M. Hasan, Azmain Yakin Srizon, Abu Sayeed, Md. Al Mehedi Hasan
{"title":"High Performance Classification of Caltech-101 with a Transfer Learned Deep Convolutional Neural Network","authors":"M. Hasan, Azmain Yakin Srizon, Abu Sayeed, Md. Al Mehedi Hasan","doi":"10.1109/ICICT4SD50815.2021.9396917","DOIUrl":"https://doi.org/10.1109/ICICT4SD50815.2021.9396917","url":null,"abstract":"Numerous models and working schemes have been proposed through decades for the successful recognition of the objects. Significant contributions are notable in the field of object recognition. However, near accurate recognition is still a challenge in this domain. In this research, we considered the Caltech-101 dataset having 102 diverse and imbalanced classes i.e., people, animals, landscapes, structures, furniture, etc. which made the recognition more complicated. We proposed and utilized modified InceptionV3 and modified EfficientNetB6 architectures for the recognition of objects which obtained 99.65% and 99.72% overall accuracy respectively. We further showed via experimental analysis that the softmax-averaging technique can further boost the accuracy to 99.85% and all three proposed procedures suppressed the previous studies by a notable boundary as well.","PeriodicalId":239251,"journal":{"name":"2021 International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114651706","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Analyze the Impact of Demand Side Management on Grid Power for an Isolate Zone in a Sustainable IEEE 14 Bus System","authors":"S. M. Salam, N. Mohammad","doi":"10.1109/ICICT4SD50815.2021.9396975","DOIUrl":"https://doi.org/10.1109/ICICT4SD50815.2021.9396975","url":null,"abstract":"Power Grid is a complex network system for power transmission, generation and distribution which is rapidly changing with time as per user requirement. Computational analyze and representation is a prime requirement to find out the impact of Demand Side Management(DSM) within a network. For the sustainability of the new grid system, it is essential to modify the load requirement for the system model and analyze the system performance for the sustainability of the network. In this paper, the IEEE 14 bus system load model is modified for DSM in a local grid network and the impact of the grid system is analyzed by MATLAB/Simulink software model for different network condition.","PeriodicalId":239251,"journal":{"name":"2021 International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD)","volume":"138 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116355969","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shuvo Chandra Pall, Md. Mehedi Hasan Tanim, M. Jabiullah, Ekhtiak Ebna Ahsan, Md. Galib Hasan
{"title":"BER Performance Analysis of Different O-OFDM Techniques in VLC","authors":"Shuvo Chandra Pall, Md. Mehedi Hasan Tanim, M. Jabiullah, Ekhtiak Ebna Ahsan, Md. Galib Hasan","doi":"10.1109/ICICT4SD50815.2021.9396866","DOIUrl":"https://doi.org/10.1109/ICICT4SD50815.2021.9396866","url":null,"abstract":"Day by day, as people began to use daily communication, visible light communication is changing globally. Visible light communication (VLC) uses a visual modulation method named orthogonal frequency division multiplexing. This paper proposes a comparison of bit error rate (BER) of numerous 16 QAM and 64 QAM hybrid optical OFDM techniques including asymmetrically clipped absolute value optical OFDM (AAO-OFDM), spectral efficiency augmented hybrid asymmetrically clipped optical OFDM (SEA-HACO-OFDM), layered asymmetrically clipped optical OFDM (LACO-OFDM), asymmetrically clipped DC biased optical OFDM (ADO-OFDM) hybrid asymmetrically clipped optical OFDM (HACO-OFDM), asymmetrically clipped DC biased optical OFDM (EADO-OFDM), DC biased optical OFDM (DCO-OFDM), asymmetrically clipped optical OFDM (ACO-OFDM). The principle of each modulation method is demonstrated, and performance is compared in respect of bit error rate. BER is an essential parameter for the characterization of systems quality and efficacy. Signal strength will be better when the bit error rate is too small. Simulation results show that SEA-HACO-OFDM for both cases exhibits the best BER performance in comparison with other modulation techniques.","PeriodicalId":239251,"journal":{"name":"2021 International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125991309","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Noninvasive Heart Rate and Blood Glucose Level Estimation Using Photoplethysmography","authors":"Suvojit Kumar Singha, Mohiudding Ahmad","doi":"10.1109/ICICT4SD50815.2021.9396849","DOIUrl":"https://doi.org/10.1109/ICICT4SD50815.2021.9396849","url":null,"abstract":"Heart rate (HR) and blood glucose level (BGL) are two important indicators of fitness of human health and it is imperative to measure these parameters regularly to avoid severe damages. At present, we need at least two devices such as electrocardiography (ECG) device and invasive BGL measuring devices to measure heart rate and blood glucose level accurately which are costly. In this paper, an easy method has been described to measure both HR and BGL values simultaneously without pricking a finger and without a trained person using a single device only based on photoplethysmography (PPG). In this research work, a pulse sensor is used for PPG signal collection. Then this PPG signal is passed through a bandpass filter for noise removal and amplification to the microcontroller unit (MCU). Then the data obtained from MCU are analyzed using MATLAB for finding HR and BGL values. The peak finding technique is used for finding HR value. For determining BGL values, the dominant feature of the PPG signal is extracted which has the highest correlation with BGL, and then using regression analysis a relationship is established from which we can measure the BGL values without pricking blood. Finally, correlation analysis is also done to analyze the relationship between HR and BGL. As PPG signal acquisition is less costly than ECG and other BGL measuring methods, it will be very helpful for determining HR and BGL values simultaneously at ease. Thus this research work will provide opportunities for early diagnosis of problems related to heart rate and blood glucose level.","PeriodicalId":239251,"journal":{"name":"2021 International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD)","volume":"2013 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125664386","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Imtiaj Ahmed Probal, Fahim Ahmed Hamim, Md. Ashraful Islam, Md. Abu Bakar Siddik
{"title":"Mars Rover and Its Strategic Implementation","authors":"Imtiaj Ahmed Probal, Fahim Ahmed Hamim, Md. Ashraful Islam, Md. Abu Bakar Siddik","doi":"10.1109/ICICT4SD50815.2021.9396853","DOIUrl":"https://doi.org/10.1109/ICICT4SD50815.2021.9396853","url":null,"abstract":"Rovers, essential vehicle for the exploration of solar system, can help scientists in surveillance operation to observe and analyze the structural components in different parts of planet Mars. In this study, we have proposed a prototype of rover which is capable of adapting to different terrains and weather conditions with a view to ensuring functional exploration. Moreover, it is endowed to travel to different areas, collect earth and rock samples, measure pH, determine weight, monitor weather conditions like temperature, humidity, and wind speed, detect the presence of harmful gas as well as carry out rescue operation if the life of an astronaut or cosmonaut is in danger. In addition to this, the wheels of the rover are made of lightweight PVC materials and a special suspension system based on rocker bogie mechanism is incorporated for smooth movement of the rover. Furthermore, integration of global positioning system facilitates distant controlling and autonomous movement in case of the connectivity with the ground station is lost or the circumstance becomes inauspicious. Hence, the proposed prototype has a good prospect in the quest for natural resources and minerals over and above the purpose of exploration and scientific observation.","PeriodicalId":239251,"journal":{"name":"2021 International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD)","volume":"45 11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125690309","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Dementia Identification for Diagnosing Alzheimer's Disease using XGBoost Algorithm","authors":"L. Akter, Ferdib-Al-Islam","doi":"10.1109/ICICT4SD50815.2021.9396777","DOIUrl":"https://doi.org/10.1109/ICICT4SD50815.2021.9396777","url":null,"abstract":"Dementia is an aggregate term used to portray different side effects of psychological decay as oblivion. Around the globe, closely 50 million humans produce dementia, and there are very nearly 10 million fresh cases every year. The roadblock to the clinician is to determine the complex illness, for example, different kinds of dementia, Alzheimer's Disease, and Parkinson's Disease. Uncommonly, Alzheimer's disease is a bit complex to analyze as far as indications as they cover in numerous perspectives at the beginning phase. Along these lines, it is important to examine the cycle of analytic with more enhanced performance with various parameters of the disease. In this paper, we have classified dementia into three classes (AD Dementia, No Dementia, and Uncertain Dementia) for identifying Alzheimer's disease in its beginning phase using Extreme Gradient Boosting (XGBoost) algorithm and also shown the feature importance scores. We got an enhanced performance in terms of accuracy (81%), precision (85%), and other performance metrics, and “ageAtEntry” was the most important feature.","PeriodicalId":239251,"journal":{"name":"2021 International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD)","volume":"2274 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130258452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shammi Akhtar, M. Dipti, Tahasina Afroze Tinni, Pallab Khan, Raihan Kabir, Md. Rashedul Islam
{"title":"Analysis on Handwriting Using Pen-Tablet for Identification of Person and Handedness","authors":"Shammi Akhtar, M. Dipti, Tahasina Afroze Tinni, Pallab Khan, Raihan Kabir, Md. Rashedul Islam","doi":"10.1109/ICICT4SD50815.2021.9397018","DOIUrl":"https://doi.org/10.1109/ICICT4SD50815.2021.9397018","url":null,"abstract":"Human handwriting has some unique properties to express the behaviors and personality of any person. The state-of-the-art systems analyze the handwriting on paper manually and demand human expertness. Moreover, it is very difficult to recognize a person by analyzing the handwriting manually. Thus, this paper proposes a system for person identification along with another system to identify the handedness of individuals using handwriting data analysis. The used handwritten texts for these studies have been collected by the pen-tablet. Six parameters are captured and distinguishable features are extracted to identify the person's handwriting attributes. To identify the person and handedness using extracted features, two different classification algorithms, i.e., Support Vector Machine with linear kernel and Random Forest Classifier, are used. The proposed systems show 69.39% and 72.44 % accuracy using SVM and random forest classifier respectively for person identification and 97.25% accuracy for handedness identification after using SVM.","PeriodicalId":239251,"journal":{"name":"2021 International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD)","volume":"187 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123003086","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Preterm Baby Birth Prediction using Machine Learning Techniques","authors":"M. Begum, Reduanul Momtaj Redoy, Anusree Das Anty","doi":"10.1109/ICICT4SD50815.2021.9396933","DOIUrl":"https://doi.org/10.1109/ICICT4SD50815.2021.9396933","url":null,"abstract":"Premature or preterm is the word that refers to early. Consequently, preterm birth is the birth of a baby inborn earlier than 37 weeks of maternity. Premature birth is one of the foremost important influential factors in infant death. It is approximate 15 million premature babies square measure born every year; growing this number of pre-term babies has a more significant impact on developing countries. As a result, predicting preterm birth may be an active research field. Although pre-term babies have several health issues, early identification of the factors for pre-term delivery will decrease the number of premature babies and mothers will also know the reasons which are causing premature babies. During this research, we've got developed a system to predict pre-term babies. Initially, we tend to study and recognized the main factors corresponding to premature babies with the consultancies of specialized doctors. The main factors are the mother's weight before pregnancy, mother's age, number of the previous preemie, mother's BMI, cervical problem, etc. After that, the dataset is pre-processed and normalized. Finally, four binary classifiers i.e. KNN, Decision Tree, SVM, and Naïve Bayes are trained and tested. The investigation result shows the effectiveness of the projected system with 99% accuracy.","PeriodicalId":239251,"journal":{"name":"2021 International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131824948","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Bengali Sentiment Analysis of E-commerce Product Reviews using K-Nearest Neighbors","authors":"M. Akter, M. Begum, Rashed Mustafa","doi":"10.1109/ICICT4SD50815.2021.9396910","DOIUrl":"https://doi.org/10.1109/ICICT4SD50815.2021.9396910","url":null,"abstract":"The sentiment analysis of the Bengali language is converting into a trendy research topic nowadays. Sentiment analysis is a useful technique in opinion mining, emotion extraction, and trend predictions. By sentiment analysis, the actual sentiment of a text review can be extracted. Every day, every second's people use the internet for different purposes and leave their opinions or perspective views in various places on the internet as a text format. The opinion or review on the internet can contain the author's positive, negative, and neutral views of the statement. This study proposed a machine learning-based model to predict a user's sentiment (positive, neutral, and negative) of a Bangla text review. We have applied five machine learning algorithms in our dataset, which we manually gathered from a Bangladeshi e-commerce site called “Daraz.” We have experimented with Random Forest classifier, Logistic Regression, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and XGBoost algorithms with our dataset. KNN performs great among all these five algorithms in all the performance measures of accuracy, precision, recall, and f1-score. KNN shows 96.25% accuracy, 0.96 in each of precision, recall, and f1-score.","PeriodicalId":239251,"journal":{"name":"2021 International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD)","volume":"236 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134192501","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
M. Alam, Umma Hany, Iftekher Alam Rahat, Md. Musfiqur Rahman, Md. Isteak Parvez, Md. Jahid Hasan, Md. Saiam
{"title":"An Autonomic LASER Tag Combat Simulation System Based on IPS Using RSSI from Wireless Devices","authors":"M. Alam, Umma Hany, Iftekher Alam Rahat, Md. Musfiqur Rahman, Md. Isteak Parvez, Md. Jahid Hasan, Md. Saiam","doi":"10.1109/ICICT4SD50815.2021.9397050","DOIUrl":"https://doi.org/10.1109/ICICT4SD50815.2021.9397050","url":null,"abstract":"A combat simulation has been prepared in an effort to make the Bangladesh Army more up-to-date and efficient, with the aim of eliminating the use of real weapons in practice field and saving the cost of ammunition. It is possible to conduct combat simulation with low-cost laser tag without using conventional real guns or ammunition. The system has three major components. These are the laser blaster, battle station, and shot detection vest. Each team can split into two and do combat simulation with laser blaster. This simulation will take place in an indoor room which will be consider as battle station, where one person sitting can watch over the whole simulation from base station. In this combat simulation, a player will gain as many points as possible from the opponent player by shooting laser from the laser blaster, each player will be wearing a vest, which will detect the opponent's laser. Each player's vest will have his health bar, points and how many bullets left in that laser blaster. As this simulation will take place in an indoor room so it is important to calculate the position of each player, while it is too difficult to get the exact position of a player. So, for this it's needed to test two simulation and get the error rate approximately zero and get the actual position of a player.","PeriodicalId":239251,"journal":{"name":"2021 International Conference on Information and Communication Technology for Sustainable Development (ICICT4SD)","volume":"314 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134445598","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}