stiff and it may happen any portion of his body. PD influences the central apprehensive system which is happening because of the hardship of dopaminergic neurons brought about in a neuro-degenerative incubation. It is grouped beneath advancement clutter as patients who have PD appear with tremor, unyielding nature, postural shifts, and lessen in unconstrained advancements. There is no particular diagnosis process for PD. PD varies from one person to another person and the situation and history. MRI, CT, ultrasound of the brain, PET scans are common imaging tests to figure out this disease but these tests are not particularly effective. In this research, several tests ran on two types of data group - control and PD affected people. The dataset is collected from the Parkinson’s Progression Markers Initiative (PPMI) repository. Then MRI slices are processed from selected data group into the CNN models. Three CNN models are sent into this thesis work to extract features from the data group. The CNN models are InceptionV3, VGG16 and VGG19. These models are used in this research to compare and get better accuracy. Among these models VGG19 worked best in the dataset because the accuracy for VGG19 is 91.5% where VGG16 gives 88.5% and inceptionV3 gives 89.5% on detecting PD.
Handwritten Bangla Numeral and Basic Character Recognition Using Deep Convolutional Neural Network
(Faculty of Electrical and Computer Engineering, CUET, 7-Feb-2019) Hakim, S M Azizul; Asaduzzaman
In this paper, the problem of recognizing handwritten
IOT Based on a Human Security System
(North South University, 2020-08-30) Asaduzzaman; Md Daud Hossain; Zannatul Ferdouse Zerin; Tanzina Akther Momo; Dr. Md Shahriar Karim
This world is running by innovation. A mind-blowing estimation of the world growing step by step. So, security is the fundamental concern. We have made a device which can detect any unwanted condition automatically by taking body movement data and can sent an alert with victim’s location to the preset number. It’s a IOT base system that control massaging from a server. We made a hardware that continuously receive body movement data and analyses the data set. If the analyzation gives any unwanted situation, the device sent a signal to the server and server sent massage to the preset number. This device analyses two type of body movement data like acceleration and gyroscope that means this device can detect irregular body movement and angle and sent the signal. This device is very helpful for women because the number of rape case increasing day by day.
Performance Analysis of Machine Learning Techniques to Predict Diabetes Mellitus
(Faculty of Electrical and Computer Engineering, CUET, 7-Feb-2019) Faruque, Md. Faisal; Asaduzzaman; Sarker, Iqbal H.
Diabetes mellitus is a common disease of human
Spatial variation of house rent in Dhaka city
(Department of Urban and Regional Planning, 2001-10) Asaduzzaman; Jahan, Dr. Sarwar
For abstracts please see full text
Spectral domain speech enhancement method based on noise compensations in both magnitude and phase spectra
(Department of Electrical and Electronic Engineering (EEE), 2012-07) Asaduzzaman; Shahnaz, Dr. Celia
In this thesis, a noisy speech enhancement method based on noise compensation performed on short time magnitude as well phase spectra is presented unlike the conventional spectral subtraction method. Here, the noise estimate to be subtracted from the noisy speech spectrum is proposed to be determined exploiting the low frequency regions of noisy speech of current frame rather than depending only on the initial silence frames. We argue that this approach of noise estimation o ers the capability of tracking the time variation of the non-stationary noise thus resulting in a noise compensated magnitude spectrum. By employing the noise estimates thus obtained, a procedure is formulated to compensate the distortion in the phase spectrum,which is kept unchanged in the typical speech enhancement methods. The noise compensated phase spectrum is then recombined with the noise compensated magnitude spectrum to produce a modi ed complex spectrum thus synthesizing an enhanced frame. Extensive simulations are carried out using NOIZEUS database in order evaluate the performance of the proposed method. It is shown in terms of objective measures, spectrogram analysis and informal subjective listening test that the proposed method consistently outperforms some of the state-of-the-art methods of speech enhancement from noisy speech corrupted by white or train or babble noise of very low levels of SNR.
Vibration based smart pavement monitoring system using vehicle dynamics and smartphone
(Department of Civil Engineering (CE), BUET, 2020-11-30) Asaduzzaman; Shohel Rana, Dr.
The main purpose of this research is to study the feasibility of pavement monitoring using the dynamic response of conventional vehicle and smartphone sensors. The estimation of unknown vehicle dynamic parameters is also studied. In this study, the pavement condition has been monitored in terms of International Roughness Index (IRI) as it is used all over the world to measure road roughness as a part of pavement condition monitoring strategy. The conventional vehicle is modeled using a Quarter-Car (QC) vehicle model. For estimation of unknown vehicle dynamic parameters, Grey-Box model algorithm is used to establish a relationship between the input (speed bump) and output (vertical accelerations) of the QC dynamic system. Numerical simulations as well as field tests are performed for unknown vehicle parameter estimations. In numerical simulations, two parameters are estimated within 12.1% of the actual values and another parameter within 22.5%. In field testing, the coefficient of variance for the estimated parameters are found to be within 15.2%. In this study, Inverse State-Space Representation of QC dynamic system is developed for reconstruction of pavement profiles of the unknown pavements to be monitored using vertical acceleration of the vehicle as the input and the pavement profile is obtained as the output. From the reconstructed pavement profile, IRI of the pavement is estimated. The measurements of IRI of the pavements are done both numerically and practically. In numerical simulations, IRI of the reconstructed pavement profiles are found to be within 4.1% of the IRI of the actual profiles. In practical experiment, the proposed method is found to be repeatable and fairly accurate. The results of the study indicate that it is possible to determine IRI of a pavement surface with reasonable accuracy using conventional vehicles and smartphone. The large-scale application of the proposed method will result in a cost effective, quick, flexible and convenient method for pavement monitoring with the potential of including the entire road network into the pavement monitoring system.