Filters








8,333 Hits in 6.0 sec

Fast feature selection by analyzing class regions approximated by ellipsoids

K. Kawaratani, S. Abe
2000 Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium  
In our previous work, we have developed the backward feature selection method based on class regions approximated by ellipsoids.  ...  In this paper, we accelerate feature selection by the forward selection search, the symmetric Cholesky factorization, and deletion of duplicated calculations between consecutive factorizations.  ...  Approximation of Class Regions by Ellipsoids Using the m-dimensional training data belonging to class i we approximate the class region by an ellipsoid with the m-dimensional center vector: c i = 1 |X  ... 
doi:10.1109/ijcnn.2000.861366 dblp:conf/ijcnn/KawarataniA00 fatcat:fazasmci7rd5llienmtwii4pmm

Fast Forward Feature Selection by Analyzing Class Regions Approximated by Ellipsoids

Kota KAWARATANI, Shigeo ABE
2001 Transactions of the Institute of Systems Control and Information Engineers  
features Table 4 4 Recognition rate for numeral data Table 5 Recognition rate for blood cell data Table 7 7 Computation time for feature selection  ...  選 択 時 の認 識 率 が 安 定 して い る とい わ れ て い る が,楕 円領 域 に よる特 徴 選 択 に 関 して は順 方 向 の 特 徴 選 択 で Fig. 5 5 Fig. 5 Computation time for thyroid data Table 1 1 Benchmark data specifications Table 2 2 Number of selected  ... 
doi:10.5687/iscie.14.265 fatcat:wc4usec5nbbh3ccv4l2gd4ds2q

A Decision Support System for the Prediction of the Trabecular Fracture Zone [chapter]

Vasileios Korfiatis, Simone Tassani, George K. Matsopoulos
2012 Lecture Notes in Computer Science  
The DSS comprises of two main modules: the feature selection module and the classifier.  ...  The feature selection module is used for reducing the initial size of the input features' subset (29 features) and for keeping the most informative features in order to increase the classification's module  ...  The automatic tool allowed performing a fast and accurate three-dimensional identification of the broken region.  ... 
doi:10.1007/978-3-642-33212-8_15 fatcat:sy4kuk2qbjcybg7jclegfpfjkm

Anomaly Intrusion Detection Based on Hyper-ellipsoid in the Kernel Feature Space

2015 KSII Transactions on Internet and Information Systems  
The SVDD has some limited classification capability, because the hyper-sphere, even in feature space, can express only a limited region of the target class.  ...  This paper presents an anomaly detection algorithm for mitigating the limitations of the conventional SVDD by finding the minimum volume enclosing ellipsoid in the feature space.  ...  Next, we approximate the decision boundary with the hyper-ellipsoid in the feature space.  ... 
doi:10.3837/tiis.2015.03.019 fatcat:gjiqfla46nehdbj5dlzwtc4my4

Evolutionary Mahalanobis Distance-Based Oversampling for Multi-Class Imbalanced Data Classification

Leehter Yao, Tung-Bin Lin
2021 Sensors  
EMDO utilizes a set of ellipsoids to approximate the decision regions of the minority class.  ...  The number of synthetic minority samples generated by EMDO in every ellipsoid is determined based on the density of minority samples in every ellipsoid.  ...  Let the decision region of the jth class samples in the feature space be denoted as j ; j is approximated by the union of α j ellipsoids, that is, j ∼ = ∪ n=1...α j Φ j n . (4) The GKA is used for learning  ... 
doi:10.3390/s21196616 pmid:34640936 fatcat:4ycz6yqabzbuhpfs64h54lysam

AUTOMATIC TREE-CROWN DETECTION IN CHALLENGING SCENARIOS

Dimitri Bulatov, Isabell Wayand, Hendrik Schilling
2016 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
by assigning trees to one of the few types in the output model.  ...  technical contributions of the paper are: Employing watershed transformation combined with classification results, applying hotspots detectors for identifying treetops in groups of trees, and correcting NDSM by  ...  Figure 5 : 5 Top row: The first three images show selected trees which are approximated by conics as shown in the three images of the bottom row. Top right: Part of the DSM.  ... 
doi:10.5194/isprs-archives-xli-b3-575-2016 fatcat:o7uyahs6kbahjdqshqcit3lo2q

AUTOMATIC TREE-CROWN DETECTION IN CHALLENGING SCENARIOS

Dimitri Bulatov, Isabell Wayand, Hendrik Schilling
2016 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
by assigning trees to one of the few types in the output model.  ...  technical contributions of the paper are: Employing watershed transformation combined with classification results, applying hotspots detectors for identifying treetops in groups of trees, and correcting NDSM by  ...  Figure 5 : 5 Top row: The first three images show selected trees which are approximated by conics as shown in the three images of the bottom row. Top right: Part of the DSM.  ... 
doi:10.5194/isprsarchives-xli-b3-575-2016 fatcat:p22vwfsnjnaifjxkzlsfbc5ibm

The Ellipsoid Factor for Quantification of Rods, Plates, and Intermediate Forms in 3D Geometries

Michael Doube
2015 Frontiers in Endocrinology  
Ellipsoid Factor utilizes an ellipsoid optimization algorithm, which assumes that maximal ellipsoids are centered on the medial axis, then dilates, rotates, and translates slightly each ellipsoid until  ...  strongly oblate (discus-shaped) ellipsoids, to +1 for strongly prolate (javelin-shaped) ellipsoids.  ...  It is a convenient and intuitive method for displaying and analyzing ellipsoid geometry.  ... 
doi:10.3389/fendo.2015.00015 pmid:25762979 pmcid:PMC4329874 fatcat:vgci3opuszha3ic46cizear47m

Detection of Visual Signatures of Marine Mammals and Fish within Marine Renewable Energy Farms using Multibeam Imaging Sonar

Francisco Francisco, Jan Sundberg
2019 Journal of Marine Science and Engineering  
Harbour seals occurred as bright torpedo-like fast moving targets, whereas grey seals occurred as bulky-ellipsoidal targets with serpentine movements.  ...  The sonar images showed that cod occurred as bright, 0.9 m long, ellipsoidal targets shoaling in groups.  ...  ellipsoidal streamlined-bulkier ellipsoidal ellipsoidal ellipsoidal- bulkier Observed swimming behavior slow -fast slow -fast fast slow -fast fast static -slow © 2019 by the  ... 
doi:10.3390/jmse7020022 fatcat:twos5dhxjjbepjacqotogzvjp4

Optical Analysis of the Internal Void Structure in Polymer Membranes for Gas Separation

Chiara Muzzi, Alessio Fuoco, Marcello Monteleone, Elisa Esposito, Johannes C. Jansen, Elena Tocci
2020 Membranes  
This work presents a study on the correlation of the structural features of different membrane materials, analyzed by means of molecular dynamics simulation, and their gas diffusivity/selectivity.  ...  The main differences in the structural features of three different classes of polymers are investigated in this work (glassy polymers, superglassy perfluoropolymers and high free volume polymers of intrinsic  ...  As a result, this strategy provides a fast and versatile tool for screening promising materials in silico, by quickly linking their void's size with their size selectivity.  ... 
doi:10.3390/membranes10110328 pmid:33167364 fatcat:xns7lfdw6valjmpswmcd3rqotq

Feature-Based Laser Scan Matching For Accurate and High Speed Mobile Robot Localization

Ali Akbar Aghamohammadi, Hamid D. Taghirad, Amir Hossein Tamjidi, Ehsan Mihankhah
2007 European Conference on Mobile Robots  
The feature extraction algorithm proposed in this paper uses a global information of the whole scan data and local information around feature points.  ...  Uncertainty of each feature is represented using covariance matrices determined due to observation and quantization error.  ...  The ellipsoids are illustrated by a scaling factor of 10, indicating the 99 percent confidence region of each feature's covariance.  ... 
dblp:conf/emcr/AghamohammadiTTM07 fatcat:mpsykjd7xzhpjlhswkcf5kwpvu

Orthogonal relation between wavefield polarization and fastSwave direction in the Val d'Agri region: An integrating method to investigate rock anisotropy

M. Pischiutta, M. Pastori, L. Improta, F. Salvini, A. Rovelli
2014 Journal of Geophysical Research - Solid Earth  
1] Wavefield polarization is investigated using 200 seismograms recorded by a network of 20 stations installed on rock outcrops in the Val d'Agri region that hosts the largest oil fields in the southern  ...  Rovelli (2014), Orthogonal relation between wavefield polarization and fast S wave direction in the Val d'Agri region: An integrating method to investigate rock anisotropy,  ...  We gratefully acknowledge Luisa Valoroso for providing us seismic data recorded by the temporary network and 3-D hypocentral locations.  ... 
doi:10.1002/2013jb010077 fatcat:kjoy7n4na5hq5kqsvvtdaryneu

Ellipsoid Segmentation Model for Analyzing Light-Attenuated 3D Confocal Image Stacks of Fluorescent Multi-Cellular Spheroids

Michaël Barbier, Steffen Jaensch, Frans Cornelissen, Suzana Vidic, Kjersti Gjerde, Ronald de Hoogt, Ralph Graeser, Emmanuel Gustin, Yolanda T. Chong, Thomas Abraham
2016 PLoS ONE  
The spheroids are assumed to be approximately ellipsoid in shape.  ...  Therefore they are increasingly complemented by more physiologically relevant 3D models, such as spheroid microtumor cultures.  ...  Validation of the ellipsoid approximation.  ... 
doi:10.1371/journal.pone.0156942 pmid:27303813 pmcid:PMC4909318 fatcat:2tcwxyc4cnhc7pwmlbalforpxq

Accurate Color Classification and Segmentation for Mobile Robots [chapter]

Raziel Alvarez, Erik Millan, Alejandro Aceves-Lopez, Ricardo Swain-Oropez
2007 Mobile Robots: Perception & Navigation  
color class. d) YUV Ellipsoid color class.  ...  By creating an ellipsoidal primitive, color regions are bounded more precisely, discarding highlights that may also be caused by noise.  ... 
doi:10.5772/4772 fatcat:lrbepvcdhzeevigetptwb6hxni

A Trajectory Estimation Method for Badminton Shuttlecock Utilizing Motion Blur [chapter]

Hidehiko Shishido, Itaru Kitahara, Yoshinari Kameda, Yuichi Ohta
2014 Lecture Notes in Computer Science  
We acquire a movement speed vector based on the shape of a motion blur region.  ...  For example, Yang et. al.proposes a visual tracking method applying color and gradient orientation histogram features using a particle filter [17] . Another strategy applying Haar-like features and  ...  In this case, however, we can estimate the velocity information by analyzing the shape of the motion blur region.  ... 
doi:10.1007/978-3-642-53842-1_28 fatcat:qz5exfx4rne5fnfxoqwni4u76m
« Previous Showing results 1 — 15 out of 8,333 results