Rotation Invariant Method Research Articles

Computationally Efficient Direction Finding for Conformal MIMO Radar.

The use of conformal arrays offers a significant advancement in Multiple-Input-Multiple-Output (MIMO) radar, enabling the placement of antennas on irregular surfaces. For joint Direction-of-Departure (DOD) and Direction-of-Arrival (DOA) estimation in conformal-array MIMO radar, the current spectrum-searching methods are computationally too expensive, while the existing rotation-invariant method may suffer from phase ambiguity caused by the non-Nyquist spacing of the sensors. In this paper, an improved rotationally invariant technique is proposed. The core function of the proposed algorithm is to estimate the phase differences between the adjacent sensors; then, it eliminates phase ambiguity via the previous estimated standard phase difference. Thereafter, DODs and DOAs are obtained via Least Squares (LS) fitting. The proposed method provides closed-form estimates for joint DOD and DOA estimation, which is much more efficient than the existing spectrum-searching techniques. Numerical simulations show that the proposed algorithm can accurately determine 2D DODs and DOAs of targets, only requiring approximately 1% of the running time required by existing spectrum-searching approaches.

Rotation invariance and equivariance in 3D deep learning: a survey

Deep neural networks (DNNs) in 3D scenes show a strong capability of extracting high-level semantic features and significantly promote research in the 3D field. 3D shapes and scenes often exhibit complicated transformation symmetries, where rotation is a challenging and necessary subject. To this end, many rotation invariant and equivariant methods have been proposed. In this survey, we systematically organize and comprehensively overview all methods. First, we rewrite the previous definition of rotation invariance and equivariance by classifying them into weak and strong categories. Second, we provide a unified theoretical framework to analyze these methods, especially weak rotation invariant and equivariant ones that are seldom analyzed theoretically. We then divide existing methods into two main categories, i.e., rotation invariant ones and rotation equivariant ones, which are further subclassified in terms of manipulating input ways and basic equivariant block structures, respectively. In each subcategory, their common essence is highlighted, a couple of representative methods are analyzed, and insightful comments on their pros and cons are given. Furthermore, we deliver a general overview of relevant applications and datasets for two popular tasks of 3D semantic understanding and molecule-related. Finally, we provide several open problems and future research directions based on challenges and difficulties in ongoing research.

LPHOG: A Line Feature and Point Feature Combined Rotation Invariant Method for Heterologous Image Registration

Remote sensing image registration has been a very important research topic, especially the registration of heterologous images. In the research of the past few years, numerous registration algorithms for heterogenic images have been developed, especially feature-based matching algorithms, such as point feature-based or line feature-based matching methods. However, there are few matching algorithms that combine line and point features. Therefore, this study proposes a matching algorithm that combines line features and point features while achieving good rotation invariance. It comprises LSD detection of line features, keypoint extraction, and HOG-like feature descriptor construction. The matching performance is compared with state-of-the-art matching algorithms on three heterogeneous image datasets (optical–SAR dataset, optical–infrared dataset, and optical–optical dataset), verifying our method’s rotational invariance by rotating images in each dataset. Finally, the experimental results show that our algorithm outperforms the state-of-the-art algorithms in terms of matching performance while possessing very good rotation invariance.

Radiomics for Parkinson's disease classification using advanced texture-based biomarkers

Parkinson's disease (PD) is one of the neurodegenerative diseases and its manual diagnosis leads to time-consuming process. MRI-based computer-aided diagnosis helps medical experts to diagnose PD more precisely and fast. Texture-based radiomic analysis is carried out on 3D MRI scans of T1 weighted and resting-state modalities. 43 subjects from Neurocon and 40 subjects from Tao-Wu dataset were examined, which consisted of 36 scans of healthy controls and 47 scans of Parkinson's patients. Total 360 2D MRI images are selected among around 17000 slices of T1-weighted and resting scans of selected 72 subjects. Local binary pattern (LBP) method was applied with custom variants to acquire advanced textural biomarkers from MRI images. LBP histogram helped to learn discriminative local patterns to detect and classify Parkinson's disease. Using recursive feature elimination, data dimensions of around 150-300 LBP histogram features were reduced to 13-21 most significant features based on score, and important features were analysed using SVM and random forest algorithms. Variant-I of LBP has performed well with highest test accuracy of 83.33%, precision of 84.62%, recall of 91.67%, and f1-score of 88%. Classification accuracies were obtained from 61.11% to 83.33% and AUC-ROC values range from 0.43 to 0.86 using four variants of LBP.•Parkinson's classification is carried out using an advanced biomedical texture feature. Texture extraction using four variants of uniform, rotation invariant LBP method is performed for radiomic analysis of Parkinson's disorder.•Proposed method with support vector machine classifier is experimented and an accuracy of 83.33% is achieved with 10-fold cross validation for detection of Parkinson's patients from MRI-based radiomic analysis.•The proposed predictive model has proved the potential of textures of extended version of LBP, which have demonstrated subtle variations in local appearance for Parkinson's detection.

Feature Extraction Based on ORB- AKAZE for Echocardiogram View Classification

In computer vision, the extraction of robust features from images to construct models that automate image recognition and classification tasks is a prominent field of research. Handcrafted feature extraction and representation techniques become critical when dealing with limited hardware resource settings, low-quality images, and larger datasets. We propose two state-of-the-art handcrafted feature extraction techniques, Oriented FAST and Rotated BRIEF (ORB) and Accelerated KAZE (AKAZE), in combination with Bag of Visual Word (BOVW), to classify standard echocardiogram views using Machine learning (ML) algorithms. These novel approaches, ORB and AKAZE, which are rotation, scale, illumination, and noise invariant methods, outperform traditional methods. The despeckling algorithm Speckle Reduction Anisotropic Diffusion (SRAD), which is based on the Partial Differential Equation (PDE), was applied to echocardiogram images before feature extraction. Support Vector Machine (SVM), decision tree, and random forest algorithms correctly classified the feature vectors obtained from the ORB with accuracy rates of 96.5%, 76%, and 97.7%, respectively. Additionally, AKAZE's SVM, decision tree, and random forest algorithms outperformed state-of-the-art techniques with accuracy rates of 97.7%, 90%, and 99%, respectively.

Classification of airborne 3D point clouds regarding separation of vegetation in complex environments.

Classification of outdoor point clouds is an intensely studied topic, particularly with respect to the separation of vegetation from the terrain and manmade structures. In the presence of many overhanging and vertical structures, the (relative) height is no longer a reliable criterion for such a separation. An alternative would be to apply supervised classification; however, thousands of examples are typically required for appropriate training. In this paper, an unsupervised and rotation-invariant method is presented and evaluated for three datasets with very different characteristics. The method allows us to detect planar patches by filtering and clustering so-called superpoints, whereby the well-known but suitably modified random sampling and consensus (RANSAC) approach plays a key role for plane estimation in outlier-rich data. The performance of our method is compared to that produced by supervised classifiers common for remote sensing settings: random forest as learner and feature sets for point cloud processing, like covariance-based features or point descriptors. It is shown that for point clouds resulting from airborne laser scans, the detection accuracy of the proposed method is over 96% and, as such, higher than that of standard supervised classification approaches. Because of artifacts caused by interpolation during 3D stereo matching, the overall accuracy was lower for photogrammetric point clouds (74-77%). However, using additional salient features, such as the normalized green-red difference index, the results became more accurate and less dependent on the data source.

Rotation Invariant Predictor-Corrector for Smoothed Particle Hydrodynamics Data Visualization

In order to extract the vortex features more accurately, a new method of vortex feature extraction on the Smoothed Particle Hydrodynamics data is proposed in the current study by combining rotation invariance and predictor-corrector method. There is a limitation in the original rotation invariance, which can only extract the vortex features that perform equal-speed rotations. The limitation is slightly weakened to a situation that the rotation invariance can be used, given that a specific axis is existed in the fluid to replace the axis needed for it. Therefore, as long as the axis exists, the modified rotation invariant method can be used. Meanwhile, the vortex features are extracted by predictor-corrector method. By calculating the cross product of the parallel vector field, the seed candidates of vortex core lines can be obtained, and the real seed points can be gained from the rotation invariant Jacobian. Finally, the seed point and a series of candidates based on the predictor-corrector method are connected to draw the vortex core lines. Compared with the original method, the rotation invariant predictor-corrector method not only expands the application scope, but also ensures the accuracy of extraction. Our method adds the steps of calculating the rotation invariant Jacobian, the performance is slightly lower, but with the increase of the particle number, the performance gradually tends to the original method.

Slave-boson analysis of the two-dimensional Hubbard model

We present a comprehensive study of the 2D one-band Hubbard model applying the spin rotation invariant slave-boson method. We utilize a spiral magnetic mean field and fluctuations around a paramagnetic mean field to determine the magnetic phase diagram and find the two approaches to be in good agreement. Apart from the commensurate phases characterized by ordering wave vectors $\boldsymbol Q = (\pi, \pi), (0, \pi)$, and $(0, 0)$ we find incommensurate phases where the ordering wave vectors $\boldsymbol Q = (Q, Q)$, and $(Q, \pi)$ vary continuously with filling, interaction strength or temperature. The mean field quantities magnetization and effective mass are found to change discontinuously at the phase boundaries separating the $(Q, Q)$ and $(Q, \pi)$ phases, indicating a first order transition. The band structure and Fermi surface is shown in selected cases. The dynamic spin and charge susceptibilities as well as the structure factors are calculated and discussed, including the emergence of collective modes of the zero sound and Mott insulator type. The dynamical conductivity is calculated in dependence of doping, interaction strength and temperature. Finally, a temperature-interaction strength phase diagram is established.

Eigen-Inference Precoding for Coarsely Quantized Massive MU-MIMO System With Imperfect CSI

This work considers the precoding problem in massive multiuser multiple-input multiple-output (MU-MIMO) systems equipped with low-resolution digital-to-analog converters (DACs). In previous literature on this topic, it is commonly assumed that the channel state information (CSI) is perfectly known. However, in practical applications the CSI is inevitably contaminated by noise. In this paper, we propose, for the first time, an eigen-inference (EI) precoding scheme to improve the error performance of the coarsely quantized massive MU-MIMO systems under imperfect CSI, which is mathematically modeled by a sum of two rectangular random matrices (RRMs). Instead of performing analysis based on the RRM, using Girko's Hermitization trick, the proposed method leverages the block random matrix theory by augmenting the RRM into a block symmetric channel matrix (BSCA). Specially, we derive the empirical distribution of the eigenvalues of the BSCA and establish the limiting spectra distribution connection between the true BSCA and its noisy observation. Then, based on these theoretical results, we propose an EI-based moments matching method for CSI-related noise level estimation and a rotation invariant estimation method for CSI reconstruction. Based on the cleaned CSI, the quantized precoding problem is tackled via the Bussgang theorem and the Lagrangian multiplier method. The prosed methods are lastly verified by numerical simulations and the results demonstrate the effectiveness of the proposed precoder.

Aeroplane detection from high-resolution remotely sensed imagery using bag-of-visual-words based hough forests

ABSTRACT This paper presents a rotation-invariant method for detecting aeroplanes from high-resolution remotely sensed images. First, a superpixel-based strategy is proposed to generate salient and distinctive feature regions. Second, a bag-of-visual-words representation is adopted to characterize spectral statistical features of feature regions. Third, a multi-scale rotation-invariant Hough forest with embedded scale factors and orientation information is trained to cast rotation-invariant votes for estimating aeroplane centroids. Quantitative evaluations on the images collected from Google Earth service show that a completeness, correctness, quality, and F 1-measure of 0.980, 0.973, 0.954, and 0.976, respectively, are obtained. Comparative studies with five existing methods also demonstrate the superior performance of the proposed method in accurately and correctly detecting arbitrarily-orientated and varying-sized aeroplanes.

Local-Entropy Based Approach for X-Ray Image Segmentation and Fracture Detection.

The paper proposes a segmentation and classification technique for fracture detection in X-ray images. This novel rotation-invariant method introduces the concept of local entropy for de-noising and removing tissue from the analysed X-ray images, followed by an improved procedure for image segmentation and the detection of regions of interest. The proposed local Shannon entropy was calculated for each image pixel using a sliding 2D window. An initial image segmentation was performed on the entropy representation of the original image. Next, a graph theory-based technique was implemented for the purpose of removing false bone contours and improving the edge detection of long bones. Finally, the paper introduces a classification and localisation procedure for fracture detection by tracking the difference between the extracted contour and the estimation of an ideal healthy one. The proposed hybrid method excels at detecting small fractures (which are hard to detect visually by a radiologist) in the ulna and radius bones—common injuries in children. Therefore, it is imperative that a radiologist inspecting the X-ray image receives a warning from the computerised X-ray analysis system, in order to prevent false-negative diagnoses. The proposed method was applied to a data-set containing 860 X-ray images of child radius and ulna bones (642 fracture-free images and 218 images containing fractures). The obtained results showed the efficiency and robustness of the proposed approach, in terms of segmentation quality and classification accuracy and precision (up to and , respectively).

A rotation and translation invariant method for 3D organ image classification using deep convolutional neural networks.

Three-dimensional (3D) medical image classification is useful in applications such as disease diagnosis and content-based medical image retrieval. It is a challenging task due to several reasons. First, image intensity values are vastly different depending on the image modality. Second, intensity values within the same image modality may vary depending on the imaging machine and artifacts may also be introduced in the imaging process. Third, processing 3D data requires high computational power. In recent years, significant research has been conducted in the field of 3D medical image classification. However, most of these make assumptions about patient orientation and imaging direction to simplify the problem and/or work with the full 3D images. As such, they perform poorly when these assumptions are not met. In this paper, we propose a method of classification for 3D organ images that is rotation and translation invariant. To this end, we extract a representative two-dimensional (2D) slice along the plane of best symmetry from the 3D image. We then use this slice to represent the 3D image and use a 20-layer deep convolutional neural network (DCNN) to perform the classification task. We show experimentally, using multi-modal data, that our method is comparable to existing methods when the assumptions of patient orientation and viewing direction are met. Notably, it shows similarly high accuracy even when these assumptions are violated, where other methods fail. We also explore how this method can be used with other DCNN models as well as conventional classification approaches.

Iris recognition system based on a new combined feature extraction method

Recent science studies are interested in automatic systems without human intervention. The security field is in a great need of automatic identification system based on biometric properties. The uniqueness of the texture present in the human iris is a natural password. In this paper, we propose a novel iris recognition approach based on a combination of two systems. The first system extracts the local variation of the mean and the variance. The second system is based on the rotation invariant neighbourhoodbased binary pattern method (Hamouchene and Aouat, 2014a). Two sets of support vector machines (SVM) are used to train each system. Dempster-Shafer theory is used to distribute unitary mass over the two output sets of SVMs. Finally, the combined belief measures are transformed to a probability by applying the Dezert-Smarandache theory. In the experiments, the proposed system is compared to famous iris recognition systems and obtained better recognition rates.

Entropy based Local Binary Pattern (ELBP) feature extraction technique of multimodal biometrics as defence mechanism for cloud storage

Cloud Computing (CC) is a technology that is growing by leaps and bounds and has attracted wide spectrum of users. The extensive usage of cloud technology is influenced by multiple factors like ease of use, pay-per usage strategy, easy access, cost-effectiveness etc. Though it is a widely used technology, challenges exist in the form of security threats. There are a variety of services that are offered by cloud. These include Software as a Service (SaaS), Infrastructure as a Service (IaaS) and Platform as a Service (PaaS). Storage is one of the key service offerings under IaaS. To provide a secure digital platform for users to work with, this research work proposes a novel security architecture for secured storage in cloud that provides a robust authentication by employing multiple biometric modalities from users and allow/deny access accordingly. The crux of better authentication relies on the way the features are extracted from multiple biometric sensors and matched with registered users. For this purpose, a novel feature extraction technique is proposed in this research work. Entropy Based Local Binary Pattern (ELBP) is a new texture-based feature extraction technique proposed to describe the entropy information into Local Binary Pattern histogram in one-dimensional space. ELBP feature extraction technique needs no quantization. Biometric images exhibit higher uniqueness and hence incorporating entropy values into local regions add higher information content to these images, thus leading to better feature extraction. The experiments are performed on biometric images from Chinese Academy of Science, Institute of Automation (CASIA) Iris, Face and Fingerprint databases and the results show that the proposed ELBP feature extraction achieves substantial improvement, in terms of various classification metrics like accuracy, precision, recall etc. in comparison with the conventional rotation invariant LBP methods. The Receiver Operating Characteristics Curve (ROC) also bears testimony to the performance of the authentication system.

Thermodynamics of the kagome-lattice Heisenberg antiferromagnet with arbitrary spin S

We use a second-order rotational invariant Green's function method (RGM) and the high-temperature expansion (HTE) to calculate the thermodynamic properties, of the kagome-lattice spin-$S$ Heisenberg antiferromagnet with nearest-neighbor exchange $J$. While the HTE yields accurate results down to temperatures of about $T/S(S+1) \sim J$, the RGM provides data for arbitrary $T \ge 0$. For the ground state we use the RGM data to analyze the $S$-dependence of the excitation spectrum, the excitation velocity, the uniform susceptibility, the spin-spin correlation functions, the correlation length, and the structure factor. We found that the so-called $\sqrt{3}\times\sqrt{3}$ ordering is more pronounced than the $q=0$ ordering for all values of $S$. In the extreme quantum case $S=1/2$ the zero-temperature correlation length is only of the order of the nearest-neighbor separation. Then we study the temperature dependence of several physical quantities for spin quantum numbers $S=1/2,1,\dots,7/2$. As increasing $S$ the typical maximum in the specific heat and in the uniform susceptibility are shifted towards lower values of $T/S(S+1)$ and the height of the maximum is growing. The structure factor ${\cal S}(\mathbf{q})$ exhibits two maxima at magnetic wave vectors $\mathbf{q}={\mathbf{Q}_i}, i=0,1,$ corresponding to the $q=0$ and $\sqrt{3}\times\sqrt{3}$ state. We find that the $\sqrt{3}\times \sqrt{3}$ short-range order is more pronounced than the $q=0$ short-range order for all temperatures $T \ge 0$. For the spin-spin correlation functions, the correlation lengths and the structure factors, we find a finite low-temperature region $0 \le T < T^*\approx a/S(S+1)$, $a \approx 0.2$, where these quantities are almost independent of $T$.

A Contactless Rotation-Invariant Palm Vein Recognition System

There is a high demand for contactless biometrics due to social concern for more convenient and hygienic technology. However, contactless biometrics like palm vein recognition faces a number of challenges. Hand rotation is among the main challenges to be solved. A number of methods have been introduced to extract palm region of interest (ROI) and have reported high accuracy, but the maximum rotation that can be handled by the methods are never mentioned. Usually, the ROI detection method fails to work if a large rotation exists in the hand position. In this paper, we propose a rotation-invariant method that can detect a palm ROI in any rotation angles. The method can sustain the recognition accuracy despite a substantial hand rotation of up to 360°. ROI detection in different angles causes minor ROI dislocation, thus a feature extraction technique called Winner-Take-All hashing (WTA) is developed. WTA serves as a feature enhancer, and is equipped with the Fisher’s Discriminant criterion to obtain a discriminative representation from the palm ROI. Experiment result shows that the proposed method is able to achieve a maximum recall of 94% and an EER at 7.12% for a challenging scenario containing hand rotation of up to 360°.

Comparison of rotation invariant local frequency, LBP and SFTA methods for breast abnormality classification

Comparison of rotation invariant local frequency, LBP and SFTA methods for breast abnormality classification

Comparison of rotation invariant local frequency, LBP and SFTA methods for breast abnormality classification

Breast cancer is the second most prominent cancer diagnosed among women. Digital mammography is one of the effective imaging modalities used to detect breast cancer in early stages. Computer-aided detection systems help radiologists to detect and diagnose abnormalities earlier and faster in a mammogram. In this paper, a comprehensive study is carried out on different feature extraction methods for classification of abnormal areas in a mammogram. The prominent techniques used for feature extraction in this study are local binary pattern (LBP), rotation invariant local frequency (RILF) and segmented fractal texture analysis (SFTA). Features extracted from these techniques are then fed to a support vector machine (SVM) classifier for further classification via 10-fold cross-validation method. The evaluation is performed using image retrieval in medical applications (IRMA) database for feature extraction. Our statistical analysis shows that the RILF technique outperforms the LBP and SFTA techniques.

Global and Local Oriented Edge Magnitude Patterns for Texture Classification

In this paper, we propose a gray-scale texture descriptor, name the global and local oriented edge magnitude patterns (GLOEMP), for texture classification. GLOEMP is a framework, which is able to effectively combine local texture, global structure information and contrast of texture images. In GLOEMP, the principal orientation is determined by Histogram of Gradient (HOG) feature, then each direction is respectively shown in detail by a local binary patterns (LBP) occurrence histogram. Due to the fact that GLOEMP characterizes image information across different directions, it contains very abundant information. The global-level rotation compensation method is proposed, which shifts the principal orientation of the HOG to the first position, thus allowing GLOEMP to be robust to rotations. In addition, gradient magnitudes are used as weights to add to the histogram, making GLOEMP robust to lighting variances as well, and it also possesses a strong ability to express edge information. The experimental results obtained from the representative databases demonstrate that the proposed GLOEMP framework is capable of achieving significant improvement, in some cases reaching classification accuracy of 10% higher than over the traditional rotation invariant LBP method.

The transform performing algorithm for frequency domain search

The algorithm basic for a rotation invariant method producing frequency transform is offered. The execution is based on a blocked matrixes usage. Each group of blocks consists of four units. Transform calculation direction is different for each block in a group. The transform results have a lot of fragments for which invariance for rotation is present for angles divisible by 90 degrees. This limitation is determined by a data representation rectangular grid. The consideration is based on two-dimensional digital cosine transform, but it could be used for different transform basics and any dimensions. The Hadamard matrix can be considered as such an example. The same algorithm could be used for compressed domain image retrievals based on DCT-processing.