Blind Feature Extraction for Time-Series Classification Using Haar Wavelet Transform

2005 ◽  
pp. 605-610 ◽  
Author(s):  
Hui Zhang ◽  
Tubao Ho ◽  
Wei Huang
Keyword(s):  
Time Series ◽  
Feature Extraction ◽  
Wavelet Transform ◽  
Haar Wavelet ◽  
Time Series Classification ◽  
Haar Wavelet Transform

Adaptive Feature Extraction and Image matching Based on Haar Wavelet Transform and SIFT

2012 ◽  
Vol 6 (7) ◽  
pp. 1-8 ◽  
Author(s):  
Mengmeng Zhang ◽  
Zeming Li ◽  
ChangNian Zhang ◽  
Huihui Bai
Keyword(s):  
Feature Extraction ◽  
Wavelet Transform ◽  
Image Matching ◽  
Haar Wavelet ◽  
Haar Wavelet Transform ◽  
Adaptive Feature Extraction

scholarly journals IMPROVING TIME SERIES CLASSIFICATION ACCURACY: COMBINING GLOBAL AND LOCAL INFORMATION IN THE SIMILARITY CRITERION

2014 ◽  
Vol 44 (3) ◽  
pp. 225-229
Author(s):  
X. HE ◽  
C. SHAO ◽  
Y. XIONG
Keyword(s):  
Time Series ◽  
Nearest Neighbor ◽  
Similarity Criterion ◽  
Haar Wavelet ◽  
Local Information ◽  
Time Series Classification ◽  
Shape Information ◽  
Haar Wavelet Transform ◽  
Invariant Distance ◽  
Global And Local

Given the widespread use of time series classification in many domains, how to improve the accuracy of classification has attracted considerable focus. In this paper, a new similarity measure (SIMscl) based on the global and local information has been proposed for improving the precision rate of one nearest neighbor (1NN) classifier. Specifically, the global information records the intrinsic properties of time series, and is reflected by two indicators: the shape information and the complexity; the local information pays attention to the exact match of value, and is realized by LB_keogh. Simultaneously, a method based on multi-scale discrete haar wavelet transform, key point extraction, and symbolization has been put forward to extract the shape information. To test the efficacy of the proposed shape similarity SIMshape and hybrid similarity SIMscl, the experiments are conducted on two data sets: star light curve and beef. Experimental evaluations show that SIMshape can deal with some time series misclassified by Euclidean Distance (ED), LB_keogh, and Complexity Invariant Distance (CID), and SIMscl has higher precision than ED, LB_keogh, and CID in time series 1NN classification.

Novel Model for Automatic Classification of the Epileptic Seizures Using Fast Fourier Series-Haar Wavelet Transform

2021 ◽  
Vol 11 (12) ◽  
pp. 3209-3214
Author(s):  
P. Geetha ◽  
S. Nagarani
Keyword(s):  
Feature Extraction ◽  
Image Segmentation ◽  
Fourier Series ◽  
Wavelet Transform ◽  
Epileptic Seizures ◽  
Medical Application ◽  
Scale Space ◽  
Haar Wavelet ◽  
Sample Length ◽  
Haar Wavelet Transform

The disorder based on neurological can be considered as epilepsy that leads to the recurrent seizures in occurrence. The electronic characteristics of brain can be monitor by the electroencephalogram (EEG). It is most commonly used in the medical application. The function monitoring records can be non linear as well as non stationary functioning. The present work produce a novel methodology, it is depend on Fast Fourier series (FFS) and wavelet transform based on Haar. These methods are used for the various kinds of epileptic seizure the electroencephalogram based signal. The detection of boundary is occur by the representation of scale-space and it also adapted to the image segmentation of the spectrum depends on the FBSE that can be obtained with the electroencephalogram based signal and the purpose of the EWT is also used to attain the narrow sub band based signals. These image segmentation and classification process implementation by FPGA based microprocessor and systems. The FFS-HMT can produce the sub band signal from the Hilbert marginal spectrum it is represented as HMS. The HMS can be used to compute the line length and the entropy characteristics due to the corresponding various kinds of the level based oscillatory of the electroencephalogram signal. Here we apply the selected feature extraction depends on the ranking parallel vector. With the use of an electroencephalogram signal, the robust random forest is utilized to classify selected feature extraction in normal and epileptic participants. The assessment of performance based on classification can be measured in FPGA microprocessor the term of classification accuracy for different sample length of EEG. The current methodology aids neurologists in distinguishing between healthy and epileptic people using electroencephalogram signals.

Image feature extraction with the optical Haar wavelet transform

1995 ◽  
Vol 34 (4) ◽  
pp. 1238 ◽  
Author(s):  
Guofan Jin
Keyword(s):  
Feature Extraction ◽  
Wavelet Transform ◽  
Haar Wavelet ◽  
Image Feature ◽  
Image Feature Extraction ◽  
Haar Wavelet Transform

scholarly journals Iris Feature Extraction based on Haar Wavelet Transform

2014 ◽  
Vol 8 (4) ◽  
pp. 265-272 ◽  
Author(s):  
Yao Zhu Wen ◽  
Jun Zhou ◽  
Yu feng Wu ◽  
Ming Jun Wang
Keyword(s):  
Feature Extraction ◽  
Wavelet Transform ◽  
Haar Wavelet ◽  
Haar Wavelet Transform ◽  
Iris Feature Extraction

Feature extraction from printed Persian sub-words using Haar wavelet transform

2011 ◽  
Author(s):  
Samira Nasrollahi Dizajyekan ◽  
Afshin Ebrahimi
Keyword(s):  
Feature Extraction ◽  
Wavelet Transform ◽  
Haar Wavelet ◽  
Haar Wavelet Transform

Using the Haar wavelet transform in the semiparametric specification of time series

2009 ◽  
Vol 26 (2) ◽  
pp. 392-403 ◽  
Author(s):  
Larry W. Taylor
Keyword(s):  
Time Series ◽  
Wavelet Transform ◽  
Haar Wavelet ◽  
Haar Wavelet Transform
Sign in / Sign up

Export Citation Format

Share Document