XAI-FusionNet: Diabetic foot ulcer detection based on multi-scale feature fusion with explainable artificial intelligence.

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 differences between FusionNet and other sub-networks. Finally, three eXplainable Artificial Intelligence (XAI) algorithms - SHAP (SHapley Additive exPlanations), LIME (Local Interpretable Model-agnostic Explanations), and Grad-CAM (Gradient-weighted Class Activation Mapping) - were integrated into FusionNet to enhance transparency and explainability. The FusionNet classifier achieved exceptional classification results with 99.05 % accuracy, 98.18 % recall, 100.00 % precision, 99.09 % AUC, and 99.08 % F1 score. We believe that our proposed FusionNet will be a valuable tool in the medical field to distinguish DFU from healthy skin.

A visual dataset for recognition of rice varieties.

This article presents a comprehensive dataset sourced from various markets across Bangladesh, highlighting 20 distinct rice varieties predominantly consumed locally. The dataset encompasses a diverse range of rice strains, including Subol Lota, Bashmoti (Deshi), Ganjiya, Shampakatari, Sugandhi Katarivog, BR-28, BR-29, Paijam, Bashful, Lal Aush, BR-Jirashail, Gutisharna, Birui, Najirshail, Pahari Birui, Polao (Katari), Polao (Chinigura), Amon, Shorna-5, and Lal Binni. Using a smartphone camera, low-resolution images capturing the essence of each rice variety were meticulously obtained, resulting in a total of 4,730 images with a non-uniform distribution. The dataset also includes augmented data, totaling 23,650 images. This precisely curated dataset holds significant promise and utility, showcasing diverse attributes, including the unique representation of 20 rice varieties, each characterized by distinct colors, sizes, and potential applications within the agricultural sector.

A comprehensive dragon fruit image dataset for detecting the maturity and quality grading of dragon fruit.

Dragon fruit, often referred to as pitaya, is a tropical fruit with various types, including both white-fleshed and red-fleshed varieties. Its distinctive appearance is complemented by a range of potential health advantages. These include its abundance of nutrients and antioxidants, which contribute to a robust immune system, aid in blood sugar regulation, and support the well-being of the heart, bones, and skin. Consequently, the global desire for dragon fruit is yielding substantial economic advantages for developing nations like Bangladesh, which in turn underscores the pressing need for an automated system to identify the optimal harvest time and differentiate between fresh and defective fruits to ensure quality. To accomplish this objective, this paper introduces an extensive collection of high-resolution dragon fruits because effective detection by machine learning models necessitates a substantial amount of data. The dataset was painstakingly gathered during a span of four months from three distinct locations in Bangladesh, with the valuable assistance of domain experts. Possible application of the dataset encompasses quality evaluation, robotic harvesting, and packaging systems, ultimately boosting the effectiveness of dragon fruit production procedures. The dataset has the potential to be a valuable resource for researchers interested in dragon fruit cultivation, offering a solid foundation for the application of computer vision and deep learning methods in the agricultural industry.

A dataset for assessing real-time attention levels of the students during online classes.

This dataset offers a comprehensive compilation of attention-related features captured during online classes. The dataset is generated through the integration of key components including face detection, hand tracking, head pose estimation, and mobile phone detection modules. The data collection process involves leveraging a web interface created using the Django web framework. Video frames of participating students are collected following institutional guidelines and informed consent through their webcams, subsequently decomposed into frames at a rate of 20 FPS, and transformed from BGR to RGB color model. The aforesaid modules subsequently process these video frames to extract raw data. The dataset consists of 16 features and one label column, encompassing numerical, categorical, and floating-point values. Inherent to its potential, the dataset enables researchers and practitioners to explore and examine attention-related patterns and characteristics exhibited by students during online classes. The composition and design of the dataset offer a unique opportunity to delve into the correlations and interactions among face presence, hand movements, head orientations, and phone interactions. Researchers can leverage this dataset to investigate and develop machine learning models aimed at automatic attention detection, thereby contributing to enhancing remote learning experiences and educational outcomes. The dataset in question also constitutes a highly valuable resource for the scientific community, enabling a thorough exploration of the multifaceted aspects pertaining to student attention levels during online classes. Its rich and diverse feature set, coupled with the underlying data collection methodology, provides ample opportunities for reuse and exploration across multiple domains including education, psychology, and computer vision research.

A dataset for successful recognition of cucumber diseases.

Plant disease is a common impediment to the productivity of the world's agricultural production, which adversely affects the quality and yield of crops and causes heavy economic losses to farmers. The cucumber is a frequently grown creeping vine plant that has few calories but is high in water and several vital vitamins and minerals. Due to the unfavorable ecological environment and non-biological circumstances, cucumber diseases will adversely harm the quality of cucumber and cause heavy financial loss. Early identification and protection of crop diseases are essential for disease management, crop yield enhancement, cost reduction, and boosting agricultural production. The traditional diagnosis of crop diseases is often time-consuming, laborious, ineffective, and subjective. To cope with this scenario, the development of a machine-based model which can detect cucumber diseases is a demand of time for increasing agricultural production. This article offers a major cucumber dataset to build an effective machine vision-based model which can detect more variety of cucumber diseases. The dataset includes eight different types of classes containing disease-affected and disease-free cucumber images (Anthracnose, Bacterial Wilt, Belly Rot, Downy Mildew, Pythium Fruit Rot, Gummy Stem Blight, Fresh leaves, and Fresh cucumber) which were collected from the 6th to 30th of May 2022 from real fields with the cooperation of an expert from an agricultural institution. The dataset is hosted by the Department of Computer Science and Engineering, Jahangirnagar University, and is freely accessible at https://data.mendeley.com/datasets/y6d3z6f8z9/1.

An extensive real-world in field tomato image dataset involving maturity classification and recognition of fresh and defect tomatoes.

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 dataset for successful recognition of fresh and rotten fruits.

Detection of rotten fruits is very crucial for agricultural productions and fruit processing as well as packaging industries. Usually, the detection of fresh and rotten fruits is done manually which is an ineffective and lengthy process for farmers. For this reason, the development of a new classification model is required which will reduce human effort, cost, and production time in the agriculture industry by recognizing defects in the fruits. This article offers a major dataset to the researchers to develop effective algorithms for recognizing more variety of fruits and overcome the limitations by increasing accuracy as well as decreasing computation time. This dataset contains sixteen types of fruit classes, namely fresh grape, rotten grape, fresh guava, rotten guava, fresh jujube, rotten jujube, fresh pomegranate, rotten pomegranate, fresh strawberry, rotten strawberry, fresh apple, rotten apple, fresh banana, rotten banana, fresh orange, rotten orange. We collected various fresh and rotten fruit images from 16th to 31st March 2022 from different fruit shops and real fields with the help of a domain specialist from an agricultural organization. The dataset is hosted by the Department of Computer Science and Engineering, Jahangirnagar University, and is freely available at https://data.mendeley.com/datasets/bdd69gyhv8/1.

VegNet: An organized dataset of cauliflower disease for a sustainable agro-based automation system.

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. The dataset is hosted by the Institute - National Institute of Textile Engineering and Research (NITER),the Department of Computer Science and Engineering and is available at the link following: https://data.mendeley.com/datasets/t5sssfgn2v/3.

Analyzing the public sentiment on COVID-19 vaccination in social media: Bangladesh context.

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.

An extensive sunflower dataset representation for successful identification and classification of sunflower diseases.

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.

Covid-19 detection in chest X-ray through random forest classifier using a hybridization of deep CNN and DWT optimized features.

Chest X-ray image contains sufficient information that finds wide-spread applications in diverse disease diagnosis and decision making to assist the medical experts. This paper has proposed an intelligent approach to detect Covid-19 from the chest X-ray image using the hybridization of deep convolutional neural network (CNN) and discrete wavelet transform (DWT) features. At first, the X-ray image is enhanced and segmented through preprocessing tasks, and then deep CNN and DWT features are extracted. The optimum features are extracted from these hybridized features through minimum redundancy and maximum relevance (mRMR) along with recursive feature elimination (RFE). Finally, the random forest-based bagging approach is used for doing the detection task. An extensive experiment is performed, and the results confirm that our approach gives satisfactory performance compare to the existing methods with an overall accuracy of more than 98.5%.

Restoration of uneven illumination in light sheet microscopy images.

Light microscopy images suffer from poor contrast due to light absorption and scattering by the media. The resulting decay in contrast varies exponentially across the image along the incident light path. Classical space invariant deconvolution approaches, while very effective in deblurring, are not designed for the restoration of uneven illumination in microscopy images. In this article, we present a modified radiative transfer theory approach to solve the contrast degradation problem of light sheet microscopy (LSM) images. We confirmed the effectiveness of our approach through simulation as well as real LSM images.

An Efficient Watermarking Approach Based on LL and HH Edges of DWT-SVD.

Digital watermarking is playing a vital role in the improvement of authentication, security, and copyright protection in today's digital transformation. The performance of this technique is shown to be impressive around the globe. Text, audio, video, and image data are acted as watermarks in the digital platform. In this article, a hybrid watermarking scheme is proposed to furnish the robustness and protection of digital data. This hybrid scheme is a form of discrete wavelet transform (DWT) and singular value decomposition (SVD). The embedding and extracting features are carried out through multi-level operations of DWT and SVD. Various attacks are added to the proposed method to justify the robustness of the watermark. In the end, the suggested approach is contrasted with existing methods to confirm the supremacy.

Component-based face recognition using statistical pattern matching analysis.

The aim of this research is to develop a fusion concept to component-based face recognition algorithms for features analysis of binary facial components (BFCs), which are invariant to illumination, expression, pose variations and partial occlusion. To analyze the features, using statistical pattern matching concepts, which are the combination of Chi-square (CSQ), Hu moment invariants (HuMIs), absolute difference probability of white pixels (AbsDifPWPs) and geometric distance values (GDVs) have been proposed for face recognition. The individual grayscale face image is cropped by applying the Viola-Jones face detection algorithm from a face database having variations in illumination, appearance, pose and partial occlusion with complex backgrounds. Doing illumination correction through histogram linearization technique, the grayscale face components such as eyes, nose and mouth regions are extracted using the 2D geometric positions. The binary face image is created by applying cumulative probability distribution function with Otsu adaptive thresholding method and then extracted BFCs such as eyes, nose and mouth regions. Five statistical pattern matching tools such as the standard deviation of CSQ values with probability of white pixels (PWPs), standard deviation of HuMIs with Hu's seven moment invariants, AbsDifPWPs and GDVs are developed for recognition purpose. GDVs are determined between two similar facial corner points (FCPs) and nine FCPs are extracted from binary whole face and BFCs. Pixel Intensity Values (PIVs) which are determined using L2 norms from grayscale values of the whole face and grayscale values of the face components. Experiment is performed using BioID Face Database on the basis of these pattern matching tools and appropriate threshold values with logical and conditional operators and gives the best expected results from true positive rate perspective.

Multiclass EEG signal classification utilizing Rényi min-entropy-based feature selection from wavelet packet transformation.

This paper proposes a novel feature selection method utilizing Rényi min-entropy-based algorithm for achieving a highly efficient brain-computer interface (BCI). Usually, wavelet packet transformation (WPT) is extensively used for feature extraction from electro-encephalogram (EEG) signals. For the case of multiple-class problem, classification accuracy solely depends on the effective feature selection from the WPT features. In conventional approaches, Shannon entropy and mutual information methods are often used to select the features. In this work, we have shown that our proposed Rényi min-entropy-based approach outperforms the conventional methods for multiple EEG signal classification. The dataset of BCI competition-IV (contains 4-class motor imagery EEG signal) is used for this experiment. The data are preprocessed and separated as the classes and used for the feature extraction using WPT. Then, for feature selection Shannon entropy, mutual information, and Rényi min-entropy methods are applied. With the selected features, four-class motor imagery EEG signals are classified using several machine learning algorithms. The results suggest that the proposed method is better than the conventional approaches for multiple-class BCI.

A field theoretical restoration method for images degraded by non-uniform light attenuation : an application for light microscopy.

Microscopy has become a de facto tool for biology. However, it suffers from a fundamental problem of poor contrast with increasing depth, as the illuminating light gets attenuated and scattered and hence can not penetrate through thick samples. The resulting decay of light intensity due to attenuation and scattering varies exponentially across the image. The classical space invariant deconvolution approaches alone are not suitable for the restoration of uneven illumination in microscopy images. In this paper, we present a novel physics-based field theoretical approach to solve the contrast degradation problem of light microscopy images. We have confirmed the effectiveness of our technique through simulations as well as through real field experimentations.

Modeling and classification of voluntary and imagery movements for brain-computer interface from fNIR and EEG signals through convolutional neural network.

Practical brain-computer interface (BCI) demands the learning-based adaptive model that can handle diverse problems. To implement a BCI, usually functional near-infrared spectroscopy (fNIR) is used for measuring functional changes in brain oxygenation and electroencephalography (EEG) for evaluating the neuronal electric potential regarding the psychophysiological activity. Since the fNIR modality has an issue of temporal resolution, fNIR alone is not enough to achieve satisfactory classification accuracy as multiple neural stimuli are produced by voluntary and imagery movements. This leads us to make a combination of fNIR and EEG with a view to developing a BCI model for the classification of the brain signals of the voluntary and imagery movements. This work proposes a novel approach to prepare functional neuroimages from the fNIR and EEG using eight different movement-related stimuli. The neuroimages are used to train a convolutional neural network (CNN) to formulate a predictive model for classifying the combined fNIR-EEG data. The results reveal that the combined fNIR-EEG modality approach along with a CNN provides improved classification accuracy compared to a single modality and conventional classifiers. So, the outcomes of the proposed research work will be very helpful in the implementation of the finer BCI system.

High-voltage electron microscopy tomography and structome analysis of unique spiral bacteria from the deep sea.

Structome analysis is a useful tool for identification of unknown microorganisms that cannot be cultured. In 2012, we discovered a unique deep-sea microorganism with a cell structure intermediate between those of prokaryotes and eukaryotes and described its features using freeze-substitution electron microscopy and structome analysis (quantitative and three-dimensional structural analysis of a whole cell at the electron microscopic level). We named it Myojin parakaryote Here we describe, using serial ultrathin sectioning and high-voltage electron microscopy tomography of freeze-substituted specimens, the structome analysis and 3D reconstruction of another unique spiral bacteria, found in the deep sea off the coast of Japan. The bacteria, which is named as 'Myojin spiral bacteria' after the discovery location and their morphology, had a total length of 1.768 ± 0.478 µm and a total diameter of 0.445 ± 0.050 µm, and showed either clockwise or counter-clockwise spiral. The cells had a cell surface membrane, thick fibrous layer, ribosomes and inner fibrous structures (most likely DNA). They had no flagella. The bacteria had 322 ± 119 ribosomes per cell. This ribosome number is only 1.2% of that of Escherichia coli and 19.3% of Mycobacterium tuberculosis and may reflect a very slow growth rate of this organism in the deep sea.