Chittagong University of Engineering & Technology

About: Chittagong University of Engineering & Technology is a education organization based out in Chittagong, Bangladesh. It is known for research contribution in the topics: Renewable energy & Dielectric. The organization has 1200 authors who have published 1444 publications receiving 10418 citations. The organization is also known as: Engineering College, Chittagong & Bangladesh Institute of Technology, Chittagong.

Handwritten Bangla Numeral and Basic Character Recognition Using Deep Convolutional Neural Network

Abstract: In this paper, the problem of recognizing handwritten Bangla characters is addressed. Handwritten character recognition is one of the most practiced tasks in computer vision. Over the past few years Convolutional Neural Network has produced the best results in case of English handwritten character recognition. Although Bangla the official language of Bangladesh and several Indian states with over 200 million native speakers Bangla handwritten character recognition is quite far behind. We present a 9 layer sequential Convolutional Neural Network model to recognize 60 (10 numerals+ 50 basic characters) Bangla handwritten characters. BanglaLekha-Isolated dataset is used as train-validation set. A new dataset of 6000 images is created for cross validation. Our proposed model trained to recognize 60 characters achieves state-of-the-art 99.44% accuracy on BanglaLekha-Isolated dataset and 95.16% accuracy on prepared test set. Experiments on recognizing Bangla numerals separately also show state-of-the-art performance.

Thermodynamic analysis of an air conditioner using R22/HC blend with TiO2/Mineral oil nanolubricant to retrofit R22/POE oil

Abstract: This research was performed experimentally and determined the first and second law performance (energetic and exergetic) of a split type air conditioner. A blend of R22 and R600a (80:20 by mass) was used as a refrigerant with TiO2/Mineral oil (MO) as a lubricant for a varied concentration of TiO2 (0.1–0.4%) as a drop-in replacement of R22/POE. The obtained data were analyzed to investigate the thermodynamic performance including compression work, compressor discharge temperature, cooling effect, coefficient of performance (COP), and exergy destroyed in the compressor, total exergy destruction, sustainability, and exergy efficiency. The analysis was based on temperature, pressure readings recorded from the appropriate gauge, and the physical and thermal properties of the refrigerants were extracted from REFRPROP 7.0 software. Experimental outcomes showed that both energetic and exergetic performance was improved for a blend with nanolubricants compared to R22. The compressor outlet temperature was found to be decreased by 15 ​°C for the blend with 0.4% TiO2/MO nanolubricant. Again, COP was improved by 10–19.5% for the various concentration of nanolubricant. The exergetic performance was also increased with the increase of concentration of nanolubricants and exergy destruction was reduced by 20.1–37.4% for the nanolubricant. It was found that exergy efficiency was also increased by 38–42.88% compared to R22. Also, exergy destruction and entropy generation were found to be maximum for the compressor compared to other components. Based on this experimental work, it can be concluded that nanolubricant and refrigerant blends worked safely without any modification in this system.

Forecasting COVID-19 Dynamics and Endpoint in Bangladesh: A Data-driven Approach

Abstract: On December 31, 2019, the World Health Organization (WHO) was informed that atypical pneumonia-like cases have emerged in Wuhan City, Hubei province, China WHO identified it as a novel coronavirus and declared a global pandemic on March 11th, 2020 At the time of writing this, the COVID-19 claimed more than 440 thousand lives worldwide and led to the global economy and social life into an abyss edge in the living memory As of now, the confirmed cases in Bangladesh have surpassed 100 thousand and more than 1343 deaths putting startling concern on the policymakers and health professionals; thus, prediction models are necessary to forecast a possible number of cases in the future To shed light on it, in this paper, we presented data-driven estimation methods, the Long Short-Term Memory (LSTM) networks, and Logistic Curve methods to predict the possible number of COVID-19 cases in Bangladesh for the upcoming months The results using Logistic Curve suggests that Bangladesh has passed the inflection point on around 28-30 May 2020, a plausible end date to be on the 2nd of January 2021 and it is expected that the total number of infected people to be between 187 thousand to 193 thousand with the assumption that stringent policies are in place The logistic curve also suggested that Bangladesh would reach peak COVID-19 cases at the end of August with more than 185 thousand total confirmed cases, and around 6000 thousand daily new cases may observe Our findings recommend that the containment strategies should immediately implement to reduce transmission and epidemic rate of COVID-19 in upcoming days

An Efficient Technique of Hemoglobin Level Screening Using Machine Learning Algorithms

Abstract: Hemoglobin (Hb), a very significant parameter for the human body and deficiency of it causes anemia. During pregnancy, menstruation and ICU deficiency of it can be very risky and even caused death. So, it is important to diagnose it continuously. Usually, physicians examine it by conducting a blood test to confirm it is painful, time-consuming and costly. The major concept of this study is to screen Hb levels within a short period of time. In this study, the data of clinical blood Hb levels of a total of 104 people (54 males and 50 females) are collected along with an eye conjunctiva image. The images are taken with a Smartphone camera of constant resolution and lighting. Using MATLAB, image processing method, the percentages of the red, green and blue pixels are extracted. Taking those features, the Hb level is plotted. The 104 data have been split into two sets where the first 81 data for training purposes, the remaining 23 data have been considered for testing. To train the model of 81 data, Multivariate Linear Regression (MLR), Decision Tree (Medium), Linear Support Vector Regression (SVR) are taken and the lowest percentage of error of 11.01% has been found in the Decision Tree (Medium) while testing the 23-test data.