Browsing by Author "Salehin, Imrus"

Now showing 1 - 13 of 13

A Computer Vision System for the Categorization of Citrus Fruits Using Convolutional Neural Network
(2021 International Symposium on Electronics and Smart Devices (ISESD), IEEE, 2021-08-12) Hasan, Md. Mehedi; Salehin, Imrus; Moon, Nazmun Nessa; Kamruzzaman, T. M.; -Ul-Islam, Baki; Hasan, Mehedi
The recognition of citrus fruit is one of the most challenging and crucial measures in citrus yield mapping. Several artificial vision systems have been proposed to solve the issue of fruits recognition problem with sundry effects. In this study, we developed an automated system to categorize citrus fruit images using Convolutional Neural Network. We categorized two different citrus fruits, Orange (Citrus Sinensis) and Kinnow (Citrus Reticulate). Firstly the images of Orange and Kinnow were collected and preprocessed. Secondly, the fruit images and their background were segmented by image segmentation and edge detection. Four main features of Orange and Kinnow fruit were extracted based on image segmentations such as fruit size, surface color fruit shape and fruit surface defects. These features were examined through Convolutional Neural Network. We implemented three separate Convolutional Neural Network models to further experiment and tested recognition rates for different parameters. We have used the classical measurements including precision, recall, F1 score, ROC and accuracy for performance evaluation. Among the three experimented models, the third model was outperformed by 92.25% percent accuracy.
A Dynamic Study on Energy Forecasts and The Potential of Renewable Energy Sources
(Daffodil International University, 2022-05-21) Noman, S. M.; Salehin, Imrus; Hasan, Mohammad Mahedy; Islam, Baki-Ul-; Haque, Obaidul; Haque, Ifranul
With developing countries, energy demand has been rapid over the years. Bangladesh contributes just 5% of the entire energy ratio in a proportion of renewable energy, and desire to 10% of the year 2025. Bangladesh is based on imported fossil fuel, which is quite expensive and uses natural gas for about 65% of energy production. Bangladesh is fortunate to have a small amount of fossil fuel, i.e., natural gas assets, but these are not adequate to meet the desired robust growth for glorious achievements in the government and non-government sectors. The sustainable power source is a key segment for improvement and has just made significant track in succeeding the greater part of the nation's power request; both in Urban and Rural territories. The motivation behind this study is to analyse energy forecasting for upcoming days and the potential of alternative renewable energy sources to assimilate a comparison in the general diagram of the power utilization of Bangladesh from a global perspective. Anticipating the future demand and looking at the competent outcomes of renewable energy may make urgency among the developing countries to satisfy the high future need for clean energy. In this article, we have focused our attention on the examination of sustainable power sources and the potential outcomes for future energy.
A Smart Polluted Water Overload Drainage Detection and Alert System
(2021 International Mobile, Intelligent, and Ubiquitous Computing Conference (MIUCC),IEEE, 2021-06-09) Salehin, Imrus; Islam, Baki-Ul-; Noman, S. M.; Hasan, Md. Mehedi; Dip, Sadia Tamim; Hasan, Mehedi
A smart city constructed through the Internet of Things is a great and best medium nowadays. In our study, we have designed an advanced and automated device that can identify overloaded polluted drainage, which is responsible for water-borne disease and unexpected floods. We are using a smart ultrasonic sensor with an Ethernet shield integrated Arduino UNO. This proposed model's most vital side is the remote data access system using IP address and the webserver. This research is adequate for city corporations to advance their city, Develop their city to be more delighter, and reduce their fund used for extra human resources of city corporations cleaner. For data access methods, we are using the city Wi-Fi router and a monitoring station. We also proposed a separate web page designed to show the report. Materials used here are very uncomplicated and inexpensive but adequate to make an advanced automation intelligence system. In this modern scientific era, to lead a comfortable life, an automation system has helped develop a city more than the old system.
An Advanced Method of Treating Agricultural Crops Using Image Processing Algorithms and Image Data Processing Systems
(2020-10) Salehin, Imrus; Talha, Iftakhar Mohammad; Saifuzzaman, Mohd.; Moon, Nazmun Nessa; Nur, Fernaz Narin
Smart agriculture has involved evolution, judgment, and application of new methods of using modern technology. Technological advances in agriculture will enable farmers to enhance their skills in farming. We planned technology for farming by combining an app and a SMS system through the mobile phone. Different types of virus, fungus, and bacterial infection causes a great loss of farming product. Modern technologies in various computer science fields such as image processing, data mining can be applied in this infrastructure. We use the Scale-Invariant Feature Transform (SIFT) algorithm in this paper to identify crop diseases based on various types of datasets. The SFT technique is a well-known method that is applied to find the image data with pixel integrated. Firstly, we find out all key points and store all unique data from image for next steps. After processing every pixel, we match the main key point for major disease detection. In this study, our contribution is that we are trying to identify all diseases. We are trying to provide some solutions with the help of a solution bank using SMS services and live web portals.
An Artificial Intelligence Based Rainfall Prediction Using LSTM and Neural Network
(IEEE, 2020-12) Salehin, Imrus; Talha, Iftakhar Mohammad; Hasan, Md. Mehedi; Dip, Sadia Tamim; Saifuzzaman, Mohd.; Moon, Nazmun Nessa
The most difficult task of meteorology is to predict rainfall. In our study, we proposed an amount of rainfall prediction model that can be easily determined using artificial intelligence and LSTM techniques. This is an advanced method to find out the rainfall. The deep learning approach is most valuable for this type of method implementation and its accuracy finds out. A long short-term memory algorithm is applied to memory sequence data measurement and calculate previous data very fast and create the best prediction. The people of this country are mostly dependent on agriculture so that this prediction system is very necessary. Timely rainfall assessment will increase crop yields and reduce costs in agriculture. Considering all these factors, we have created our model which will help us to determine the amount of rainfall. We have collected data from 6 regions to do this. To predict, we have taken 6 parameters (temperature, dew point, humidity, wind pressure, wind speed, and wind direction). After analyzing all our data, we got 76% accuracy in our work. We also focus on a vast dataset in long time weather for the better result.
Analysis of Student Sentiment During Video Class with Multilayer Deep Learning Approach
(Daffodil International University, 2022-08-08) Salehin, Imrus; Moon, Nazmun Nessa; Talha, Iftakhar Mohammad; Hasan, Md. Mehedi; Nur, Farnaz Narin; Hakim, Md. Azizul; Haque, Farhan Al
The modern education system is an essential part of the rise of technology. The E-learning education system is not just an experimental system; it is a vital learning system for the whole world over the last few months. In our research, we have developed our learning method in a more effective and modern way for students and teachers. For significant implementation, we are implementing convolutions neural networks and advanced data classifiers. The expression and mood analysis of a student during the onlineclass is the main focus of our study. For output measure, we divide the final output result as attentive, inattentive, understand, and neutral. Showing the output in real-time online class and for sensory analysis, we have used support vector machine(SVM)and OpenCV. The level of 5*4 neural network is created for this work. An advanced learning medium is proposed through our study. Teachers can monitor the live class and different feelings of a student during the class period through this system.
Electricity energy dataset “BanE-16”: Analysis of peak energy demand with environmental variables for machine learning forecasting
(Scopus, 2024) Salehin, Imrus; Noman, S.M.; Hasan, Mohammad Mahedy
The “BanE-16” dataset is a comprehensive repository integrating electricity grid dynamics with meteorological variables for machine learning-based energy forecasting. Featuring peak energy demand, environmental factors (temperature, wind speed, atmospheric pressure), and electricity generation statistics, this dataset enables intricate analysis of weather-energy correlations. Its multidimensional nature facilitates predictive modeling, exploring intricate dependencies, and optimizing energy infrastructure. Leveraging machine learning methodologies, this dataset stands as a catalyst for innovative forecasting models and informed decision-making in energy management. Its diverse variables offer a holistic perspective, empowering researchers to delve into nuanced interrelationships, paving the way for sustainable energy planning and predictive analytics in dynamic energy ecosystems. Its multivariate nature empowers sophisticated machine-learning models, enabling precise energy forecasts and infrastructure optimizations. Researchers leveraging this dataset unlock the potential to delve deeper into intricate weather-energy relationships, driving advancements in predictive analytics for sustainable energy management. The integration of diverse variables lays the groundwork for innovative methodologies, steering the trajectory of informed decision-making in dynamic energy landscapes.
Human Behaviour Impact to Use of Smartphones with the Python Implementation Using Naive Bayesian
(11th International Conference on Computing, Communication and Networking Technologies, ICCCNT 2020, IEEE, 2020-10-15) Talha, Iftakhar Mohammad; Salehin, Imrus; Debnath, Susanta Chandra; Saifuzzaman, Mohd.; Moon, Nazmun Nessa; Nur, Fernaz Narin
A change of behavior in special groups and many sustainable smart populations increasing day by day for excessive uses of smartphones. In recent years, the use of smartphones and mental imbalances have become a major problem with increasing negative effects. In our study, we find out the major problem of the negative side and its different sources like mental imbalance, stress, depression, loneliness, etc. Bayes' theorem and classifier, support vector machine, special data set of human behavior, and probability are used to calculate accuracy. For collecting data from three major sections, we use the physical methods, virtual methods, and medical reports. So, a vast data set is trained by data to compare method, and also probability is used for predicting the validity of the data model. Naive Bayes' theorem accurate 71% positive which is indicated the negative impact of human behavior. Based on the SVM classifier, we separate the barrier between the impact of positive and negative data. In SVM, we set up a parameter to measure negative and positive values. Python library function is a major component to calculate all instructions and also use for data training. Finally, we compare the results obtained by our proposed specialization with the results obtained from the three baseline landmarks.
IFSG
(Indonesian Journal of Electrical Engineering and Computer Science, 2021) Salehin, Imrus; Noman, S. M.; Ul-Islam, Baki; Lopa, Israt Jahan; Angon, Prodipto Bishnu; Habiba, Ummya; Moon, Nazmun Nessa
The agricultural and technological combination is blessed for modern world life. Internet of things (IoT) is essential for comfort and development to our agriculture side. In our study, we detected the various pest using different types of sensors and this information has automatically sent to the farmer's mobile for the alert. All these sensors had a central database. Those sensors collect all the data and display the results compared to the central data. The High-image sensor will be able to detect all the rays emitted from the plant and another one is the gas sensor which is able to detect all the gases coming from the diseased plant. We mainly use sound sensor, MQ138, CMOSOV-7670, AMG-8833 for a better automation system. We test it with real-time environment conditions (40°C≤TA≤14°C). Crop pest detection automatic process is more efficient than the other detection process according to testing output. As a result, far-reaching changes in the agricultural sector are possible. To reduce extra cost and increasing more farming ability we need to IoT and Agriculture combinations more.
Natural Language Processing Based Advanced Method of Unnecessary Video Detection
(International Journal of Electrical and Computer Engineering, 2021) Moon, Nazmun Nessa; Salehin, Imrus; Parvin, Masuma; Hasan, Md. Mehedi; Talha, Iftakhar Mohammad; Debnath, Susanta Chandra; Nur, Fernaz Narin; Saifuzzaman, Mohd.
In this study we have described the process of identifying unnecessary video using an advanced combined method of natural language processing and machine learning. The system also includes a framework that contains analytics databases and which helps to find statistical accuracy and can detect, accept or reject unnecessary and unethical video content. In our video detection system, we extract text data from video content in two steps, first from video to MPEG-1 audio layer 3 (MP3) and then from MP3 to WAV format. We have used the text part of natural language processing to analyze and prepare the data set. We use both Naive Bayes and logistic regression classification algorithms in this detection system to determine the best accuracy for our system. In our research, our video MP4 data has converted to plain text data using the python advance library function. This brief study discusses the identification of unauthorized, unsocial, unnecessary, unfinished, and malicious videos when using oral video record data. By analyzing our data sets through this advanced model, we can decide which videos should be accepted or rejected for the further actions.
Predicting the Depression Level of Excessive Use of Mobile Phone
(Daffodil International University, 2021-05-31) Salehin, Imrus; Talha, Iftakhar Mohammad
In this research titled “Predicting the Depression Level of Excessive Use of Mobile Phone: Using Machine Learning Algorithm” which is applied advanced machine learning and regression analysis to find out the depression level. We have done the whole work in the research area of medical science and information technology and also built up a collaboration. In this study, we are focusing on the strength of the algorithm and also calculate the accuracy with python programming. The result expresses that smart mobile device changing the human brain day by day if spend more time around 8 to 12 hours a day. At last, we observed that a man or woman slowly going through a depression for the impact of the excessive mobile operates. In our study, we have used multiple classification algorithms to find out depression level such as Probability, Decision Tree, Random Forest, Linear Regression and SVM (Support Vector Machine). For the accuracy of our work, we have used four types of algorithms to find the optimal ratio and percentage.
Real-Time Medical Image Classification with ML Framework and Dedicated CNN–LSTM Architecture
(Hindawi Publications, 2022) Salehin, Imrus; Islam, Md. Shamiul; Amin, Nazrul; Baten, Md. Abu; Noman, S. M.; Saifuzzaman, Mohd; Yazmyradov, Serdar
In the domain of modern deep learning and classification techniques, the convolutional neural network (CNN) stands out as a highly successful and preferred method for image classification in artificial intelligence. Especially in the medical field, CNN has proven to be an ideal approach for analyzing medical data and accurately identifying diseases. Over the recent years, CNN has demonstrated significant potential and success in various computer vision tasks, with medical image classification being one of the prominent applications. In our study, we introduce a novel custom CNN model called MedvCNN, designed for classifying different types of classes. We conduct experiments with various image sizes to explore their versatility. In addition, long short-term memory (LSTM), a type of recurrent neural network (RNN), is incorporated into our approach. LSTM is specifically tailored to handle sequential data, making it ideal for time series analysis. However, its capabilities extend beyond time series data and are effectively applied to various sequential data types, including sequential vectors derived from image data. One of the key advantages of utilizing LSTM for image classification is its ability to effectively memorize and capture important features in the image data. This feature is particularly advantageous in medical image processing, where precise and accurate identification of key attributes is crucial for successful diagnosis and analysis. Furthermore, our experiments reveal that the hybrid custom LSTM model, MedvLSTM, a RNN algorithm, surpasses other methods in the domain of medical image classification. Our study places significant emphasis on attaining robust classification performance for medical image data through a sophisticated, parameter free approach, complemented by an ablation study, and comprehensive statistical analysis. This comprehensive analysis and evaluation allow us to gain a deeper understanding of the model’s effectiveness and its potential impact in the field of medical image analysis. We compare these two approaches to a baseline CNN architecture, aiming to streamline the classification process, reduce time consumption, and improve cost efficiency. Additionally, we present a real-time web-based AutoML framework along with a practical demonstration. Ultimately, our research provides a thorough investigation of the current state-of-the-art in medical image analysis accuracy, focusing on the utilization of neural networks and LSTM.
Tidal Power Plant Exploration
(Daffodil International University, 2022-12-29) Noman, S. M.; Hasan, Mohammad Mahedy; Haque, Obaidul; Islam, Baki-Ul-; Salehin, Imrus; Moon, Nazmun Nessa
Tidal energy is an environmentally benign, practical, and cost-effective renewable energy source. The hydropower potential appropriate site for electrical power production accessible in Bangladesh's coastline region is investigated in this article. We propose a model for determining the feasibility of power production via the tidal barrage system using a Compact Axial Kaplan (CAK) turbine in a low-head water turbine. Tidal power supplies enormous quantities of electrical energy in coastal locations across the globe without generating carbon or harming the environment. The world's wealthiest nations are increasingly resorting to tidal power generation. Where tidal power is available, we may be used tidal energy to alleviate the power shortage in coastal areas. A hydropower station may supply clean energy and fulfill a significant amount of the country's electrical demands. The notion of establishing a tidal power plant with a barrage model system would be a better long-term outlook for energy production and flourishing economic expansion. Dublar Char is an ideal location for tidal energy to be used to solve the electricity crisis in distant coastal regions. This article analyzes the electricity generating capability of Dublar Char, estimated at 120.17 megawatts (MW).