Browsing by Author "Islam, Samiul"

Now showing 1 - 20 of 23

A Framework for Liquefaction Susceptibility Mapping of Dhaka City Using Latest SPT Based Co-relationship and Regional Factors
(Department of Civil and Environmental Engineering(CEE), Islamic University of Technology(IUT), Board Bazar, Gazipur-1704, Bangladesh, 2023-09-30) Islam, Samiul
Integration of regional code-based provisions considering local site effects and updated analytical techniques is a must in the field of seismic site characterization and liquefaction susceptibility assessment. This study introduces provisions from latest regional guidelines to define seismic site class and assess liquefaction susceptibility, utilizing the most comprehensive database developed for Dhaka City and produces vulnerability maps, categorizing regions into various zones based on potential risk factors. Utilizing the latest SPT based co-relationship, coupled with code-based stress reduction factors, liquefaction susceptibility assessment was conducted. Additionally, classification-based supervised machine learning algorithms were utilized to evaluate the performance of the liquefaction susceptibility calculations, which were subsequently used as an input parameter for conducting geo-statistical interpolation, resulting in risk-based zonation maps in terms of liquefaction hazard for the city. The results show that the deposition type of soil plays a significant role in triggering liquefaction in different areas of Dhaka City and the majority portion of the recent artificial fill areas are subjected to high liquefaction potential for 7.0, 7.5 and 8.0 magnitude earthquake. This study also supplements the newly published mandates and provides guidelines according to the code to conduct engineering studies as per recommended seismic site class and liquefaction susceptibility for design applications. A significant increase in the coverage area of seismic site class with low shear wave velocities have been observed, necessitating special infrastructural considerations as per the new codal guidelines compared to past researches. Areas of improvement to evaluate liquefaction susceptibility in the newly published mandate also have been identified. Furthermore, this study also outlines a generalized framework with supplementary policies integrating regional factors into consideration for development of liquefaction susceptibility-based risk maps for any location. The developed liquefaction susceptibility-based zonation map provides a clear visual representation of areas prone to liquefaction, enabling better-informed decision- making for disaster preparedness, risk reduction, and sustainable urban development.
A Guava Leaf Disease Detection by Machine Learning
(Daffodil International University, 2021-06-01) Haque, Md. Radoanul; Islam, Samiul; Mamata, Nishat Anjum
Fruit diagnosis and early identification The production of healthy fruit industry is more critical for plant diseases. Farmers' general monitoring system can take time, costly and often incorrect. This paper offers an overview of target recognition through grouping of numerous images and machine learning methods for the guava leaf disease detection. Our system has been developed based on machine learning algorithm. In this work rust, white fly, leaf spot and sound disease has been detected. For the whole method, a number of machine learning programs (MLs) were used, such as Scikit-learn, Pandas, Matploatlib, Numpy. In the pre-processing of images, we have also used Scikit-learn to implement algorithms. In order to check the validity of our work we use five separate K-Nearest Neighbour(KNN), Vector Support (SVM), Tree Classifier Decisions and the Random Forest. Naive Bayes. The most effective algorithm. This five algorithms were studied. Finally, this high-precision algorithm detects guava leaf disease.
Abnormal behavior detection of human by video surveillance system
(BRAC University, 2014-12) Amin, Anisul; Anzum, Mohammad Farhan; Mondol, Mark Himel; Alom, Md. Zahangir; Islam, Samiul
In recent years, the number of surveillance cameras installed to monitor private and public spaces and areas has increased dramatically. There is an increasing demand for smarter video surveillance of public and private space using intelligent vision systems which can distinguish what is semantically meaningful to the human observer as „normal‟ and „abnormal‟ behaviors. Usually, the video streams are constantly recorded or observed by operators. In these cases an intelligent system can give more accurate performance than a human. In this thesis we present a video surveillance system that detects and predicts abnormal behavior of human. The system acquires color images from a stationary camera and analyzes the behavior of human. Behaviors that are common or frequent will not be given much attention by the system.
Automatic detection of defective rail anchors
(2014-11) Khan, Rubayat Ahmed; Islam, Samiul; Biswas, Rubel
Rail line anchors/fasteners are the metallic components that attach each line with the sleepers. These are essential rail components as absence of these often result in derailments. Therefore in order to prevent dangerous situations and ensuring safety rail lines are periodically inspected. Rail inspection in many countries especially in third world countries, like Bangladesh, is performed manually by a trained human operator who periodically walks along the track searching for visual anomalies. This manual inspection is lengthy, laborious and subjective. This paper presents a machine vision-based technique to automatically detect the presence of rail line anchors/fasteners using Shi - Tomasi and Harris - Stephen feature detection algorithms. This approach has confirmed to successfully detect scenarios with both grounded and missing anchors invoked in the experiment, with an accuracy of 83.55%, thus proving its robustness.
Automatic detection of defective rail anchors
(2014-11) Khan, Rubayat Ahmed; Islam, Samiul; Biswas, Rubel
Rail line anchors/fasteners are the metallic components that attach each line with the sleepers. These are essential rail components as absence of these often result in derailments. Therefore in order to prevent dangerous situations and ensuring safety rail lines are periodically inspected. Rail inspection in many countries especially in third world countries, like Bangladesh, is performed manually by a trained human operator who periodically walks along the track searching for visual anomalies. This manual inspection is lengthy, laborious and subjective. This paper presents a machine vision-based technique to automatically detect the presence of rail line anchors/fasteners using Shi - Tomasi and Harris - Stephen feature detection algorithms. This approach has confirmed to successfully detect scenarios with both grounded and missing anchors invoked in the experiment, with an accuracy of 83.55%, thus proving its robustness.
Automatic detection of defective rail anchors
(2014-11) Khan, Rubayat Ahmed; Islam, Samiul; Biswas, Rubel
Rail line anchors/fasteners are the metallic components that attach each line with the sleepers. These are essential rail components as absence of these often result in derailments. Therefore in order to prevent dangerous situations and ensuring safety rail lines are periodically inspected. Rail inspection in many countries especially in third world countries, like Bangladesh, is performed manually by a trained human operator who periodically walks along the track searching for visual anomalies. This manual inspection is lengthy, laborious and subjective. This paper presents a machine vision-based technique to automatically detect the presence of rail line anchors/fasteners using Shi - Tomasi and Harris - Stephen feature detection algorithms. This approach has confirmed to successfully detect scenarios with both grounded and missing anchors invoked in the experiment, with an accuracy of 83.55%, thus proving its robustness.
Automatic detection of defective rail anchors
(2014-11) Khan, Rubayat Ahmed; Islam, Samiul; Biswas, Rubel
Rail line anchors/fasteners are the metallic components that attach each line with the sleepers. These are essential rail components as absence of these often result in derailments. Therefore in order to prevent dangerous situations and ensuring safety rail lines are periodically inspected. Rail inspection in many countries especially in third world countries, like Bangladesh, is performed manually by a trained human operator who periodically walks along the track searching for visual anomalies. This manual inspection is lengthy, laborious and subjective. This paper presents a machine vision-based technique to automatically detect the presence of rail line anchors/fasteners using Shi - Tomasi and Harris - Stephen feature detection algorithms. This approach has confirmed to successfully detect scenarios with both grounded and missing anchors invoked in the experiment, with an accuracy of 83.55%, thus proving its robustness.
Automatic measurement of rail line expansion joint gaps
(© 2014 Institute of Electrical and Electronics Engineers Inc., 2015-03) Islam, Samiul; Khan, Rubayat Ahmed; Biswas, Rubel
Expansion joint gaps are the gaps which are deliberately left between the rail ends to allow for expansion of the rails in hot weather. Over gapping of these end to end gaps often result in derailments. Therefore in order to prevent dangerous situations and ensuring safety rail lines are periodically inspected. Rail inspection in many countries especially in third world countries, like Bangladesh, is performed manually by a trained human operator who periodically walks along the track searching for visual anomalies. Such manual inspection is lengthy, laborious and subjective. This paper presents a machine vision-based technique to automatically measure the length of rail line expansion joint gaps using morphological processing. This approach has confirmed to successfully detect scenarios of different condition with an accuracy of 89%, thus proving its robustness.
Automatic measurement of rail line expansion joint gaps
(© 2014 Institute of Electrical and Electronics Engineers Inc., 2015-03) Islam, Samiul; Khan, Rubayat Ahmed; Biswas, Rubel
Expansion joint gaps are the gaps which are deliberately left between the rail ends to allow for expansion of the rails in hot weather. Over gapping of these end to end gaps often result in derailments. Therefore in order to prevent dangerous situations and ensuring safety rail lines are periodically inspected. Rail inspection in many countries especially in third world countries, like Bangladesh, is performed manually by a trained human operator who periodically walks along the track searching for visual anomalies. Such manual inspection is lengthy, laborious and subjective. This paper presents a machine vision-based technique to automatically measure the length of rail line expansion joint gaps using morphological processing. This approach has confirmed to successfully detect scenarios of different condition with an accuracy of 89%, thus proving its robustness.
Automatic measurement of rail line expansion joint gaps
(© 2014 Institute of Electrical and Electronics Engineers Inc., 2015-03) Islam, Samiul; Khan, Rubayat Ahmed; Biswas, Rubel
Expansion joint gaps are the gaps which are deliberately left between the rail ends to allow for expansion of the rails in hot weather. Over gapping of these end to end gaps often result in derailments. Therefore in order to prevent dangerous situations and ensuring safety rail lines are periodically inspected. Rail inspection in many countries especially in third world countries, like Bangladesh, is performed manually by a trained human operator who periodically walks along the track searching for visual anomalies. Such manual inspection is lengthy, laborious and subjective. This paper presents a machine vision-based technique to automatically measure the length of rail line expansion joint gaps using morphological processing. This approach has confirmed to successfully detect scenarios of different condition with an accuracy of 89%, thus proving its robustness.
Automatic measurement of rail line expansion joint gaps
(© 2014 Institute of Electrical and Electronics Engineers Inc., 2015-03) Islam, Samiul; Khan, Rubayat Ahmed; Biswas, Rubel
Expansion joint gaps are the gaps which are deliberately left between the rail ends to allow for expansion of the rails in hot weather. Over gapping of these end to end gaps often result in derailments. Therefore in order to prevent dangerous situations and ensuring safety rail lines are periodically inspected. Rail inspection in many countries especially in third world countries, like Bangladesh, is performed manually by a trained human operator who periodically walks along the track searching for visual anomalies. Such manual inspection is lengthy, laborious and subjective. This paper presents a machine vision-based technique to automatically measure the length of rail line expansion joint gaps using morphological processing. This approach has confirmed to successfully detect scenarios of different condition with an accuracy of 89%, thus proving its robustness.
Capturing Spectral and Long-term Contextual Information for Speech Emotion Recognition Using Deep Learning Techniques
(Department of Computer Science and Engineering(CSE), Islamic University of Technology(IUT), Board Bazar, Gazipur-1704, Bangladesh, 2023-05-30) Haque, Md. Maksudul; Islam, Samiul; Sadat, Abu Jobayer Md.
Traditional approaches in speech emotion recognition, such as LSTM, CNN, RNN, SVM, and MLP, have limitations such as difficulty capturing long-term dependen cies in sequential data, capturing the temporal dynamics, and struggling to capture complex patterns and relationships in multimodal data. This research addresses these shortcomings by proposing an ensemble model that combines Graph Con volutional Networks (GCN) for processing textual data and the HuBERT trans former for analyzing audio signals. We found that GCNs excel at capturing Long term contextual dependencies and relationships within textual data by leveraging graph-based representations of text and thus detecting the contextual meaning and semantic relationships between words. On the other hand, HuBERT utilizes self-attention mechanisms to capture long-range dependencies, enabling the mod eling of temporal dynamics present in speech and capturing subtle nuances and variations that contribute to emotion recognition. By combining GCN and Hu BERT, our ensemble model can leverage the strengths of both approaches. This allows for the simultaneous analysis of multimodal data, and the fusion of these modalities enables the extraction of complementary information, enhancing the discriminative power of the emotion recognition system. The results indicate that the combined model can overcome the limitations of traditional methods, leading to enhanced accuracy in recognizing emotions from speech.
Clustering and detection of good and bad rail line anchors from images
(© 2015 Institute of Electrical and Electronics Engineers Inc., 2016-06) Islam, Samiul; Khan, Rubayat Ahmed
Absence of railway anchors/fasteners is a serious concern as it might lead to severe consequences such as train derailments. Hence regular inspection is an obligation to ensure safety. The third world countries choose the inspection process to be non-automatic where a trained operator moves along the rail line boarding a motor trolley checking for visual anomalies. In the previous research [1], an automatic system was proposed to overcome the cons of the running manual technique by using image processing. Two feature detection algorithms - Shi Tomasi and Harris Stephen - were used and an accuracy of 83.55% was achieved. This research presents an upgraded version of the previous work by introducing Neural Network. The addition of NN has not only speeded up the detection process but increased the accuracy significantly to approximately 93.86%.
Clustering and detection of good and bad rail line anchors from images
(© 2015 Institute of Electrical and Electronics Engineers Inc., 2016-06) Islam, Samiul; Khan, Rubayat Ahmed
Absence of railway anchors/fasteners is a serious concern as it might lead to severe consequences such as train derailments. Hence regular inspection is an obligation to ensure safety. The third world countries choose the inspection process to be non-automatic where a trained operator moves along the rail line boarding a motor trolley checking for visual anomalies. In the previous research [1], an automatic system was proposed to overcome the cons of the running manual technique by using image processing. Two feature detection algorithms - Shi Tomasi and Harris Stephen - were used and an accuracy of 83.55% was achieved. This research presents an upgraded version of the previous work by introducing Neural Network. The addition of NN has not only speeded up the detection process but increased the accuracy significantly to approximately 93.86%.
Clustering and detection of good and bad rail line anchors from images
(© 2015 Institute of Electrical and Electronics Engineers Inc., 2016-06) Islam, Samiul; Khan, Rubayat Ahmed
Absence of railway anchors/fasteners is a serious concern as it might lead to severe consequences such as train derailments. Hence regular inspection is an obligation to ensure safety. The third world countries choose the inspection process to be non-automatic where a trained operator moves along the rail line boarding a motor trolley checking for visual anomalies. In the previous research [1], an automatic system was proposed to overcome the cons of the running manual technique by using image processing. Two feature detection algorithms - Shi Tomasi and Harris Stephen - were used and an accuracy of 83.55% was achieved. This research presents an upgraded version of the previous work by introducing Neural Network. The addition of NN has not only speeded up the detection process but increased the accuracy significantly to approximately 93.86%.
Cross-cultural comparison of food appetite and restaurant features
(BRAC University, 2019-04) Islam, Md Sami Ul; Abib, Md. Shariful Huq; Islam, Samiul
Food business has expanded drastically over the years. It’s safe to say that you never have to make a loss for a restaurant business today onwards. People have become food lovers and they are always in search of new dishes and different tastes. From business meetings to engagement proposals to school assignments- all are happening over a cup of coffee or during lunch or dinner. So, it’s worth a try to understand this food culture globally. Because food represents the culture. Taste and method represent the history of a country. Price-variation tells us about the overall economy. Watching these scenarios nowadays, we got encouraged to study on people around the world to find out where they are on the same page in terms of food consumption. Which types of cuisines got popular worldwide and what are the reasons. Are the prices and environments similar to the same type of cuisines everywhere? We have adequate data of great variation to analyze these factors. We will take the help of statistic formulas and several machine learning techniques to accomplish our project.
Design and development of doctor’s dictation kit using raspberry PI
(BRAC University, 2016) Haque, Kazi Injamamul; Saha, Ullash; Biswas, Sudipto; Billah, Md. Muhtasim; Momin, Abu Saleh Al; Rahman, Mohammad Zahidur; Islam, Samiul
Natural language processing and speech to text can make a significant improve in medical dictation (transcription, radiology report, prescription etc) in a developing country like Bangladesh. In the field of telemedicine it can play a very crucial part in the absence of qualified doctors and specialists to prescribe medicine and provide with medical support in remote and rural places. This paper is based on a real time speech detection with a standalone system to implement it in a single board computer Raspberry PI that can also work in crowded place. The recognition engine used for the system is JULIUS along with the toolkit HTK to manipulate HMM(Hidden Markov Model). The acoustic model is set to such a way that it can detect selected medicine names those are widely used in Bangladesh. The accuracy rate of our trained dictionary is 84% but a silent environment and longer string prodeces 94% accuraccy which can also be imroved with more accurate training with advanced directional microphone. The intention of implementing the system in Raspberry PI was to have a future innovation of a standalone device for medical dictation and telepharmacy.
Molecular insights into withaferin A: A holistic approach to cancer therapy
(2024-12-15) Rezaul Islam, Md.; Abdur Rauf; Rakesh, M. Meenakshi; Akash, Shopnil; Fakir, M. Naeem Hossain; Islam, Samiul; Naba, Afifa Farzana; Al-Imran, Md. Ibrahim Khalil; AlOmar, Taghrid S.; Ogaly, Hanan A.; Alzahrani, Hayat E.; Thiruvengadam, Rekha; Thiruvengadam, Muthu
Withaferin A (WA), a significant phytoconstituent of Withania somnifera, belongs to the triterpenoid class of C28-steroidal lactones found in nature. It has been utilized in traditional and native medical systems to treat various diseases such as cancers. It also has pro-apoptotic, anti-inflammatory, metabolic, and anticancer properties. In addition, it interacts with NF-κB, STAT, Hsp90, ER, and p53, thereby suppressing cancer cell growth and reducing the cell cycle at the G2/M stage. This review discusses the pro-apoptotic properties of WA, such as the generation of ROS, activation of PAR-4, development of ER stress, and activation of p53. This provides a comprehensive analysis of the molecular mechanisms of WA, highlighting its intricate effects on cancer cell signaling pathways, apoptosis, and tumor microenvironment (TME) regulation. Moreover, the ability to modify the TME through immune response modulation underscores its importance in comprehensive cancer treatment. The involvement of WA in oncogenic pathways that lead to malignant neoplasms and its potential therapeutic effects when combined with other cancer treatments are promising. This review indicates that WA's strong pharmacological profile, especially in combating cancer, could be advantageous for developing new cancer therapy medications. Furthermore, it emphasizes the potential of WA as a supplement to standard medications and encourages further clinical trials to assess its efficacy and safety in various cancer types. Moreover, it indicates a growing understanding of natural chemicals in cancer therapy, which can be integrated into various treatment methods.
Railway expansion joint gaps and hooks detection using morphological processing, corner points and blobs
(BRAC University, 2014-04) Islam, Samiul; Biswas, Rubel; Alam, Jahangir
Rail inspection is an essential task in railway maintenance. It is periodically needed for preventing dangerous situations and ensuring safety in railways. In Bangladesh it has been seen many train accidents occur due to over gapping between rail lines and also due to missing of hooks which attach the tracks to the ground. At present, this task is operated manually by a trained human operator who periodically walks along the track searching for visual anomalies. This manual inspection is lengthy, laborious and subjective. This thesis presents a machine vision-based technique to automatically detect the presence of rail line hook and measure the gaps between each line to check whether the gap is safe or not. This inspection system uses real images acquired by a digital line scan camera installed under an automatic vehicle. Data are processed according to a combination of image processing and pattern recognition methods to achieve high performance automated detection. The scope of this project is strictly limited to the development of a machine vision based program capable of detecting the presence of parts of interest in rail tracks, from given rail track images.
Scrutiny of electricity data for consumption and load forecasting
(BRAC University, 2017) Islam, Samiul; Zaber, Dr. Moinul Islam
In this research, electricity data of Dhaka city, the capital city of Bangladesh has been analysed to use the insights for social good and betterment of electricity sectors. Bangladesh has a very complex electricity infrastructure for both generation and supply sector. According to Power system master plan, Bangladesh mainly produces electricity from gas mine and supply to the grid line. From the gridline, electricity supplies to households. Under the jurisdiction of the Ministry of Power, Energy and Mineral Resources (MPEMR), the Power Division (PD) oversees the whole electricity utility. There are two parts in Dhaka in the historical evolution: old Dhaka and new Dhaka. The responsible department for supplying electricity in these areas are DESCO and DPDC. DESCO is mostly responsible for new Dhaka and extended urban area of Dhaka. The data we have collected from DESCO consist of billing (monthly consumption) data, supply (hourly load) data and load shedding data. Monthly consumption data spans from 1995 to till date, hourly load data and load shedding data span from 2015 to 2016. The objectives of this research can be classified into two parts: one is to analyse monthly consumption/billing data and propose a consumption forecasting model which will predict the consumption in user level. Second, analyzing load/supply data (along with load-shedding data) to understand how legacy method works, addressing key points to ensure a better forecasting, how forecasting will help in future, a brief study of recent forecasting techniques, load shedding scenario, area specific impacts and proposing a forecasting technique which ensures granularity and relatively higher accuracy. It has been found that electricity consumption varies a lot for different tariff bracket consumers in a zone. Consumers from same tariff bracket act differently in different zones. Moreover, electricity usage is strongly correlated with temperature, seasonal change and an occasional change. If temperature increases, electricity usage also increases, usage changes a lot due to particular events. So, forecasting the demand (consumer and substation level) is a crucial part. A proper flow of information is a must from consumer level to the generation plants to predict future demand with minimal error