Browsing by Author "Uddin, Mohammad Shorif"

Now showing 1 - 20 of 22

An Explorative Analysis on the Machine-Vision-Based Disease Recognition of Three Available Fruits of Bangladesh
(Daffodil International University, 2021-09-18) Habib, Md. Tarek; Mia, Md. Jueal; Uddin, Mohammad Shorif; Ahmed, Farruk
Bangladesh, being a densely populated country, hinges on agriculture for the security of finance and food to a large extent. Hence, both the fruits’ quantity and quality turn out to be very important, which can be degraded due to the attacks of various diseases. Automated fruit disease recognition can help fruit farmers, especially remote farmers, for whom adequate cultivation support is required. Two daunting problems, namely disease detection, and disease classification are raised by automated fruit disease recognition. In this research, we conduct an intense investigation of the applicability of automated recognition of the diseases of three available Bangladeshi local fruits, viz. guava, jackfruit, and papaya. After exerting four notable segmentation algorithms, -means clustering segmentation algorithm is selected to segregate the disease-contaminated parts from a fruit image. Then some discriminatory features are extracted from these disease-contaminated parts. Nine noteworthy classification algorithms are applied for disease classification to thoroughly get the measure of their merits. It is observed that random forest outperforms the eight other classifiers by disclosing an accuracy of 96.8% and 89.59% for guava and jackfruit, respectively, whereas support vector machine attains an accuracy of 94.9% for papaya, which can be claimed good as well as attractive for forthcoming research.
An Explorative Analysis on the Machine-Vision-Based Disease Recognition of Three Available Fruits of Bangladesh
(Scopus, 2021) Habib, Md. Tarek; Mia, Md. Jueal; Uddin, Mohammad Shorif; Ahmed, Farruk
Bangladesh, being a densely populated country, hinges on agriculture for the security of finance and food to a large extent. Hence, both the fruits’ quantity and quality turn out to be very important, which can be degraded due to the attacks of various diseases. Automated fruit disease recognition can help fruit farmers, especially remote farmers, for whom adequate cultivation support is required. Two daunting problems, namely disease detection, and disease classification are raised by automated fruit disease recognition. In this research, we conduct an intense investigation of the applicability of automated recognition of the diseases of three available Bangladeshi local fruits, viz. guava, jackfruit, and papaya. After exerting four notable segmentation algorithms, K-means clustering segmentation algorithm is selected to segregate the disease-contaminated parts from a fruit image. Then some discriminatory features are extracted from these disease-contaminated parts. Nine noteworthy classification algorithms are applied for disease classification to thoroughly get the measure of their merits. It is observed that random forest outperforms the eight other classifiers by disclosing an accuracy of 96.8% and 89.59% for guava and jackfruit, respectively, whereas support vector machine attains an accuracy of 94.9% for papaya, which can be claimed good as well as attractive for forthcoming research.
An Extensive Real-World in Field Tomato Image Dataset Involving Maturity Classification and Recognition of Fresh and Defect Tomatoes
(Elsevier, 2023-10-15) Khatun, Tania; Razzak, Abdur; Md. Shofiul Islam, Md. Shofiul; Uddin, Mohammad Shorif
Tomato, a fruiting plant species within the Solanaceae family, is a widely used ingredient in culinary dishes due to its sweet and acidic flavor profile, as well as its rich nutritional content. Recognized for its potential health benefits, including reducing the risk of coronary artery disease and specific types of cancer, tomatoes have become a staple in global cuisine. Traditional methods for tomato maturity assessment, harvesting, quality grading, and packaging are often labor-intensive and economically inefficient. This paper introduces an extensive dataset of high-resolution tomato images collected over an eight-month period from the demonstration fields of Sher-E-Bangla Agricultural University in Dhaka, Bangladesh, in collaboration with plant breeding experts of the same university. The dataset was meticulously curated to ensure precision and consistency, encompassing various stages of tomato maturity, including images of both fresh and defective tomatoes. This dataset is a valuable resource for researchers, stakeholders, and individuals interested in tomato production in Bangladesh, providing a robust foundation for leveraging computer vision and deep learning techniques in the agriculture sector. The dataset's potential applications extend to automating tasks such as robotic harvesting, quality assessment, and packaging systems, ultimately enhancing the efficiency of tomato production processes.
An Extensive Sunflower Dataset Representation for Successful Identification and Classification of Sunflower Diseases
(Daffodil International University, 2022-05-13) Sara, Umme; Rajbongshi, Aditya; Shakil, Rashiduzzaman; Akter, Bonna; Sazzad, Sadia; Uddin, Mohammad Shorif
Sunflowers are agricultural seed crops that can be used for essential edible oils and ornamental purposes. This cash crop is primarily cultivated in North and South America. Sunflower crops are prone to various diseases, insects, and nematodes, resulting in a wide range of production losses. Digital image processing and computer vision approaches have been widely utilized to categorize and detect plant diseases including leaves, fruits, and flowers over the last few decades. Early diagnosis of infections in sunflowers helps to prevent them from spreading throughout the farm and reducing financial losses to the farmers. This article offers a resourceful dataset of sunflower leaves and flowers that will help the researchers in developing effective algorithms for the detection of diseases. The dataset contains healthy and affected sunflower leaves and flowers with downy mildew, gray mold, and leaf scars. The images were captured manually between 25th to 29th November 2021 from the demonstration farm of Bangladesh Agricultural Research Institute (BARI) at Gazipur in cooperation with its one domain expert when the sunflower plants were about to bloom and the maximum diseases can be found. The dataset is hosted by the Department of Computer Science and Engineering, National Institute of Textile Engineering and Research (NITER), Bangladesh and freely available at https://data.mendeley.com/datasets/b83hmrzth8/1.
An In-Depth Exploration of Automated Jackfruit Disease Recognition
(Daffodil International University, 2022-05-05) Habib, Md. Tarek; Mia, Md. Jueal; Uddin, Mohammad Shorif; Ahmed, Farruk
Bangladesh extensively depends on agriculture for the economy as well as food security owing to its huge population. In this connection, it becomes very important to efficiently grow plants and increase their yields. Quantity and quality of fruits can be degraded having attacked by various diseases. It is a matter of fact that not even a single research work has been conducted for automated recognition of jackfruit diseases to facilitate those distant farmers who need proper cultivation support. Presuming that our context is the recognition of jackfruit diseases, two challenging problems are mainly raised, i.e. detection of diseases and classification of diseases. In this research, we perform an in-depth investigation of an agro-medical expert system, which proceeds with a digital image acquired with a cellphone or other handheld device and recognizes the disease. Exhaustive experiments have been performed to assess the feasibility of our intended expert system. At first, a discriminatory feature set is selected. k-means clustering segmentation is put into action to detect disease-affected regions of an image of a disease-attacked jackfruit and extract the features from these regions. Then classification of the diseases is accomplished by using nine off-the-shelf classification algorithms in order to thoroughly assess the merits of the classifiers in the index of seven prominent performance metrics. Random forest is found outperforming all other classifiers to the amount of all metrics used by attaining an accuracy approaching to 90%. On the contrary, logistic regression shows not only the poorest result of an accuracy approaching to 75% but also some other poorest metric-values.
An in-depth Exploration of Automated Jackfruit Disease Recognition
(Journal of King Saud University - Computer and Information Sciences, 2020) Habib, Md. Tarek; Mia, Md. Jueal; Uddin, Mohammad Shorif; Ahmed, Farruk
Bangladesh extensively depends on agriculture for the economy as well as food security owing to its huge population. In this connection, it becomes very important to efficiently grow plants and increase their yields. Quantity and quality of fruits can be degraded having attacked by various diseases. It is a matter of fact that not even a single research work has been conducted for automated recognition of jackfruit diseases to facilitate those distant farmers who need proper cultivation support. Presuming that our context is the recognition of jackfruit diseases, two challenging problems are mainly raised, i.e. detection of diseases and classification of diseases. In this research, we perform an in-depth investigation of an agro-medical expert system, which proceeds with a digital image acquired with a cellphone or other handheld device and recognizes the disease. Exhaustive experiments have been performed to assess the feasibility of our intended expert system. At first, a discriminatory feature set is selected. k-means clustering segmentation is put into action to detect disease-affected regions of an image of a disease-attacked jackfruit and extract the features from these regions. Then classification of the diseases is accomplished by using nine off-the-shelf classification algorithms in order to thoroughly assess the merits of the classifiers in the index of seven prominent performance metrics. Random forest is found outperforming all other classifiers to the amount of all metrics used by attaining an accuracy approaching to 90%. On the contrary, logistic regression shows not only the poorest result of an accuracy approaching to 75% but also some other poorest metric-values.
Analyzing the Public Sentiment on COVID-19 Vaccination in Social Media
(Daffodil International University, 22-06-12) Zulfiker, Md. Sabab; Kabir, Nasrin; Biswas, Al Amin; Zulfiker, Sunjare; Uddin, Mohammad Shorif
Since December 2019, the world has been fighting against the COVID-19 pandemic. This epidemic has revealed a bitter truth that though humans have advanced to unprecedented heights in the last few decades in terms of technology, they are lagging far behind in the fields of medical science and health care. Several institutes and research organizations have stepped up to introduce different vaccines to combat the pandemic. Bangladesh government has also taken steps to provide widespread vaccinations from January 2021. The Bangladeshi netizens are frequently sharing their thoughts, emotions, and experiences about the COVID-19 vaccines and the vaccination process on different social media sites like Facebook, Twitter, etc. This study has analyzed the views and opinions that they have expressed on different social media platforms about the vaccines and the ongoing vaccination program. For performing this study, the reactions of the Bangladeshi netizens on social media have been collected. The Latent Dirichlet Allocation (LDA) model has been used to extract the most common topics expressed by the netizens regarding the vaccines and vaccination process in Bangladesh. Finally, this study has applied different deep learning as well as traditional machine learning algorithms to identify the sentiments and polarity of the opinions of the netizens. The performance of these models has been assessed using a variety of metrics such as accuracy, precision, sensitivity, specificity, and F1-score to identify the best one. Sentiment analysis lessons from these opinions can help the government to prepare itself for the future pandemic.
Audio Watermarking: A Comprehensive Review
(Scopus, 2024) Uddin, Mohammad Shorif; jaman, Ohiduj; Hasan, Mahmudul; Shimamura, Tetsuya
Audio watermarking has emerged as a potent tech nology for copyright protection, content authentication, content monitoring, and tracking in the digital age. This paper offers a comprehensive exploration of audio watermarking principles, techniques, applications, and challenges. Initially, it presents the fundamental concepts of digital watermarking, elucidating its key characteristics and functionalities. After that, different audio watermarking methods in both the time and transform domains are explained, such as feature-based, parametric, and spread spectrum methods, along with how they work, and their pros and cons. The paper further addresses critical challenges in maintaining key criteria such as imperceptibility, robustness, and payload capacity associated with audio watermarking. Addi tionally, it examines watermarking evaluation metrics, datasets, and performance findings under diverse signal-processing attacks. Finally, the review concludes by discussing future directions in audio watermarking research, emphasizing advancements in deep learning-based approaches and emerging applications..
Bangla Handwritten Digit Recognition Using Deep Convolutional Neural Network
(ACM International Conference Proceeding Series, 2020) Basri, Rabeya; Haque, Mohammad Reduanul; Akter, Morium; Uddin, Mohammad Shorif
Handwritten Bangla digit recognition is one of the most challenging computer vision problems due to its diverse shapes and writing style. Recently deep learning based convolutional neural network known as deep CNN finds wide-spread applications in recognizing different objects due to its high accuracy. This paper investigates the performance of some state-of-the-art deep CNN techniques for the recognition of handwritten digits. It considers four deep CNN architectures, such as AlexNet, MobileNet, GoogLeNet (Inception V3), and CapsuleNet models. These four deep CNNs have been experimented on a large, unbiased and highly augmented standard dataset, NumtaDB and confirmed that the AlexNet showed the best performance on the basis of accuracy and computation time.
Convolutional Neural Network Modeling for Eye Disease Recognition
(2022) Siddique, Md. Ashikul Aziz; Ferdouse, Jannatul; Habib, Md. Tarek; Mia, Md. Jueal; Uddin, Mohammad Shorif
The eye is an important sensing organ of the human body, as it reacts to light and allows vision of humans. Many Bangladeshi people become nearsighted when it comes to the awareness of vision loss due to eye disease. Many Bangladeshis people are more concerned about losing their money than getting nearsighted or blind, due to a combination of poverty and illiteracy. With this view, this paper proposes an osteopathic expert system that can deal with an image of the eye and recognize the disease. Here, we have focused on the three most common eye diseases in Bangladesh, namely cataract, chalazion, and squint. We have modeled six convolutional neural networks (CNN’s), namely VGG16, VGG19, MobileNet, Xception, InceptionV3, and DenseNet121 to recognize the diseases. We have reached the best configuration of each of these CNN models after adequate investigation. After performing satisfactory experimentation, we have found that the MobileNet model gives the best performance based on accuracy, precision, recall, and F1-score. At last, we have compared our findings with the recently reported relevant works to show their efficacy.
Crime Detection and Criminal Recognition to Intervene in Interpersonal Violence Using Deep Convolutional Neural Network with Transfer Learning
(International Journal of Ambient Computing and Intelligence, 2021) Haque, Mohammad Reduanul; Hafiz, Rubaiya; Al Azad, Alauddin; Adnan, Yeasir; Akter, Sharmin; Khatun, Amina; Uddin, Mohammad Shorif
Interpersonal violence, such as physical and sexual abuse, eve-teasing, bullying, and taking hostages, is a growing concern in our society. The criminals who directly or indirectly committed the crime often do not go into the trial for the lack of proper evidence as it is very tough to collect photographic proof of the incident. A subject's corneal reflection has the potentiality to reveal the bystander images. Motivated with this clue, a novel approach is proposed in the current paper that uses a convolutional neural network (CNN) along with transfer learning in identifying crime as well as recognizing the criminals from the corneal reflected image of the victim called the Purkinje image. This study found that off-the-shelf CNN can be fine-tuned to extract discriminative features from very low resolution and noisy images. The procedure is validated using the developed datasets comprising six different subjects taken at diverse situations. They confirmed that it has the ability to recognize criminals from corneal reflection images with an accuracy of 95.41%.
Degraded Document Enhancement through Binarization Techniques
(2019 International Conference on Sustainable Technologies for Industry 4.0 (STI), IEEE, 2019-12-25) Bonny, Moushumi Zaman; Uddin, Mohammad Shorif
Enhancement of degraded documents is one of the significant and challenging research areas. In recent years, several binarization methods are proposed and presented for the improvement of degraded documents, but, most of them are not appropriate for all kinds of degradation. In this paper, we have described some state-of-the-art binarization techniques and compared their performances using DIBCO 2016 to DIBCO 2018 databases. In addition, we briefly discussed about the challenges and possible future works of image binarization.
Image-based automated haze removal using dark channel prior
(IEEE, 2018-02-12) Uddin, Mohammad Shorif; Gautam, Bishal; Sarker, Aditi; Akter, Morium; Haque, Mohammad Reduanul
Haze, fog, rain usually hampered the performance of vision systems. So, removal of haze appearance in a scene should be the first priority for clear vision. Previously a dehazing mechanism was developed based on dark channel prior which cannot automatically set the patch size and the sky-region's transmission value. The current paper tries to fill this gap to automate these values. Experimentation has been conducted to find the performance on the basis of subjective as well as objective metrics. We have obtained satisfactory results compared to the existing techniques.
Image-Based Soft Drink Type Classification and Dietary Assessment System Using Deep Convolutional Neural Network with Transfer Learning
(Daffodil International University, 2020-09-09) Hafiz, Rubaiya; Haque, Mohammad Reduanul; Rakshit, Aniruddha; Uddin, Mohammad Shorif
Nowadays, people are taking soft drinks (carbonated nonalcoholic beverages) at an increasing rate. Health experts around the world have cautioned from time to time that these drinks lead to weight gain, raise the risk of non-communicable diseases, and so on. To develop consciousness among people, the present work describes an image-based tool to self-monitor the nutritional information of soft drinks by using a deep convolutional neural network (CNN) along with transfer learning. At first, a pre-processing function is done through noise reduction and contrast enhancement. Then the location of the drinks region is extracted through visual saliency and mean-shift segmentation technique. After removing backgrounds and segment out only the region of interest from the image a deep CNN-based transfer learning model is employed for the drink classification. Finally, the size of each drink bottle is estimated using the bag-of-feature (BoF) and distance ratio calculation to find the nutrition value from the nutrition fact table. To perform experimentation a dataset is built containing ten most consumed soft drinks in Bangladesh using images from the ImageNet dataset, internet sources and also self-capturing. The experiment confirms that our system can detect and recognize different types of drinks with an accuracy of 98.51%.
Image-based Soft Drink Type Classification and Dietary Assessment System Using Deep Convolutional Neural Network with Transfer Learning
(Journal of King Saud University - Computer and Information Sciences, 2020-09-09) Hafiz, Rubaiya; Rakshit, Aniruddha; Uddin, Mohammad Shorif; Haque, Mohammad Reduanul
Nowadays, people are taking soft drinks (carbonated nonalcoholic beverages) at an increasing rate. Health experts around the world have cautioned from time to time that these drinks lead to weight gain, raise the risk of non-communicable diseases, and so on. To develop consciousness among people, the present work describes an image-based tool to self-monitor the nutritional information of soft drinks by using a deep convolutional neural network (CNN) along with transfer learning. At first, a pre-processing function is done through noise reduction and contrast enhancement. Then the location of the drinks region is extracted through visual saliency and mean-shift segmentation technique. After removing backgrounds and segment out only the region of interest from the image a deep CNN-based transfer learning model is employed for the drink classification. Finally, the size of each drink bottle is estimated using the bag-of-feature (BoF) and distance ratio calculation to find the nutrition value from the nutrition fact table. To perform experimentation a dataset is built containing ten most consumed soft drinks in Bangladesh using images from the ImageNet dataset, internet sources and also self-capturing. The experiment confirms that our system can detect and recognize different types of drinks with an accuracy of 98.51%.
Machine Vision Based Papaya Disease Recognition
(Journal of King Saud University - Computer and Information Sciences, Elsevier, 2020-03) Habib, Md. Tarek; Majumder, Anup; Jakaria, A.Z.M.; Akter, Morium; Uddin, Mohammad Shorif; Ahmed, Farruk
Over the years little research has been performed for vision-based papaya disease recognition system in order to help distant farmers, most of whom require proper support for cultivation. Due to advancement of vision-based technology we find a good solution to this problem. Papaya disease recognition mainly involves two challenging problems: one is disease detection and another is disease classification. Considering this scenario, here we present an online machine vision-based agro-medical expert system that processes an image captured through mobile or handheld device and determines the diseases in order to help distant farmers to address the problem. Some experiments are performed to show the utility of the proposed expert system. First, we propose a set of features from the view point of distinguishing attributes. K-means clustering algorithm is used in order to segment out the disease-attacked region from the captured image and then required features are extracted to classify the diseases with the help of support vector machine. More than 90% classification accuracy has been achieved, which appears to be good as well as promising by comparing performances obtained with recently reported relevant works.
Quantitative Analysis of Deep Cnns for Multilingual Handwritten Digit Recognition
(Scopus, 2021) Haque, Mohammad Reduanul; Azam, Md. Gausul; Milon, Sarwar Mahmud; Hossain, Md. Shaheen; Molla, Md. Al-Amin; Uddin, Mohammad Shorif
Indian subcontinent is a birthplace of multilingual people, where documents such as job application form, passport, number plate identification, and so forth are composed of text contents written in different languages or scripts. These scripts consist of different Indic numerals in a single document page. Recently, deep convolutional neural networks (CNN) have achieved favorable result in computer vision problems, especially in recognizing handwritten digits but most of the works focuses on only one language, i.e., English or Hindi or Bangla, etc. However, developing a language-invariant method is very important as we live in a global village now. In this work, we have examined the performance of the ten state-of-the-art deep CNN methods for the recognition of handwritten digits using four most common languages in the Indian sub-continent that creates the foundation of a script invariant handwritten digit recognition system. Among the deep CNNs, Inception-v4 performs the best based on accuracy and computation time. Besides, it discusses the limitations of existing techniques and shows future research directions.
Spectral Analysis of Bone-conducted Speech Using Modified Linear Prediction
(Springer Nature, 2024-10-16) Ohidujjaman; Hasan, Mahmudul; Zhang, Shiming; Huda, Mohammad Nurul; Uddin, Mohammad Shorif
This paper improves the performance of linear prediction (LP) in precise spectral estimation of bone-conducted (BC) speech. Inherently, BC speech contains a wide spectral dynamic range that causes ill conditioning in the autocorrelation (ACR) method and its variants, where the Levinson–Durbin (L–D) algorithm is commonly implemented. Instead of the conventional LP-based spectral estimation methods, we utilize the covariance-based method, specifically the modified covariance (MC) method, where the orthogonal decomposition algorithm is deployed. In this paper, we derive the MC method from the least squares (LS) technique for BC speech analysis. The MC method reduces the eigenvalue expansion that compresses the spectral dynamic range of the BC speech signal. The effect of spectral dynamic range compression declines the ill-conditioned properties of LP. Through the proposed method using synthetic BC speech, the resulting power spectrum provides more accurate peaks than the conventional methods. The validity of the proposed method is also analyzed by inspecting real BC speech. This study reveals the utmost use of BC speech in speech processing systems. The experimental results demonstrate that the proposed method provides more accurate spectral estimation for synthetic and real BC speeches compared with conventional spectral estimation methods.
VegNet
(Daffodil International University, 2022-06-26) Sara, Umme; Rajbongshi, Aditya; Shakil, Rashiduzzaman; Akter, Bonna; Uddin, Mohammad Shorif
Cauliflower, a winter seasoned vegetable that originated in the Mediterranean region and arrived in Europe at the end of the 15th century, takes the lead in production among all vegetables. It's high in fiber and can keep us hydrated, and have medicinal properties like the chemical glucosinolates, which may help prevent cancer. If proper care is not given to the plants, several significant diseases can affect the plants, reducing production, quantity, and quality. Plant disease monitoring by hand is extremely tough because it demands a great deal of effort and time. Early detection of the diseases allows the agriculture sector to grow cauliflower more efficiently. In this scenario, an insightful and scientific dataset can be a lifesaver for researchers looking to analyze and observe different diseases in cauliflower development patterns. So, in this work, we present a well-organized and technically valuable dataset “VegNet’ to effectively recognize conditions in cauliflower plants and fruits. Healthy and disease-affected cauliflower head and leaves by black rot, downy mildew, and bacterial spot rot are included in our suggested dataset. The images were taken manually from December 20th to January 15th, when the flowers were fully blown, and most of the diseases were observed clearly. It is a well-organized dataset to develop and validate machine learning-based automated cauliflower disease detection algorithms.
XAI-FusionNet: Diabetic foot ulcer detection based on multi-scale feature fusion with explainable artificial intelligence
(Scopus, 2024) Biswas, Shuvo; Mostafiz, Rafid; Uddin, Mohammad Shorif; Paul, Bikash Kumar
Diabetic foot ulcer (DFU) poses a significant threat to individuals affected by diabetes, often leading to limb amputation. Early detection of DFU can greatly improve the chances of survival for diabetic patients. This work introduces FusionNet, a novel multi-scale feature fusion network designed to accurately differentiate DFU skin from healthy skin using multiple pre-trained convolutional neural network (CNN) algorithms. A dataset comprising 6963 skin images (3574 healthy and 3389 ulcer) from various patients was divided into training (6080 images), validation (672 images), and testing (211 images) sets. Initially, three image preprocessing techniques - Gaussian filter, median filter, and motion blur estimation - were applied to eliminate irrelevant, noisy, and blurry data. Subsequently, three pre-trained CNN algorithms -DenseNet201, VGG19, and NASNetMobile - were utilized to extract high-frequency features from the input images. These features were then inputted into a meta-tuner module to predict DFU by selecting the most discriminative features. Statistical tests, including Friedman and analysis of variance (ANOVA), were employed to identify significant