Shortest Euclidean Distance Research Articles

Face Recognition Based on a Combination of Increasing Virtual Samples and Kernel Principal Component Analysis

In pattern recognition such as face recognition, the recognition result is not only limited by the quality and quantity of samples, but also limited by the extracted principal components. For improving the quality and quantity of training samples and for extracting more efficient principal components, this paper presents a recognition method combing the increased virtual samples and kernel principal component analysis (KPCA), which doubly weakens the influence of nonlinear factors on face recognition. New database is generated with the pose-changed and the mirror-like virtual images. Then KPCA is used for dimension reduction and feature extraction. The shortest Euclidean distance is applied to measure similarity. A series of experiments are conducted in the ORL and YALE face database and the experimental results show the efficiency of the proposed method.

Multivariate analysis of morphological traits of the Nigerian Muscovy ducks (Cairina Moschata)

This study was designed to investigate the possibility of classifying Nigerian Muscovy ducks into distinct ecotypes using stepwise discriminant analysis and Euclidean distance. Data were collected on ten morphological traits (body weight, body length, body girth, wing length, shank length, shank circumference, thigh length, total leg length, bill length and bill width) of one thousand and twenty (1020) adult Muscovy ducks comprising 203, 598 and 289 randomly sampled ducks from the rain forest, derived savanna and Guinea savanna ecotypes, respectively. The results of the descriptive statistics revealed highly (p<0.001) significant effects of ecotype on eight of the morphological traits except bill width and body length. Stepwise discriminant analysis indicated that bill length and body length had highest discriminating powers among the six distinguishing variables. Result of the cross validation of the correct assignment of ducks into distinct ecotypes was low; 53.7, 41.8 and 42.3 % of samples drawn from the rain forest, derived savanna and Guinea savanna ecotypes were correctly classified into their expected ecotypes. Euclidean distance between ecotypes was small; shortest distance was between rain forest and derived savanna ducks (2010) while the longest distance (3758) was between rain forest and Guinea savanna ducks. It is evident that due to the low classification success rate of discriminant analysis and short Euclidean genetic distance between ecotypes, they could not be classified into distinct ecotypes on the bases of morphostructural traits and appre-ciable heterosis is practically impossible from crossing Nigerian Muscovy ducks originating from different ecotypes.

WE‐D‐9A‐01: A Novel Mesh‐Based Deformable Surface‐Contour Registration

Purpose:Initial guess is vital for 3D‐2D deformable image registration (DIR) while dealing with large deformations for adaptive radiation therapy. A fast procedure has been developed to deform body surface to match 2D body contour on projections. This surface‐contour DIR will provide an initial deformation for further complete 3D DIR or image reconstruction.Methods:Both planning CT images and come‐beam CT (CBCT) projections are preprocessed to create 0–1 binary mask. Then the body surface and CBCT projection body contours are extracted by Canny edge detector. A finite element modeling system was developed to automatically generate adaptive meshes based on the image surface. After that, the projections of the CT surface voxels are computed and compared with corresponding 2D projection contours from CBCT scans. As a result, the displacement vector field (DVF) on mesh vertices around the surface was optimized iteratively until the shortest Euclidean distance between the pixels on the projections of the deformed CT surface and the corresponding CBCT projection contour is minimized. With the help of the tetrahedral meshes, we can smoothly diffuse the deformation from the surface into the interior of the volume. Finally, the deformed CT images are obtained by the optimal DVF applied on the original planning CT images.Results:The accuracy of the surface‐contour registration is evaluated by 3D normalized cross correlation increased from 0.9176 to 0.9957 (sphere‐ellipsoid phantom) and from 0.7627 to 0.7919 (H&amp;N cancer patient data). Under the GPU‐based implementation, our surface‐contour‐guided method on H&amp;N cancer patient data takes 8 seconds/iteration, about 7.5 times faster than direct 3D method (60 seconds/iteration), and it needs fewer optimization iterations (30 iterations vs 50 iterations).Conclusion:The proposed surface‐contour DIR method can substantially improve both the accuracy and the speed of reconstructing volumetric images, which is helpful for applying in adaptive radiotherapy.This research is supported by CPRIT individual investigator award RP110329.

Based on the Hamilton Improved Circle of Scraps Joining Together

By using Hamilton improved circle, transversely and longitudinally cut up paper join together. By scanning image and image extraction techniques, the images of shredded paper are obtained. The process of shredding splicing is mainly divided into two parts, the image preprocessing and edge matching. Then the images conver into grey value matrixes and binarization are implemented on the image with appropriate threshold. The images of shredded paper are classified into nineteen lines. With the shortest Euclidean distance between the edge of shredding and Hamilton improved circle, shredded paper is transversely spliced. The test shows that accuracy is as high as 100%.

SU‐E‐J‐81: A Contour‐Guided Deformable Image Registration Scheme for Organ Surface Deformation

Purpose: Accurate deformation vector field (DVF) acting on organ surface is vital for surface dose summation of deformable cavitary organs in adaptive radiotherapy. The aim of this study is to develop a contour‐guided deformable image registration (DIR) scheme to establish accurate DVF on an organ surface. Methods: A critical cavitary organ, for example bladder, is delineated by a clinician on fraction 1 and fraction 2 CT images used for cervical high‐dose rate (HDR) brachytherapy planning. Both CT sets are preprocessed by creating binary images by setting one inside of the organ under evaluation and zero outside. The organ surface is subsequently extracted and discretized with a triangular mesh. The DVF on designed vertices is estimated through an inverse‐consistent demons‐DIR between the paired binary images. Due to shortcomings in DIR algorithms, the DVF derived through DIR always has limited accuracy. To increase the DVF accuracy, an iterative closest point (ICP) algorithm is adopted in this study to match corresponding vertices of the paired surface meshes through an affine registration. The accuracy of the resulting registration is then evaluated. Results: Using contours drawn by a clinician as ground truth we evaluated DVF accuracy through organ surface‐to‐surface distance (SSD), defined as the shortest Euclidean distance between the vertices on the target and the deformed surfaces. In our evaluation case, the maximum SSD is 27.8mm before registration, decreases to 15.0mm following demons registration, and 8.33mm after further affine alignment. The 95 percentile SSD decreases from 23.9mm to 10.8mm and to 3.8mm, respectively. Conclusion: The proposed contour‐guided DIR scheme substantially improves the accuracy of DVF on organ surfaces. Results from the evaluation case demonstrate that the proposed contour‐guided DIR scheme can provide accurate DVF making it a useful approach to be applied in adaptive radiotherapy.

A robust evolutionary algorithm for the recovery of rational Gielis curves

Gielis curves (GC) can represent a wide range of shapes and patterns ranging from star shapes to symmetric and asymmetric polygons, and even self intersecting curves. Such patterns appear in natural objects or phenomena, such as flowers, crystals, pollen structures, animals, or even wave propagation. Gielis curves and surfaces are an extension of Lamé curves and surfaces (superquadrics) which have benefited in the last two decades of extensive researches to retrieve their parameters from various data types, such as range images, 2D and 3D point clouds, etc. Unfortunately, the most efficient techniques for superquadrics recovery, based on deterministic methods, cannot directly be adapted to Gielis curves. Indeed, the different nature of their parameters forbids the use of a unified gradient descent approach, which requires initial pre-processings, such as the symmetry detection, and a reliable pose and scale estimation. Furthermore, even the most recent algorithms in the literature remain extremely sensitive to initialization and often fall into local minima in the presence of large missing data. We present a simple evolutionary algorithm which overcomes most of these issues and unifies all of the required operations into a single though efficient approach. The key ideas in this paper are the replacement of the potential fields used for the cost function (closed form) by the shortest Euclidean distance (SED, iterative approach), the construction of cost functions which minimize the shortest distance as well as the curve length using R-functions, and slight modifications of the evolutionary operators. We show that the proposed cost function based on SED and R-function offers the best compromise in terms of accuracy, robustness to noise, and missing data.

Gaze Recognition and Application

This paper asserts a method of controlling the cursor based on the Fisher linear discriminant (FLD) analysis to recognize six classifications of human face movements, which are face movements upwards, downwards, leftwards, rightwards, blinking of the right eye and blinking of the left eye. These classifications represent cursor movements upward, downwards, leftwards, rightwards, and the right or left button on the mouse respectively. This method can be separated into two areas: face detection and gaze recognition. Face detection is to convert the RGB color space into YCbCr in order to segment skin tone areas, where the human face area is located by using the connect component labeling method. Gaze recognition is accomplished by building a gaze recognition training model parameter through FLD algorithm. Then, by using the Euclidean distance as the rule of decision, match the detected facial image to the parameters of this model to find the shortest Euclidean distance and its corresponding classification to control cursor movements.

Selection of lubricant using combined multiple attribute decision-making method

The main objective of this paper is to select a right lubricant from amongst a number of lubricants during the machining of En-31 steel work piece with tungsten carbide inserts by using combined multiple attribute decision–making method. The procedure is based on a combined TOPSIS and AHP method. The selection of an optimal material for an engineering design from a list of available alternative materials on the basis of two or more attributes in multiple attribute decision making problem. The analytic hierarchy process, being a simple, but powerful decision making tool, is being applied to solve different manufacturing problems. TOPSIS method is based on the concept that the chosen alternative should have the shortest Euclidean distance from the ideal solution and the farthest from the negative ideal solution. TOPSIS thus gives a solution that is not only closest to the hypothetical best, which is also the farthest from the hypothetically worst. Lubricant selection factors are identified and these are chip-tool interface temperature, cutting force, tool wear and surface roughness. Combined multi-attribute decision-making is aimed at integrating different measures into a single global lubricant index helps to select right lubricant and rank the given lubricant for a steel turning operation. The framework that is used in steel turning operation could serve as one of the tools for making a strategic decision. The effectiveness of our model is demonstrated through an actual experimental work.

An EFQM‐Rembrandt excellence model based on the theory of displaced ideal

PurposeThis paper aims to propose a new benchmarking framework that uses a series of existing intuitive and analytical methods to systematically capture both objective data and subjective beliefs and preferences from a group of decision makers (DMs).Design/methodology/approachThe proposed framework combines the excellence model developed by the European Foundation for Quality Management with the Rembrandt method, the entropy concept, the weighted‐sum approach, and the theory of the displaced ideal. Hard data and personal judgments are synthesized to evaluate a set of business units (BUs) with two overall performance scores plotted in a four quadrant model.FindingsThe two performance scores are used to benchmark the performance of the BUs in accordance with their Euclidean distance from the “ideal” BU. Quadrants are used to classify the BUs as efficacious, productive ineffectual, proficient unproductive, and inefficacious. The efficacious BUs, referred to as “excellent”, fall in the competency zone and have the shortest Euclidean distance from the ideal BU relative to their peers.Originality/valueThe benchmarking framework presented in this study has some obvious attractive features. First, the generic nature of the framework allows for the subjective and objective evaluation of a finite number of BUs by a group of DMs. Second, the information requirements of the framework are stratified hierarchically allowing DMs to focus on a small area of the large problem. Third, the framework does not dispel subjectivity; it calibrates the subjective weights with the objective weights determined through the entropy concept.

Morphological variation of Juniperus oxycedrus subsp. macrocarpa (Cupressaceae) in three Italian localities

The intra- and interpopulational geographic variation of three distant populations of &lt;em&gt;J. oxycedrus &lt;/em&gt;subsp. &lt;em&gt;macrocarpa &lt;/em&gt;was examined biometrically on the basis of morphological key-characters of needles, seeds and cones. The shortest Euclidean distances were used in the agglomerative grouping of closest neighbourhood and discrimination analysis with principal component analysis (PCA) was performed to verify the south-north differentiation of the taxon in Italy. Most of the cone and seed dimensional characters are correlated. No significant correlations were found between cone and needle characters. The sampled populations differ only insignificantly from one another, while the intrapopulational variation is slightly higher. Cone length as well as seed number, width and thickness were significantly correlated with geographic latitude, while the other four characters are not. Differences in cone shape were observed between individuals, so that several morphotypes can be distinguished.

Multi-modality morphological correlation of axillary lymph nodes

The goal of this study is to develop a computerized method that identifies a specific axillary lymph node (ALN) seen on ultrasound (US) with its most likely corresponding node on breast MRI (BMRI). This goal is an important step in developing a preoperative non-invasive method for staging breast cancer on the basis of multi-modality imaging. Twenty patients with newly diagnosed breast cancer were scanned on US and MRI. Two expert breast imaging radiologists independently correlated ALNs seen on US with BMRI, and this correlation was used as the gold standard. To correlate ALNs on US and BMRI, the cortex and hilum of each ALN was segmented using an ellipse fitting algorithm, then the ALN long and short axes and maximum cortical thickness (MCT) were computed. Three ALNs were chosen as candidates from the BMRI datasets for each lymph node seen on US. Finally, the Euclidean distances across all measurements between the US ALN and each of the three BMRI candidates were computed, and the smallest distance was reported as the correlation result. Using the expert radiologists identified correlated BMRI slice as the ground truth, the shortest Euclidean distance successfully identified the same lymph node as the radiologists in 13 out of 16 ALNs (81.25%). In negative ALNs, the standard deviation for long and short axes was relatively large but that of maximum cortical thickness was small. Average maximum cortical thickness and its standard deviation measured in US were very close to those measured in MRI. There were no significant differences among the long axis, short axis, and MCT measurements between US and MRI-T2 weighted sequence (P > 0.05 paired t-test). We performed a feasibility study which showed that computerized measurements of ALNs might be used to identify the same ALN on different modalities such as US and BMRI. This type of correlation would be valuable as it would allow the use of combined imaging parameters to be applied to the evaluation of ALNs in patients with breast cancer. It is hoped that the combined multi-modality information would provide a more robust non-invasive method of staging the axilla than is currently available.

An MRF-based kernel method for nonlinear feature extraction

Nonlinear kernel-based feature extraction algorithms have recently been proposed to alleviate the loss of class discrimination after feature extraction. When considering image classification, a kernel function may not be sufficiently effective if it depends only on an information resource from the Euclidean distance in the original feature space. This study presents an extended radial basis kernel function that integrates multiple discriminative information resources, including the Euclidean distance, spatial context, and class membership. The concepts related to Markov random fields (MRFs) are exploited to model the spatial context information existing in the image. Mutual closeness in class membership is defined as a similarity measure with respect to classification. Any dissimilarity from the additional information resources will improve the discrimination between two samples that are only a short Euclidean distance apart in the feature space. The proposed kernel function is used for feature extraction through linear discriminant analysis (LDA) and principal component analysis (PCA). Experiments with synthetic and natural images show the effectiveness of the proposed kernel function with application to image classification.

Hierarchical Organization of Human Cortical Networks in Health and Schizophrenia

The complex organization of connectivity in the human brain is incompletely understood. Recently, topological measures based on graph theory have provided a new approach to quantify large-scale cortical networks. These methods have been applied to anatomical connectivity data on nonhuman species, and cortical networks have been shown to have small-world topology, associated with high local and global efficiency of information transfer. Anatomical networks derived from cortical thickness measurements have shown the same organizational properties of the healthy human brain, consistent with similar results reported in functional networks derived from resting state functional magnetic resonance imaging (MRI) and magnetoencephalographic data. Here we show, using anatomical networks derived from analysis of inter-regional covariation of gray matter volume in MRI data on 259 healthy volunteers, that classical divisions of cortex (multimodal, unimodal, and transmodal) have some distinct topological attributes. Although all cortical divisions shared nonrandom properties of small-worldness and efficient wiring (short mean Euclidean distance between connected regions), the multimodal network had a hierarchical organization, dominated by frontal hubs with low clustering, whereas the transmodal network was assortative. Moreover, in a sample of 203 people with schizophrenia, multimodal network organization was abnormal, as indicated by reduced hierarchy, the loss of frontal and the emergence of nonfrontal hubs, and increased connection distance. We propose that the topological differences between divisions of normal cortex may represent the outcome of different growth processes for multimodal and transmodal networks and that neurodevelopmental abnormalities in schizophrenia specifically impact multimodal cortical organization.

An Improved Computer-aided Diagnosis Scheme Using the Nearest Neighbor Criterion for Determining Histological Classification of Clustered Microcalcifications

Our purpose was to evaluate the potential usefulness of the nearest neighbor case which was assumed to be the similar case in a CAD scheme for determining the histological classification of clustered microcalcifications. Our database consisted of current and previous magnification mammograms obtained from 93 patients before and after three-month follow-up examination. It included 11 invasive carcinomas, 19 noninvasive carcinomas of the comedo type, 25 non-invasive carcinomas of the noncomedo type, 23 mastopathies, and 15 fibroadenomas. Six objective features on clustered microcalcifications were first extracted from each of the current and the previous images. The nearest neighbor case was then identified by the Euclidean distance in the previous and current feature-space. The histological classification of an unknown new case in question was assumed to be the same as that of the nearest neighbor case which has the shortest Euclidean distance in our database. The classification accuracies were 90.9% for invasive carcinoma, 89.5% for noninvasive carcinoma of the comedo type, 96.0% for noninvasive carcinoma of the noncomedo type, 82.6% for mastopathy, and 93.3% for fibroadenoma. These results were substantially higher than those with our previous CAD scheme. The nearest neighbor criterion was useful in a CAD scheme for determining the histological classification.

Finding shortest path in the presence of barriers: An alternate approach

This paper proposes an alternate graphical approach with an increased accuracy to compute the shortest path distance in the presence of barriers between the pair of nodes. This methodology is used to compute the curvilinear path distance in a single-step between the pair of nodes. It is not possible to compute the shortest Euclidean distance in a single-step in the presence of barriers between the pair of nodes. The rectilinear distance in a single-step or multiple steps results in the maximum length. In such cases, the proposed methodology is used to compute the curvilinear path distance. The results of the proposed methodology to compute the curvilinear path distance are compared with the existing analytical approach in the presence of barriers between the pairs of nodes for the real life problem of the urban transportation system of the Kadapa in Andhra Pradesh State, India.

Hyperspectral images classification based on KLT

We propose an approach, based on KLT—Karhunen-Loeve Transform, in order to automatically classify hyperspectral images. We used an AVIRIS image with 224 bands taken from the surroundings of Alto Paraiso City, on Goias State, in the middle of Brazil. Our approach performs dimensionality reduction, using only the eigenvectors with the highest eigenvalues, generating an eigenspace of low dimension. We also consider the spectral signatures of each class we work with. The classification is done finding the shortest Euclidean distance among the primitives of the new images and the primitives of the classes. We built a thematic map with 4 different classes.

Non-perturbative renormalization of lattice operators in coordinate space

We present the first numerical implementation of a non-perturbative renormalization method for lattice operators, based on the study of correlation functions in coordinate space at short Euclidean distance. The method is applied to compute the renormalization constants of bilinear quark operators for the non-perturbative O(a)-improved Wilson action in the quenched approximation. The matching with perturbative schemes, such as MS¯, is computed at the next-to-leading order in continuum perturbation theory. A feasibility study of this technique with Neuberger fermions is also presented.

Incorporating heterogeneous distance metrics within block layout design

This paper focuses on optimal design of block layouts when using more than one distance metric within a single facility. Previous work in block layout has assumed a single distance metric, usually the shortest rectilinear distance between department centroids, during the design step. However, most facilities have more than one method of material handling and alternative material handling systems can imply alternative distance metrics and cost structures. Specifically, up to three distance metrics within a single facility are considered--the shortest rectilinear distance between centroids (appropriate for automated guided vehicles and forklift trucks), the Tchebychev (maximum) distance (appropriate for overhead cranes) and the shortest Euclidean distance between centroids (appropriate for conveyor lines). Optimal block layouts using each of these distance metrics individually and then collectively are compared and contrasted. This approach can also be used to compare layouts when the choice of material handling system is not clear. It is argued that incorporating the distance metric that best reflects the planned material handling device is more realistic than previous formulations, avoids block layouts that are sub-optimal for the material handling systems installed, and is quite workable within a heuristic optimization framework.

Estimation of an overidentified equation in an econometric model by the method of generalized least squares

As a mathematical preliminary we first prove by construction in Section III, that there always exists a matrix B of the form mxn and of rank r, which has the shortest Euclidean distance from a given matrix A of the form m×n and of rank greater than or equal to r. B may be called the rank r nearest Euclidean neighbour of the matrix A. This result becomes handy in Section IV where we show how to estimate the coefficiants of an overidentified equation of an econometric model, by the model of generalized least squares. Linear estimation appropriate for a multivariate linear model is first used to estimate the coefficients of the matrix A of an econometic model in its reduced form Next, these estimators are used to derive the estimators of the coefficients of the overidentified equation together with improved estimtators of the coefficients of the matrix A. Although presented in the context of econometrics, the statistical model and the estimation procedure will fit in in other areas of research as well.