Hu Moments Research Articles

Performance Evaluation of Bagging Meta-Estimator in Lung Disease Detection: A Case Study on Imbalanced Dataset

In this study, titled "Performance Evaluation of Bagging Meta-Estimator in Lung Disease Detection: A Case Study on Imbalanced Dataset," we explore the effectiveness of the Bagging Meta-Estimator in diagnosing lung diseases, focusing on the challenges of imbalanced datasets. Utilizing a dataset segmented and characterized by Hu moments and encompassing categories of Normal, Bacterial Pneumonia, and Tuberculosis, the algorithm's performance was assessed through a 5-fold cross-validation. Results indicated moderate effectiveness with an average accuracy of 60.574%, precision of 60.749%, recall of 59.753%, and F1-Score of 59.416%, highlighting variable performance across folds. These findings suggest that while the Bagging Meta-Estimator has potential in medical imaging, further refinement is needed for consistent and reliable lung disease detection, especially in imbalanced datasets.

Performance Metrics of AdaBoost and Random Forest in Multi-Class Eye Disease Identification: An Imbalanced Dataset Approach

This study presents a comprehensive evaluation of AdaBoost and Random Forest Classifier algorithms in the classification of eye diseases, focusing on a challenging scenario involving an imbalanced dataset. Eye diseases, particularly Cataract, Diabetic Retinopathy, Glaucoma, and Normal eye conditions, pose significant diagnostic challenges, and the advent of machine learning offers promising avenues for enhancing diagnostic accuracy. Our research utilizes a dataset preprocessed with Canny edge detection for image segmentation and Hu Moments for feature extraction, providing a robust foundation for the comparative analysis. The performance of the algorithms is assessed using a 5-fold cross-validation approach, with accuracy, precision, recall, and F1-score as the key metrics. The results indicate that the Random Forest Classifier outperforms AdaBoost across these metrics, albeit with moderate overall performance. This finding underscores the potential and limitations of using advanced machine learning techniques for medical image analysis, particularly in the context of imbalanced datasets. The study contributes to the field by providing insights into the effectiveness of different machine learning algorithms in handling the complexities of medical image classification. For future research, it recommends exploring a diverse range of image processing techniques, delving into other sophisticated machine learning models, and extending the study to encompass a wider array of eye diseases. These findings have practical implications in guiding the selection of machine learning tools for medical diagnostics and highlight the need for continuous improvement in automated systems for enhanced patient care.

Optimizing Neurodegenerative Disease Classification with Canny Segmentation and Voting Classifier: An Imbalanced Dataset Study

This study explores the efficacy of a Voting Classifier, combining Logistic Regression, Random Forest, and Gaussian Naive Bayes, in the classification of neurodegenerative diseases, focusing on Alzheimer's Disease (AD), Parkinson’s Disease (PD), and control groups. Utilizing a dataset pre-processed with Canny segmentation and Hu Moments feature extraction, the research aimed to address the challenges posed by imbalanced datasets in medical image classification. The classifier's performance was evaluated through a 5-fold cross-validation approach, with metrics including accuracy, precision, recall, and F1-Score. The results revealed a consistent recall rate of approximately 46% across all folds, indicating the model's effectiveness in identifying cases of neurodegenerative diseases. However, the precision and F1-Score were notably lower, averaging around 22% and 29%, respectively, underscoring the difficulties in achieving accurate classification in imbalanced datasets. The study contributes to the understanding of machine learning applications in medical diagnostics, specifically in the challenging context of neurodegenerative disease classification. It highlights the potential of using advanced image processing techniques combined with machine learning ensembles in enhancing diagnostic accuracy. However, it also draws attention to the inherent challenges in such approaches, particularly regarding precision in imbalanced datasets. Recommendations for future research include exploring data balancing techniques, alternative feature extraction methods, and different machine learning algorithms to improve the precision and overall performance. Additionally, applying the model to a broader and more diverse dataset could provide more generalizable and robust findings. This study is significant for researchers and practitioners in medical imaging and machine learning, offering insights into the complexities and potential of automated disease classification

Assessing Bagging-meta Estimator in Imbalanced CT Kidney Disease Classification: A Focus on Sobel and Hu Moment Techniques

This study investigates the efficacy of the Bagging-meta estimator in classifying CT kidney diseases, focusing on an imbalanced dataset processed through Sobel segmentation and Hu moment feature extraction. The research utilized a quantitative approach, applying the Bagging-meta estimator to a dataset comprising CT images classified into four categories: Normal, Cyst, Tumor, and Stone. These images were preprocessed using Sobel segmentation to highlight critical structures and Hu moment feature extraction for robust classification features. The study employed a 5-fold cross-validation method to evaluate the model's performance, assessing metrics such as accuracy, precision, recall, and F1-Score. The results indicated a significant variation in the model's performance across different folds, with accuracy ranging from 49.86% to 66.17%, precision between 51.86% and 65.93%, recall from 57.95% to 64.44%, and F1-Scores spanning 48.26% to 60.74%. These findings suggest that while the Bagging-meta estimator can achieve reasonable accuracy in classifying kidney diseases from CT images, its performance is affected by the imbalanced nature of the dataset. This study contributes to the understanding of the challenges and potential of machine learning in medical imaging, particularly in the context of imbalanced datasets. It highlights the need for specialized approaches to handle such datasets and underscores the importance of preprocessing techniques in enhancing model performance. Future research directions include exploring methods to address data imbalance, investigating alternative feature extraction techniques, and testing the model on diverse datasets to enhance its generalizability and reliability in clinical settings. This research offers valuable insights into the development of automated diagnostic tools in medical imaging and advances the field of computer-aided diagnosis in nephrology.

Evaluating Thresholding-Based Segmentation and Humoment Feature Extraction in Acute Lymphoblastic Leukemia Classification using Gaussian Naive Bayes

This study, titled "Evaluating Thresholding-Based Segmentation and HuMoment Feature Extraction in Acute Lymphoblastic Leukemia Classification using Gaussian Naive Bayes," investigates the application of image processing and machine learning techniques in the classification of Acute Lymphoblastic Leukemia (ALL). Utilizing a dataset of microscopic blood smear images, the research focuses on the efficacy of thresholding-based segmentation and Hu moment feature extraction in distinguishing between benign and malignant cases of ALL. Gaussian Naive Bayes, known for its simplicity and effectiveness, is employed as the classification algorithm. The study adopts a 5-fold cross-validation approach to evaluate the model's performance, with particular emphasis on metrics such as accuracy, precision, recall, and F1-score. Results indicate a high precision rate across all folds, averaging approximately 84.13%, while exhibiting variability in accuracy, recall, and F1-scores. These findings suggest that while the model is effective in identifying malignant cases, further refinements are necessary for improving overall accuracy and consistency. This research contributes to the field of medical image analysis by demonstrating the potential of combining simple yet efficient techniques for the automated diagnosis of hepatological diseases. It highlights the importance of integrating image processing with machine learning to enhance diagnostic accuracy in medical applications.

Automatic Weight of Color, Texture, and Shape Features in Content-Based Image Retrieval Using Artificial Neural Network

Image retrieval is the process of finding images in the database that are similar to the query image by measuring how close the feature values of the query image are to other images. Image retrieval is currently dominated by approaches that combine several different representations or features. The optimal weight of each feature is needed in combining the image features such as color features, texture features, and shape features. In this study, we use a multi-layer perceptron artificial neural network (MLP) method to obtain feature weights automatically and simultaneously look for optimal weights. The color moment is used to find nine color features, Gray Level Co-occurrence Matrix (GLCM) to find four texture features, and Hu Moment to find seven shape features totaling 20 neurons and all of these features become the input layer in our MLP network. Three neurons in output layers become the automatic weight of each feature. These weights are used to combine each feature's role in obtaining the relevant image. Euclidean Distance is used to measure similarity. The average precision values obtained using automatic feature weights are 93.94% for the synthetic dataset, 91.19% for the Coil-100 dataset, and 54.31% for the Wang dataset. These results have an average difference of 5.06% with the target so automatic feature weighting works well. This value is obtained at a hidden layer size of 11 and a learning rate of 0.1. In addition, the use of automatic feature weighting gives more accurate results compared to manual feature weighting.

Seismic Image Identification and Detection Based on Tchebichef Moment Invariant

The research focuses on the analysis of seismic data, specifically targeting the detection, edge segmentation, and classification of seismic images. These processes are fundamental in image processing and are crucial in understanding the stratigraphic structure and identifying oil and natural gas resources. However, there is a lack of sufficient resources in the field of seismic image detection, and interpreting 2D seismic image slices based on 3D seismic data sets can be challenging. In this research, image segmentation involves image preprocessing and the use of a U-net network. Preprocessing techniques, such as Gaussian filter and anisotropic diffusion, are employed to reduce blur and noise in seismic images. The U-net network, based on the Canny descriptor is used for segmentation. For image classification, the ResNet-50 and Inception-v3 models are applied to classify different types of seismic images. In image detection, Tchebichef invariants are computed using the Tchebichef polynomials’ recurrence relation. These invariants are then used in an optimized multi-class SVM network for detecting and classifying various types of seismic images. The promising results of the SVM model based on Tchebichef invariants suggest its potential to replace Hu moment invariants (HMIs) and Zernike moment invariants (ZMIs) for seismic image detection. This approach offers a more efficient and dependable solution for seismic image analysis in the future.

A Comparative of Two-Dimensional Statistical Moment Invariants Features in Formulating an Automated Probabilistic Machine Learning Identification Algorithm for Forensic Application

IBIS, ALIS, EVOFINDER, and CONDOR are the massive ballistics computerised technological machines that have typically been utilised in forensic laboratories to automatically locate similarities between images of cartridge cases and bullets. However, it imposed a long execution time and requires physical interpretation to consolidate the analysis results when employing these market-available technologies to accomplish ballistics matching tasks. Therefore, the principal objective of this study is to propose an improvised automated probabilistic machine learning identification algorithm by extracting the two-dimensional (2D) statistical moment invariants from the segmented region of interest (ROI) corresponding to the cartridge case and bullets images. To pursue this principal objective, several 2D statistical moment invariants have been compared and tested to determine the most suitable feature set applied in the proposed identification algorithm. The 2D statistical moment invariants employed include Orthogonal Legendre moments (OLM), Hu moments (HM), Tsirikolias-Mertzois moments (TMM), Pan-Keane moments (PKM), and Central Geometric moments (CGM). Moreover, the proposed identification algorithm is also tested in different scenarios, including based on the classification of strength association measurements between the extracted feature sets. The empirical results in this article revealed that the proposed identification algorithm applied with the CGM comprising the weak association classification yielded the best identification accuracy rates, which are >96.5% across all the sample sizes of the training set. These empirical results also conveyed that the superior proposed identification algorithm in this research could be developed as a mobile application for ballistics identification that can significantly reduce the time taken and conveniently perform the ballistics identification tasks.

Research on an Underwater Target-Tracking Method Based on Zernike Moment Feature Matching

Sonar images have the characteristics of lower resolution and blurrier edges compared to optical images, which make the feature-matching method in underwater target tracking less robust. To solve this problem, we propose a particle filter (PF)-based underwater target-tracking method utilizing Zernike moment feature matching. Zernike moments are used to construct the feature-description vector for feature matching and contribute to the update of particle weights. In addition, the particle state transition method is optimized by using a first-order autoregressive model. In this paper, we compare Hu moments and Zernike moments, and we also compare whether to optimize the particle state transition on the tracking results or not based on the effects of each option. The experimental results based on the AUV (autonomous underwater vehicle) prove that the robustness and accuracy of this innovative method is better than the other combined methods mentioned in this paper.

Multi-descriptor Enabled Leaf Disease Detection using Machine Learning Methods

India is an agricultural country. Two-thirds of the Indian populations work in agriculture, making it the backbone of the country's economic system. Leaf disease is a major challenge in agriculture that affects crop yield and quality. In recent years, advancements methods for accurately and effectively detecting leaf diseases have been developed thanks to advances in machine learning and computer vision. The objective of this work is to create a machine learning-based Random Forest Regression Algorithm for automatically detecting leaf diseases. In the suggested methodology, leaf pictures are acquired, subjected to pre-processing which includes RGB image and HSV image conversion, multi-descriptor feature extraction includes Hu moments, Haralick, color histogram and classification using Random Forest Regression Algorithm. Metrics including accuracy, precision, recall, and f1-score are used to assess the system's performance by implementing in Python 3.8. The findings of the study indicate that the proposed system achieves precision 98 %, recall 98 %, f1-score 98%, test accuracy 97.81 %, while validation accuracy is 95.93 %. The developed system has potential applications in precision agriculture, enabling farmers to detect and treat plant diseases early, thereby reducing crop losses and increasing yields.

ACO-KELM: Anti Coronavirus Optimized Kernel-based Softplus Extreme Learning Machine for classification of skin cancer

Due to the presence of redundant and irrelevant features in large-dimensional biomedical datasets, the prediction accuracy of disease diagnosis can often be decreased. Therefore, it is important to adopt feature extraction methodologies that can deal with problem structures and identify underlying data patterns. In this paper, we propose a novel approach called the Anti Coronavirus Optimized Kernel-based Softplus Extreme Learning Machine (ACO-KSELM) to accurately predict different types of skin cancer by analyzing high-dimensional datasets. To evaluate the proposed ACO-KSELM method, we used four different skin cancer image datasets: ISIC 2016, ACS, HAM10000, and PAD-UFES-20. These dermoscopic image datasets were preprocessed using Gaussian filters to remove noise and artifacts, and relevant features based on color, texture, and shape were extracted using color histogram, Haralick texture, and Hu moment extraction approaches, respectively. Finally, the proposed ACO-KSELM method accurately predicted and classified the extracted features into Basal Cell Carcinoma (BCC), Squamous Cell Carcinoma (SCC), Actinic Keratosis (ACK), Seborrheic Keratosis (SEK), Bowen’s disease (BOD), Melanoma (MEL), and Nevus (NEV) categories. The analytical results showed that the proposed method achieved a higher rate of prediction accuracy of about 98.9%, 98.7%, 98.6%, and 97.9% for the ISIC 2016, ACS, HAM10000, and PAD-UFES-20 datasets, respectively.

Multiclass Classification of Paddy Leaf Diseases Using Random Forest Classifier

With increase in population, improving the quality and quantity of food is essential. Paddy is a vital food crop serving numerous people in various continents of the world. The yield of paddy is affected by numerous factors. Early diagnosis of disease is needed to prevent the plants from successive stage of disease. Manual diagnosis by naked eye is the traditional method widely adopted by farmers to identify leaf diseases. However, when the task involves manual disease diagnosis, problems like the hiring of domain experts, time consumption, and inaccurate results will arise. Inconsistent results may lead to improper treatment of plants. To overcome this problem, automatic disease diagnosis is proposed by researchers. This will help the farmers to accurately diagnose the disease swiftly without the need for expert. This manuscript develops model to classify four types of paddy leaf diseases bacterial blight, blast, tungro and brown spot. To begin with, the image is preprocessed by resizing and conversion to RGB Red, Green and Blue (RGB) and Hue, Saturation and Value (HSV) color space. Segmentation is done. Global features namely: hu moments, Haralick and color histogram are extracted and concatenated. Data is split in to training part and testing part in 70:30 ratios. Images are trained using multiple classifiers like Logistic Regression, Random Forest Classifier, Decision Tree Classifier, K-Nearest Neighbor (KNN) Classifier, Linear Discriminant Analysis (LDA),Support Vector Machine (SVM) and Gaussian Naive Bayes. This study reports Random Forest classifier as the best classifier. The Accuracy of the proposed model gained 92.84% after validation and 97.62% after testing using paddy disordered samples. 10 fold cross validation is performed. Performance of classification algorithms is measured using confusion matrix with precision, recall, F1- score and support as parameters.

Application of the K-Nearest Neighbors (KNN) Algorithm on the Brain Tumor Dataset

Brain tumors pose significant challenges in the medical domain, necessitating advanced diagnostic techniques for early and accurate detection. This research paper presents a comprehensive study on the application of the K-Nearest Neighbors (KNN) algorithm to a dataset comprising brain tumor images. The methodology involved segmenting the images using the Canny method, extracting relevant features via Hu Moments, and subsequently employing the KNN algorithm for classification. Using a 5-fold cross-validation, the system consistently achieved an average accuracy of approximately 62%. These findings highlight the potential of traditional machine learning algorithms in medical imaging, providing valuable insights for both researchers and practitioners. While the results are promising, the study also underscores the importance of integrating such algorithms with other diagnostic methods for optimal results

Effectiveness Evaluation of the RandomForest Algorithm in Classifying CancerLips Data

Lip cancer, though less commonly discussed, remains a significant concern in the realm of oncology. Early detection and diagnosis are paramount for improved patient outcomes. This research evaluated the effectiveness of the RandomForest algorithm in classifying the CancerLips dataset, a collection of lip images processed using the Canny segmentation method and described using Hu moments. Using a 5-fold cross-validation approach, the algorithm achieved an average accuracy of approximately 70.96%. The results highlight the potential of machine learning techniques, specifically RandomForest, in aiding lip cancer detection. However, the choice of preprocessing methods and feature extraction plays a crucial role in determining the outcome. The study underscores the need for further research, focusing on algorithm optimization and comparisons with other datasets or feature extraction methods, to enhance diagnostic precision in medical imaging.

Enhancing Gastrointestinal Disease Diagnosis with KNN: A Study on WCE Image Classification

This study explores the application of the K-Nearest Neighbors (KNN) algorithm, following Sobel segmentation and Hu Moment feature extraction, to classify Wireless Capsule Endoscopy (WCE) images into Normal and Ulcerative Colitis conditions. Through a rigorous 5-fold cross-validation approach, the research aimed to determine the KNN algorithm's accuracy, precision, recall, and F1-score on the WCE Curated Colon Disease Dataset. The findings revealed high performance across all metrics, with accuracy rates extending up to 90.625%. The confusion matrix provided further validation, illustrating a high true positive rate coupled with a low false negative rate. These results substantiate the hypothesis that employing edge detection and shape descriptors as pre-processing techniques can significantly enhance the efficacy of machine learning algorithms in medical image classification. The study’s contribution is twofold: it reaffirms the potential of machine learning in the advancement of medical diagnostics and provides a methodological framework for automated image classification that can assist clinicians. It is recommended that future research extends to broader datasets and explores various algorithms to enhance diagnostic precision. In practice, integrating this research into a clinical decision support system could revolutionize diagnostic processes, offering a non-invasive, accurate, and efficient tool for gastroenterological diagnostics.

Classification Optimization of Skin Cancer Using the Adaboost Algorithm

Early detection of melanoma skin cancer is crucial in improving prognosis and saving lives. This research aimed to optimize the classification of melanoma images using the Adaboost algorithm. Employing a dataset of 10,000 melanoma images, the study combined the Canny method for image segmentation, Hu Moments for feature extraction, and the Adaboost algorithm for classification. The 5-fold cross-validation results revealed an average accuracy of 61.52%. While the precision consistently surpassed recall, indicating the model's conservative nature in predicting positive cases. The outcomes align with previous research, emphasizing the challenges in melanoma classification. This study contributes to the domain by showcasing the potential and areas of improvement for machine learning in early melanoma detection. Future research is recommended to explore hybrid models and diversify data sources for enhanced robustness and generalizability.

Image Retrieval based on the Fusion of Graph Method with Color Moments, GLCM, and Hu Moments

Retrieving images that are similar to the query image in the image database means determining the similarity between the images. This study aims to use a graph method with region adjacency graph representation in conjunction with a non-graph method in image retrieval. We represented an image as a graph and used the Graph Edit Distance (GED) method to calculate the similarity between two graphs. The feature extraction of the image graph, which exposes the content and the relationships between existing content, is a key step in image retrieval based on the graph method. The extraction of graph features is accomplished by the image segmentation method, which divides the image into regions and represents them as a region-adjacency graph (RAG), in which vertices represent regions and edges indicate two neighboring regions. Image retrieval based on the graph method is combined with low-level approaches like Color Moments, Gray Level Co-occurrence Matrix (GLCM), and Hu Moments to boost accuracy. All obtained features are normalized, weighted, and then compared between images to get the similarity value using Euclidean Distance. An image retrieval prototype based on the combined graph method and non-graph method was successfully created in this work, using four datasets: synthetic, batik, COIL-100, and Wang. The MAP of the four datasets is 67.84 percent, but when combined with the low-level feature approach, it rises to between 79.73 and 89.71 percent. The combination of graph and non-graph algorithms improves image retrieval outcomes.

Micro-Scale Surface Recognition via Microscope System Based on Hu Moments Pattern and Micro Laser Line Projection

The surface engineering of metals develops high technology to detect microscale convex, concave and flat surface patterns. It is because the manufacturing industry requires technologies to recognize microscale surface features. Thus, it is necessary to develop microscopic vision technology to recognize microscale concave, convex and flat surfaces. This study addresses microscale concave, convex and flat surface recognition via Hu moments’ patterns based on micro-laser line contouring. In this recognition, a Hu moments’ pattern is generated from a Bezier model to characterize the surface recovered through microscopic scanning. The Bezier model is accomplished by employing a genetic algorithm and surface coordinates. Thus, the flat, convex and concave surfaces are recognized based on the Hu moments’ pattern of each one. The microscope system projects a 40 μm laser line on the object and a camera acquires the object’s contour reflection to retrieve topographic coordinates. The proposed technique enhances the microscale convex, concave, flat, and surface recognition, which is performed via optical microscope systems. The contribution of microscopic shape recognition based on the Hu moments’ pattern and microscopic laser line is elucidated by a discussion based on the microscopic shape recognition performed through the optical microscopic image processing.

Feature aggregation for nutrient deficiency identification in chili based on machine learning

Macronutrient deficiency inhibits the growth and development of chili plants. One of the non-destructive methods that plays a role in processing plant image data based on specific characteristics is computer vision. This study uses 5166 image data after augmentation process for six plant health conditions. But the analysis of one feature cannot represent plant health condition. Therefore, a careful combination of features is required. This study combines three types of features with HSV and RGB for color, GLCM and LBP for texture, and Hu moments and centroid distance for shapes. Each feature and its combination are trained and tested using the same MLP architecture. The combination of RGB, GLCM, Hu moments, and Distance of centroid features results the best performance. In addition, this study compares the MLP architecture used with previous studies such as SVM, Random Forest Technique, Naive Bayes, and CNN. CNN produced the best performance, followed by SVM and MLP, with accuracy reaching 97.76%, 90.55% and 89.70%, respectively. Although MLP has lower accuracy than CNN, the model for identifying plant health conditions has a reasonably good success rate to be applied in a simple agricultural environment.

Grading the Severity of Damage to the Perijunctional Actomyosin Ring and Zonula Occludens-1 of the Corneal Endothelium by Ensemble Learning Methods.

Purpose: In many epithelia, including the corneal endothelium, intracellular/extracellular stresses break down the perijunctional actomyosin ring (PAMR) and zonula occludens-1 (ZO-1) at the apical junctions. This study aims to grade the severity of damage to PAMR and ZO-1 through machine learning. Methods: Immunocytochemical images of PAMR and ZO-1 were drawn from recent studies on the corneal endothelium subjected to hypothermia and oxidative stress. The images were analyzed for their morphological (e.g., Hu moments) and textural features (based on gray-level co-occurrence matrix [GLCM] and Gabor filters). The extracted features were ranked by SHapley analysis and analysis of variance. Then top features were used to grade the severity of damage using a suite of ensemble classifiers, including random forest, bagging classifier (BC), AdaBoost, extreme gradient boosting, and stacking classifier. Results: A partial set of features from GLCM, along with Hu moments and the number of hexagons, enabled the classification of damage to PAMR into Control, Mild, Moderate, and Severe with the area under the receiver operating characteristics curve (AUC) = 0.92 and F1 score = 0.77 with BC. In contrast, a bank of Gabor filters provided a partial set of features that could be combined with Hu moments, branch length, and sharpness for the classification of ZO-1 images into four levels with AUC = 0.95 and F1 score of 0.8 with BC. Conclusions: We have developed a workflow that enables the stratification of damage to PAMR and ZO-1. The approach can be applied to similar data during drug discovery or pathophysiological studies of epithelia.