Browsing by Author "Rahman, W."

Now showing 1 - 9 of 9

Automated Detection of Harmful Insects in Agriculture: A Smart Framework Leveraging IoT, Machine Learning, and Blockchain
(Institute of Electrical and Electronics Engineers Inc., 2024-05-08) Rahman, W.; Hossain, M.M.,; Iqbal, M.S.; Rahman, M.M.; Fida Hasan, K.; Moni, M.A.
Paddy cultivation is a significant global economic sector, with rice production playing a crucial role in influencing worldwide economies. However, insects in paddy farms predominantly impact the growth rate and ecological equilibrium of the agricultural field. Hence, the precise and timely identification of insects in agricultural settings presents a potential strategy for addressing this issue. This study aims to implement an automated system for paddy farming by employing a real-time framework that incorporates the Internet of Things (IoT), blockchain technology, and Deep Learning (DL) algorithms. The primary emphasis of the DL-based system is on the timely identification of pests. In contrast, integrating the IoT and blockchain technologies facilitates stablishing a fully automated system with security within the agricultural domain. The DL-based system includes a secondary dataset of paddy insects, and then preprocessing, feature extraction, and identification have been performed. Besides, an IoT-based system is embodied with a camera module and microprocessor, accompanied by some apparatus required to automate the whole system. In addition, the research also includes the blockchain to secure each individual data transmission among the several IoT components and the cloud server. While examining the proposed solution, various experimental data have been systematically documented and analyzed. The proposed framework attained a peak accuracy of 98.91% using the VGG19 model and ensemble classifiers to detect the pest with a specificity of 99.14% and a precision of 98.21%. The study additionally quantifies the mean duration of the cloud response when integrated with IoT, yielding an average time of 1.71 s after pest identification. Nevertheless, the system has exhibited a high level of efficacy in the context of real-time monitoring and automation of paddy farms.
BDWaste: A comprehensive image dataset of digestible and indigestible waste in Bangladesh
(Elsevier Inc., 2024-04-24) Rahman, W.; Akter, M.; Sultana, N.; Farjana, M.; Uddin, A.; Mazrur, M.B.,; Rahman, M.M.
The “BDWaste” dataset contains two significant categories of waste, namely digestible and indigestible, in Bangladesh. Each category represents 10 distinct species of waste. The digestible categories are sugarcane husk, fish ash, potato peel, paper, mango peel, rice, shell of malta, lemon peel, banana peel, and egg shell. On the other hand, the indigestible species are polythene, cans, plastic, glass, wire, gloves, empty medicine packets, chip packets, bottles, and masks. The research uploaded the primarily collected dataset on Mendeley, and the dataset contains a total of 2497 raw images, of which 1234 were digestible and 1263 belonged to indigestible species. Each species is stored in a fixed file based on its name and categories. Also, each species contains an indoor (with a visible surface) and an outdoor (with a surface that can be seen generally) image. The dataset is free from any blurry, dark, noisy, or invisible images. The research also performed waste classification with pre-trained convolutional neural network models such as MobileNetV2 and InceptionV3. The research found the highest accuracy of 96.70% in the indigestible waste classification and 99.70% in the digestible waste classification. The researchers presume that this data can be used in the future in different types of research, such as sustainable development, sustainable environments, agricultural development, recycling processes, and other computer vision-based applications
COVID-19 detection from chest CT images using optimized deep features and ensemble classification
(Academic Press, 2024-02-04) Hossain, M.M.; Walid, M.A.A.; Walid, M.A.A.; Galib, S.M.S.; Azad, M.M.; Rahman, W.; Shafi, A.S.M.; Rahman, M.M.
Diagnosis of COVID-19 positive patients is the eventual move to impede the expansion of coronavirus. Variations of coronavirus make it tough to recognize COVID-19 positive patients through symptoms. Hence, this research aims at a faster and automatic detection approach of COVID-19 disease from the chest Computed tomography (CT) scan images. For the composition of the system, this approach constructs a feature vector from the CT images through the features fusion of two Convolutional neural network (CNN) models namely VGG-19 and ResNet-50. Before the feature fusion, preprocessing techniques are applied to gain more accurate outcomes. Moreover, pertinent features are identified from the feature vector by using several feature optimization methods namely Recursive feature elimination (RFE), Principal component analysis (PCA), and Linear discriminant analysis (LDA), and among them, we have observed PCA as the best preference. Classification is performed on the optimized feature utilizing the Max voting ensemble classification (MVEC). The fused features of VGG-19 and ResNet-50, processed with PCA and MVEC, provide the best outcomes of accuracy, specificity, sensitivity, and precision at 98.51 %, 97.58 %, 99.49 %, and 97.47 %, respectively, after 5-fold cross-validation for the proposed method.
Electrolyte composition of diarrhoeal stool in children-some thoughts on proper replacement therapy[abstract]
(1968-08) Rahman, M.M.; Rahman, W.
Evaluating the Performance of State-of-the-art Methods and Classifying Covid-19 Infected Tissues
(Institute of Electrical and Electronics Engineers Inc., 2022) Kamruzzaman, M.M.; Moinuddin, M.,; Liton, A.I.,; Azad, M.M.,; Hossain, M.A.,; Rahman, W.
In this study, the Traditional Convolution Neural Network (TCNN) and state-of-the-art approaches were applied to the datasets of Chest X-ray and CT scan imaging modalities and trained them concurrently. The TCNN's performance for detecting COVID-19 infected tissues was determined through a comparison examination using state-of-the-art approaches. The accuracy of the models has been improved by lowering the model's losses and overfitting. Finally, the training data size has been enhanced utilizing various picture augmentation methods such as flip-up-down, flip-down-left-right, and so on. VGG19 and InceptionV3 were tested in this work, and accuracy scores of 97 percent (X-ray images) and 96 percent (CT-scan images) were obtained. The model's loss functions, Precision, Recall, and F1-Score, were extracted and interpreted in the study. We examined the researchers' modified DL models and discovered that they were 65 percent accurate on X-ray data and 62 percent accurate on CT scan images. Experiments have demonstrated that when the number of sample images rises, the VGG19 and InceptionV3 perform well.
Medicinal Plant Classification Using Particle Swarm Optimized Cascaded Network
(Institute of Electrical and Electronics Engineers Inc., 2024-01-12) Islam, M.T.; Rahman, W.; Hossain, M.S.; Roksana, K.; Azpiroz, I.D.,; Diaz, R.M.; Ashraf, I.; Samad, M.A.
Medicinal plants are essential to healthcare since ancient times and are integral to developing drugs and other medical treatments. More than 25% of medicines in developed countries are produced from medicinal plants, while in developing countries, approximately 80% of individuals receive primary healthcare from these plants. Traditionally, these plants are identified manually by experts, which is tedious, time-consuming, subjective and dependent on the availability of experts. Furthermore, a wrong detection can result in serious health issues or death. This signifies the need for a more reliable approach to identifying medicinal plants, which is accurate and practical. Several automated methods were proposed previously, utilizing deep learning and traditional machine learning (TML) techniques, but they require singular leaf images and failed to achieve sufficient accuracy when demonstrated in a different setting. Capturing singular leaf images for each plant is also time-consuming and laborious. This paper presents a robust, accurate and practical system to identify medicinal plants from smartphone-captured plant images in the site of plants. The proposed system utilized a cascaded architecture to extract features using a pre-trained ResNet50 model, which were optimized using Particle Swarm Optimization (PSO) to classify the plants using a Support Vector Machine (SVM). The proposed ResNet50-PSO-SVM network classified seven medicinal plants with 99.60% accuracy, outperforming the state-of-the-art (99%). The system was demonstrated for three different smartphones, classifying an image in 0.15 seconds with 97.79% accuracy on average. The system's high accuracy, rapid identification time and robustness ensured its practical use.
News Modeling and Retrieving Information: Data-Driven Approach
(Tech Science Press, 2024-02-05) Hossain, E.; Alshahrani, A.; Rahman, W.
This paper aims to develop Machine Learning algorithms to classify electronic articles related to this phenomenon by retrieving information and topic modelling. The Methodology of this study is categorized into three phases: the Text Classification Approach (TCA), the Proposed Algorithms Interpretation (PAI), and finally, Information Retrieval Approach (IRA). The TCA reflects the text preprocessing pipeline called a clean corpus. The Global Vectors for Word Representation (Glove) pre-trained model, FastText, Term Frequency-Inverse Document Frequency (TF-IDF), and Bag-of-Words (BOW) for extracting the features have been interpreted in this research. The PAI manifests the Bidirectional Long Short-Term Memory (Bi-LSTM) and Convolutional Neural Network (CNN) to classify the COVID-19 news. Again, the IRA explains the mathematical interpretation of Latent Dirich-let Allocation (LDA), obtained for modelling the topic of Information Retrieval (IR). In this study, 99% accuracy was obtained by performing K-fold cross-validation on Bi-LSTM with Glove. A comparative analysis between Deep Learning and Machine Learning based on feature extraction and computational complexity exploration has been performed in this research. Furthermore, some text analyses and the most influential aspects of each document have been explored in this study. We have utilized Bidirectional Encoder Representations from Trans-formers (BERT) as a Deep Learning mechanism in our model training, but the result has not been uncovered satisfactory. However, the proposed system can be adjustable in the real-time news classification of COVID-19
Paddy insect identification using deep features with lion optimization algorithm
(Elsevier Ltd, 2024-06-05) Elmagzoub, M.A.; Rahman, W.; Roksana, K.; Islam, M.T.; Sadi, A.H.M.S.; Rahman, M.M.; Rajab, A.; Rajab, K.; Shaikh, A.
Pests are a significant challenge in paddy cultivation, resulting in a global loss of approximately 20 % of rice yield. Early detection of paddy insects can help to save these potential losses. Several ways have been suggested for identifying and categorizing insects in paddy fields, employing a range of advanced, noninvasive, and portable technologies. However, none of these systems have successfully incorporated feature optimization techniques with Deep Learning and Machine Learning. Hence, the current research provided a framework utilizing these techniques to detect and categorize images of paddy insects promptly. Initially, the suggested research will gather the image dataset and categorize it into two groups: one without paddy insects and the other with paddy insects. Furthermore, various pre-processing techniques, such as augmentation and image filtering, will be applied to enhance the quality of the dataset and eliminate any unwanted noise. To determine and analyze the deep characteristics of an image, the suggested architecture will incorporate 5 pre-trained Convolutional Neural Network models. Following that, feature selection techniques, including Principal Component Analysis (PCA), Recursive Feature Elimination (RFE), Linear Discriminant Analysis (LDA), and an optimization algorithm called Lion Optimization, were utilized in order to further reduce the redundant number of features that were collected for the study. Subsequently, the process of identifying the paddy insects will be carried out by employing 7 ML algorithms. Finally, a set of experimental data analysis has been conducted to achieve the objectives, and the proposed approach demonstrates that the extracted feature vectors of ResNet50 with Logistic Regression and PCA have achieved the highest accuracy, precisely 99.28 %. However, the present idea will significantly impact how paddy insects are diagnosed in the field.
Species classification of brassica napus based on flowers, leaves, and packets using deep neural networks
(Elsevier B.V., 2023-12) Alom, M.,; Ali, M.Y.,; Islam, M.T.,; Uddin, A.H.,; Rahman, W.
Deep learning (DL) has gradually taken the lead as the most effective approach in the agricultural fields due to the early identification and classification of plant species and diseases for improving the quality of crop production because of recent technological breakthroughs, which have had a significant impact on agriculture. Plenty of complicated problems in farming, including species classification, plant disorder identification, yield approximation, and weather and soil moisture prediction, are made simple using deep neural networks. Thus, this proposed study aims to classify Brassica Napus (B. Napus) rapeseed species based on their most significant features, like flowers, leaves, and packets. The study has adopted two types of rapeseed such as B. Rapa and B. Alba. Five contemporary deep learning-based Convolutional Neural Network (CNN) models have also been assessed for distinguishing rapeseed species. These models are DenseNet201, VGG19, InceptionV3, Xception, and ResNet50. Initially, the researchers collected data from the agricultural field, and then image pre-processing is performed to create our dataset. After that, CNN models were applied to this dataset and enumerated the experimental data accordingly. Our DenseNet201 model successfully classified both species with the highest accuracy of 100% for flowers and 97% for both packets and leaves. A comprehensive analysis with companion studies confirmed the efficacy of our preferred paradigm for the near future. Nevertheless, future studies will compare these methodologies to data from a separate metabolomics dataset from comparable crops.