Hu Invariant Moments Research Articles

A comprehensive approach with DTW-driven IMF selection, multi-domain fusion, and TSA-based feature selection for compound fault diagnosis

Fault diagnosis in industrial machinery ensures operational efficiency and prevents downtimes. However, compound faults present a significant challenge due to their complex nature. This manuscript proposes a novel methodology for identifying compound faults, integrating advanced techniques to address the challenges present in existing methods. Firstly, the ICEEMDAN algorithm is employed for signal decomposition, enabling the extraction of noise-free IMFs crucial for identifying fault signatures. A DTW-based selection criteria is introduced to enhance the accuracy of IMF selection. The Bessel Transform is utilized for time–frequency conversion. HU invariant moments are incorporated as image features, and multi-domain feature fusion is implemented to understand compound faults comprehensively. A novel TSA-based feature selection method is proposed to select optimal features from the fused feature set. The methodology’s effectiveness is evaluated through a case study of compound gear-bearing faults. The results demonstrate the superiority of the approach, with a testing accuracy of 99 %.

Image-free Hu invariant moment measurement by single-pixel detection

The global feature invariant moment plays a crucial role in capturing the feature information of a target. At present, the extraction of the target moment depends on the processing of the two-dimensional image, which leads to the high computational resource occupation due to the large amount of inherent data and the high information redundancy of the image. This reduces the real-time performance of target moment extraction to a certain extent. In order to solve this problem, we introduce a novel strategy for extracting Hu invariant moments, which is based on the combination of single-pixel detection technology and the Hu invariant moments theory. To the best of our knowledge, this is the first proposal of an image-free extraction strategy employing single-pixel detection. The extraction of moments is seamlessly integrated into the single-pixel detection process, facilitating efficient and rapid acquisition, followed by the construction of Hu invariant moments. Theoretical analysis and experimental validation of our proposed method are carried out, demonstrating the effective acquisition of Hu invariant moments. This technology holds substantial application value and prospects in various fields, including target analysis, understanding, classification, and recognition.

Segmentation-free measurement of locomotor frequency in Caenorhabditis elegans using image invariants.

An animal's locomotor rate is an important indicator of its motility. In studies of the nematode Caenorhabditis elegans (C. elegans), assays of the frequency of body bending waves have often been used to discern the effects of mutations, drugs, or aging. Traditional manual methods for measuring locomotor frequency are low in throughput and subject to human error. Most current automated methods depend on image segmentation, which requires high image quality and is prone to errors. Here, we describe an algorithm for automated estimation of C. elegans locomotor frequency using image invariants, i.e. shape-based parameters that are independent of object translation, rotation, and scaling. For each video frame, the method calculates a combination of 8 Hu's moment invariants and a set of maximally stable extremal regions (MSER) invariants. The algorithm then calculates the locomotor frequency by computing the autocorrelation of the time sequence of the invariant ensemble. Results of our method show excellent agreement with manual or segmentation-based results over a wide range of frequencies. We show that compared to a segmentation-based method that analyzes a worm's shape and a method based on video covariance, our technique is more robust to low image quality and background noise. We demonstrate the system's capabilities by testing the effects of serotonin and serotonin pathway mutations on C. elegans locomotor frequency.

Research on Digital Construction and Design of Minority Clothing Based on Multivariate Statistical Analysis

Abstract Digitizing minority costumes is a modern way to protect non-heritage. In this paper, after analyzing the Hu invariant moment features and HOG features, we propose a feature extraction method for ethnic minority costumes that integrates the fusion features of HOG features and Hu invariant moment features. On this basis, the structure of the extracted multivariate feature data is analyzed, and the KPCA method of multivariate statistical analysis is adopted to process the multivariate data of ethnic minority costumes and digitally construct and design the ethnic minority costumes. In addition, experiments were conducted to explore the effect of minority costumes on feature extraction and the effect of digitization construction and design. The results show that the extraction accuracy of six ethnic costumes is stable at 0.75, 0.45, 0.50, 0.75, 0.72, and 0.68, respectively, and its Top-10 accuracy reaches 0.61 at the highest level, while the accuracy of retrieving only some of them is less than 0.40, and the effects of constructed ethnic costume outlines are all around 0.9. The output results of ethnic dresses constructed based on this research are roughly similar to the original ethnic dresses, which is conducive to the inheritance of ethnic dress culture.

Methodology based on the Hu moment invariants for object comparison on radar satellite imagery

In this paper we present a methodology for comparing objects, identified on satellite images acquired by Synthetic Aperture Radar (SAR). The methodology describes object transformations mathematically and provides a quantitative estimate of the similarity of different shapes. To demonstrate its capabilities we examine a case study – we detect an underwater structure in the Western Black Sea, which becomes visible on radar imagery under favourable sea conditions. Our aim is to quantitatively describe if it changes over time and to determine on which dates the detections reveal the greatest similarity.The suggested methodology is based on computation of the Hu moment invariants, or the image moments. They give a set of 7 invariants, which describe object transformations, such as translation, scale, rotation and reflection. The presented methodology can be applied in different contexts on satellite imagery for quantitative description of structure transformations.

Collar recognition and matching of clothing style drawings based on complex networks

Seeking a fast and effective algorithm for describing and classifying clothing styles is currently a hot topic in the textile and clothing industry. The study first constructs a sample library of collar styles, preprocesses clothing images, and uses complex network models to describe and extract features; Implement clothing image classification based on support vector machines, and finally use image matching algorithms to match the image to be matched with the target object. It was indicated that the complex network model established by the research institute can effectively depict the characteristics of various collar styles, and the complex network model can effectively depict the characteristics of various collar styles. In 10 experiments, the overall average overall accuracy was 98%; Regarding different types of collars, the average accuracy of 10 experiments is above 96%, indicating that the collar classification accuracy based on support vector machine method is very high. The classifying results of the 10 categories obtained by the research method have little change and are relatively stable. For each category, the research method is higher than the other two methods. The results can indicate that the research method has effect and is more suitable for collar recognition and classification. Compared with traditional Hu invariant moments and HOG feature extraction algorithms, the research algorithm has stronger noise suppression ability, and using Euclidean distance as a similarity measure has great advantages.

Research on the Shape Classification Method of Rural Homesteads Based on Parcel Scale—Taking Yangdun Village as an Example

The basic information survey on homesteads requires understanding the shape of homesteads, and the shape of the homesteads based on the spatial location can reflect information such as their outline and regularity, but the current shape classification of rural homesteads at the parcel scale lacks analytical methods. In this study, we endeavor to explore a classification model suitable for characterizing homestead shapes at the parcel scale by assessing the impact of various research methods. Additionally, we aim to uncover the evolutionary patterns in homestead shapes. The study focuses on Yangdun Village, located in Deqing County, Zhejiang Province, as the research area. The data utilized comprise Google Earth satellite imagery and a vector layer representing homesteads at the parcel scale. To classify the shapes of homesteads and compare classification accuracy, we employ a combination of methods, including the fast Fourier transform (FFT), Hu invariant moments (HIM), the Boyce and Clark shape index (BCSI), and the AlexNet model. Our findings reveal the following: (1) The random forest method, when coupled with FFT, demonstrates the highest effectiveness in identifying the shape categories of homesteads, achieving an average accuracy rate of 88.6%. (2) Combining multiple methods does not enhance recognition accuracy; for instance, the accuracy of the FFT + HIM combination was 88.4%. (3) The Boyce and Clark shape index (BCSI) proves unsuitable for classifying homestead shapes, yielding an average accuracy rate of only 58%. Furthermore, there is no precise numerical correlation between the homestead category and the shape index. (4) It is noteworthy that over half of the homesteads in Yangdun Village exhibit rectangular-like shapes. Following the “homesteads reform”, square-like homesteads have experienced significant vacating, resulting in a mixed arrangement of homesteads overall. The research findings can serve as a methodological reference for the investigation of rural homestead shapes. Proficiency in homestead shape classification holds significant importance in the realms of information investigation, regular management, and layout optimization of rural land.

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.

Typical Fault Modeling and Vibration Characteristics of the Turbocharger Rotor System

To study the typical failure mechanism (rotor unbalance, rotor friction, and rotor crack) and vibration characteristics of the turbocharger rotor system, a rotor system dynamics simulation model was established by an improved four-node aggregate parameter method. The geometric and physical characteristics of the rotor system under three failure states and its dynamics under operation were analyzed. Thus, a typical failure dynamics simulation model of the rotor system was established. On this basis, the output failure simulation signal was extracted using the Hu invariant moment feature extraction method to analyze the system vibration characteristics under each typical failure state of the rotor system. The results show that the model in this paper can effectively reduce the computational volume and computational time, and the errors of numerical simulation were less than 3%. When an unbalance fault occurred in the rotor system, the shaft trajectory was “0” shaped and the response spectrum was dominated by 1X. When the rotor system was frictional, the shaft trajectory was a slightly concave “8” shape, and the response spectrum was dominated by 0.5X. When the rotor system was cracked, the axial trajectory was a “vortex”, and the response spectrum was dominated by 0.5X. Thus, the study of typical failure mechanism and vibration characteristics of a turbocharger rotor system by simulation calculation is effective and has good research prospects, providing an important technical reference for dynamic analysis and fault diagnosis of the rotor system.

Computer-Aided Ankle Ligament Injury Diagnosis from Magnetic Resonance Images Using Machine Learning Techniques.

Ankle injuries caused by the Anterior Talofibular Ligament (ATFL) are the most common type of injury. Thus, finding new ways to analyze these injuries through novel technologies is critical for assisting medical diagnosis and, as a result, reducing the subjectivity of this process. As a result, the purpose of this study is to compare the ability of specialists to diagnose lateral tibial tuberosity advancement (LTTA) injury using computer vision analysis on magnetic resonance imaging (MRI). The experiments were carried out on a database obtained from the Vue PACS-Carestream software, which contained 132 images of ATFL and normal (healthy) ankles. Because there were only a few images, image augmentation techniques was used to increase the number of images in the database. Following that, various feature extraction algorithms (GLCM, LBP, and HU invariant moments) and classifiers such as Multi-Layer Perceptron (MLP), Support Vector Machine (SVM), k-Nearest Neighbors (kNN), and Random Forest (RF) were used. Based on the results from this analysis, for cases that lack clear morphologies, the method delivers a hit rate of 85.03% with an increase of 22% over the human expert-based analysis.

Research on fault diagnosis method of turbocharger rotor based on Hu-SVM-RFE

Abstract Several parameters need to be monitored for turbocharger rotor faults and the overlap between different fault parameters as well as the redundancy of data, which leads to increased calculation time and reduced classification accuracy. To improve the recognition rate of turbocharger rotor faults, a recursive elimination method based on the support vector machine-recursive feature elimination (SVM-RFE) combined with improved Hu invariant moments is developed for the axial orbit feature extraction of turbocharger rotor with rotor fault. Firstly, improved Hu-invariant moments are extracted for different rotor fault axis orbits, and then the feature ranking and selection are performed by the SVM-RFE method to filter out the feature combinations with higher classification recognition rates. Then, the feature matrix of the Hu-SVM-RFE algorithm screening combination was identified for classification using each of the three diagnostic algorithms. The results show that the optimal feature subset obtained by the Hu-SVM-RFE method can ensure the richness of the fault information of the turbocharger rotor with small number of features. And, a high classification rate can be obtained with low time consumption in combination with the probabilistic neural network (PNN) algorithm. Therefore, Hu-SVM-RFE feature screening method combined with PNN fault diagnosis technology has high accuracy and efficiency, which is of great significance for online fault identification of the supercharger rotor.

In Situ Defect Detection in Selective Laser Melting using a Multi-feature Fusion Method

Defects in each layer during selective laser melting (SLM) process are closely associated with the final forming quality and performance of the part. The in-situ monitoring of the layer-wise images is a reliable method for controlling and retracing the quality during the SLM process. In order to improve the accuracy and efficiency of defect detection in the SLM process as well as investigate the relations between the structures of thin-wall parts and defects, a feature fusion-based method is proposed for detecting geometric deformation, debris and local bulge defects on the forming layer. Edge features, texture features and geometric features in the image of forming layer are extracted using histogram of oriented gradients (HOG), gray level co-occurrence matrix (GLCM) and Hu invariant moments. These three features are weighted and formed a feature vector as the inputs of support vector machine (SVM) algorithm to achieve classifications of defects. The performance of different feature fused strategies and classifiers to characterize defects shows that the proposed feature fusion method combined with SVM classifier is able to achieve an average accuracy of 97.11% and an average F1score of 96.91%. The forming layer has severe defects of geometric deformation, debris or local bulge when the thickness of the thin-wall part reaches its limited 0.5mm.

Fabric Image Retrieval Algorithm based on Fractal Coding and Zernike Moment under Wavelet Transform

Abstract: To help the workers retrieve the same or similar in textile companies, a fabric image retrieval algorithm based on fractal coding and Zernike moment under wavelet transform is proposed, which can quickly and accurately retrieve images similar to the fabric image to be retrieved from the fabric image database. Firstly, the low-frequency component is obtained by wavelet transform, and the transformed low-frequency sub-image is fractal encoded to obtain the coding parameters. Then, the Zernike moment of the low-frequency sub-image is calculated. The fractal coding parameters and Zernike moment under wavelet transform are combined as the fabric image retrieval characteristic. The algorithm overcomes the problems of low retrieval accuracy and time-consuming of direct feature extraction under single feature. Compared with the basic fractal algorithm (BFIC), the joint orthogonal fractal parameters with the improved Hu invariant moment algorithm (HVKF) and the sparse fractal image compression

Human Behavior Recognition of Video Surveillance System Based on Neural Network

Video surveillance system is widely used in modern society. Human behavior recognition based on neural network uses computer vision and image processing technology to analyze and judge the behavior of different objects in surveillance video. In order to improve the recognition accuracy of human behavior, this paper makes an in-depth study on the design of video surveillance system and target tracking using the relevant algorithms of neural network. In this paper, the application of relevant neural network algorithms and the technology of character behavior recognition are analyzed by means of experimental method comparison and algorithm comparison. The experimental data shows that in the detection of abnormal behavior, the highest accuracy rate of human behavior recognition is 90% by using the improved hu invariant moment model.

Image Perceptual Hashing for Content Authentication Based on Geometric Invariant Vector Distance

Image perceptual hashing is broadly applied in image content authentication, recognition, retrieval, and social media hotspot event detection. An image authentication algorithm is put forward based on the Itti visual saliency model and geometric invariant vector distance. To begin with, the image is preprocessed and weighted by the Itti model and contourlet transform. After that, the weighted image is randomly divided into blocks, and the image feature vector is constructed by calculating the geometric invariant vector distance on both Hu invariant moment vector and maximum singular value vector of the random blocks. In the end, the feature vector is quantized and encrypted to generate the ultimate hash. Experimental results illustrate that when the threshold T = 70, the true positive rate P T P R for duplicate images stands at 0.96574, while the false rate P F P R of different images is merely 0.0224, with the total error rate reaching the minimum value (0.0566). Furthermore, the AUC value of the proposed algorithm is 0.9951, which is higher than that of the comparison algorithms, indicating that the algorithm has better performance than other state-of-the-art algorithms in terms of various visual content-preserving attacks.

The Comparison of SVM and ANN Classifier for COVID-19 Prediction

Coronavirus 2 (SARS-CoV-2) is the cause of an acute respiratory infectious disease that can cause death, popularly known as Covid-19. Several methods have been used to detect COVID-19-positive patients, such as rapid antigen and PCR. Another method as an alternative to confirming a positive patient for COVID-19 is through a lung examination using a chest X-ray image. Our previous research used the ANN method to distinguish COVID-19 suspect, pneumonia, or expected by using a Haar filter on Discrete Wavelet Transform (DWT) combined with seven Hu Moment Invariants. This work adopted the ANN method's feature sets for the Support Vector Machine (SVM), which aim to find the best SVM model appropriate for DWT and Hu moment-based features. Both approaches demonstrate promising results, but the SVM approach has slightly better results. The SVM's performances improve accuracy to 87.84% compared to the ANN approach with 86% accuracy.
 

Designs of human–robot interaction using depth sensor-based hand gesture communication for smart material-handling robot operations

With rapid developments in biometric recognition, a great deal of attention is being paid to robots which interact smartly with humans and communicate certain types of biometrical information. Such human–machine interaction (HMI), also well-known as human–robot interaction (HRI), will, in the future, prove an important development when it comes to automotive manufacturing applications. Currently, hand gesture recognition-based HRI designs are being practically used in various areas of automotive manufacturing, assembly lines, supply chains, and collaborative inspection. However, very few studies are focused on material-handling robot interactions combined with hand gesture communication of the operator. The current work develops a depth sensor-based dynamic hand gesture recognition scheme for continuous-time operations with material-handling robots. The proposed approach properly employs the Kinect depth sensor to extract features of Hu moment invariants from depth data, through which feature-based template match hand gesture recognition is developed. In order to construct continuous-time robot operations using dynamic hand gestures with concatenations of a series of hand gesture actions, the wake-up reminder scheme using fingertip detection calculations is established to accurately denote the starting, ending, and switching timestamps of a series of gesture actions. To be able to perform typical template match on continuous-time dynamic hand gesture recognition with the ability of real-time recognition, representative frame estimates using centroid, middle, and middle-region voting approaches are also presented and combined with template match computations. Experimental results show that, in certain continuous-time periods, the proposed complete hand gesture recognition framework can provide a smooth operation for the material-handling robot when compared with robots controlled using only extractions of full frames; presented representative frames estimated by middle-region voting will maintain fast computations and still reach the competitive recognition accuracy of 90.8%. The method proposed in this study can facilitate the smart assembly line and human–robot collaborations in automotive manufacturing.

Fuzzy Clustering of Infrared Image Features Based on Lazy Snapping Algorithm

Substation equipment is an important part of the power grid, which undertakes the function of power transmission and conversion and directly affects the operation status of the whole substation and power system. When the substation equipment is in an abnormal working state, the temperature will change. Therefore, the temperature information of the substation equipment is used as the judgment basis to complete the judgment of the working state of the equipment, which can realize the fault diagnosis of the substation equipment and ensure that the power system works in a safe and reliable environment. In this paper, according to the characteristics of the transformer equipment shape stability, the invariant moment is used to extract the infrared image feature of the transformer equipment. The support vector machine is used to complete the classification and recognition of the image. Hu invariant moments and Zernike invariant moments are used to extracting features respectively, and the results of feature extraction are used as training samples to train support vector machines for recognition. Using the Lazy Snapping algorithm to complete the infrared image segmentation processing of substation equipment, the target region is extracted from the background completely, and the segmentation image information is completed. In the experimental test, through the test of the recognition model constructed by invariant moments, it is proved that the recognition accuracy of Zernike invariant moments combined with support vector machine in this paper is higher, and the Lazy Snapping algorithm has obvious advantages in segmentation quality compared with other methods. Therefore, the method of intelligent identification and diagnosis of the thermal fault of substation equipment studied in this paper has important practical significance for the establishment of an online diagnosis system.

Machine learning-based identification of craniosynostosis in newborns

Early closure of cranial vault sutures, defined as craniosynostosis is a relatively common condition with somehow specific head and face abnormality for each subtype. Early diagnosis results in a better prognosis but pediatricians and primary care providers are not so familiar with these abnormalities while taking 3D CT scan of skull, predisposes the growing brain to harmful effects of radiation. Thus, developing a user-friendly and accurate diagnostic system would be helpful. This study aimed to diagnose simple suture synostosis by using machine learning based methods in digital photographs of child head. Digital photos of 145 craniosynostosis infants, operated in Mofid children hospital (Tehran, Iran) are used in this study. Head border is identified by GrabCut algorithm segmentation method and then several anthropometric indices such as cranial index (CI), cranial vault asymmetry index (CVAI), anterior–midline width ratio (AMWR) and anterior–posterior​ width ratio (APWR) and left–right height ratio (LRHR) are calculated. Moreover, statistical pattern matching indices (Chi-square (CS), Hu moment invariants (HuMI), absolute difference of white pixels probability (AbsDifWPP) and pixel intensity (PI)) are calculated and compared to anthropometric indices. The classification results for statistical pattern matching indices varied in the range of 85%–92% which is statistically higher than hand crafted indices. Our proposed approach could diagnose and classify common subtypes of single suture craniosynostosis and could help pediatricians and parents in early diagnosis and follow-up of this disorder.

Simulation evaluation of knitted mesh structure using morphological operations

High-quality simulation of fabric structural features is significant for realizing the morphological prediction of nonuniform mesh structures and building numerical simulation of physical properties. For lace textiles which have hundreds and thousands of variable meshes, an objective and unified standard to evaluate structural deformation is quite challenging. It needs to comprehensively and quantitatively conclude a result, instead of visual and subjective judgment. Therefore, this paper proposed an image-based method to mathematically solve accuracy by comparing morphological features from both simulation result and real fabric photograph. Jacquardtronic lace textiles were fabricated as experimental samples. Based on morphological expansion and corrosion algorithms, non-characteristic lapping details were eroded from binary images with only featured single-pixel contours of irregular meshes. Shape descriptor of each featured image was represented by a moment vector of seven Hu invariant moments. Then the morphological vectors of both simulated and real fabric images were substituted into a defined equation of similarity measurement. This image-based evaluation model effectively avoids defects of subjective visual observation and geometric measurement methods.