The New Approach to the Invariant Feature Extraction Using Ridgelet Transform

2014 ◽  
Vol 651-653 ◽  
pp. 2241-2244
Author(s):  
Ying Yu ◽  
Cai Lin Dong ◽  
Bo Wen Sheng ◽  
Wei Dan Zhong ◽  
Xiang Lin Zou
Keyword(s):  
Feature Extraction ◽  
Fourier Transform ◽  
Feature Vector ◽  
Typical Feature ◽  
Chinese Characters ◽  
New Approach ◽  
Ridgelet Transform ◽  
One Dimensional ◽  
Invariant Feature ◽  
The One

With the aim to meet the requirements of multi-directional choice, the paper raise a new approach to the invariant feature extraction of handwritten Chinese characters, with ridgelet transform as its foundation. First of all, the original images will be rotated to the Radon circular shift by means of Radon transform. On the basis of the characteristic that Fourier transform is row shift invariant, then, the one-dimensional Fourier transform will be adopted in the Radon domain to gain the conclusion that magnitude matrixes bear the rotation-invariance as a typical feature, which is pretty beneficial to the invariant feature extraction of rotation. When such is done, one-dimensional wavelet transform will be carried out in the direction of rows, thus achieving perfect choice of frequency, which makes it possible to extract the features of sub-line in the appropriate frequencies. Finally, the average values, standard deviations and the energy values will form the feature vector which is extracted from the ridgelet sub-bands. The approaches mentioned in the paper could satisfy the requirements from the form automatic processing on the recognition of handwritten Chinese characters.

scholarly journals Application of Deep Learning in Fault Diagnosis of Rotating Machinery

Processes ◽  
2021 ◽  
Vol 9 (6) ◽  
pp. 919
Author(s):  
Wanlu Jiang ◽  
Chenyang Wang ◽  
Jiayun Zou ◽  
Shuqing Zhang
Keyword(s):  
Neural Network ◽  
Feature Extraction ◽  
Fault Diagnosis ◽  
Rotating Machinery ◽  
Generative Adversarial Networks ◽  
Extraction Ability ◽  
One Dimensional ◽  
Adversarial Networks ◽  
Diagnosis Model ◽  
The One

The field of mechanical fault diagnosis has entered the era of “big data”. However, existing diagnostic algorithms, relying on artificial feature extraction and expert knowledge are of poor extraction ability and lack self-adaptability in the mass data. In the fault diagnosis of rotating machinery, due to the accidental occurrence of equipment faults, the proportion of fault samples is small, the samples are imbalanced, and available data are scarce, which leads to the low accuracy rate of the intelligent diagnosis model trained to identify the equipment state. To solve the above problems, an end-to-end diagnosis model is first proposed, which is an intelligent fault diagnosis method based on one-dimensional convolutional neural network (1D-CNN). That is to say, the original vibration signal is directly input into the model for identification. After that, through combining the convolutional neural network with the generative adversarial networks, a data expansion method based on the one-dimensional deep convolutional generative adversarial networks (1D-DCGAN) is constructed to generate small sample size fault samples and construct the balanced data set. Meanwhile, in order to solve the problem that the network is difficult to optimize, gradient penalty and Wasserstein distance are introduced. Through the test of bearing database and hydraulic pump, it shows that the one-dimensional convolution operation has strong feature extraction ability for vibration signals. The proposed method is very accurate for fault diagnosis of the two kinds of equipment, and high-quality expansion of the original data can be achieved.

The shift lattice: an interpretation of some infinitely adaptive structures

1993 ◽  
Vol 440 (1908) ◽  
pp. 23-36 ◽  
Keyword(s):  
Fourier Transform ◽  
Fourier Transforms ◽  
Periodic Systems ◽  
Adaptive Structures ◽  
One Dimensional ◽  
Modulated Structures ◽  
Simple Generalization ◽  
The One

The structures of various ordered, but non-periodic, systems have been investigated and exhibit features which can be directly described by means of a construction which the authors call the shift lattice , which is a simple generalization of the concept of the lattice. This paper is devoted to a description of the properties of the one-dimensional shift lattice and its Fourier transform. Its applications to the phases related to L–Ta 2 O 5 and some Bi 2 TeO 5 -related systems are outlined and its relation to the theory of modulated structures and their Fourier transforms is briefly discussed.

scholarly journals On a direct method for the determination of an exact invariant for the time-dependent harmonic oscillator

1977 ◽  
Vol 20 (1) ◽  
pp. 97-105 ◽  
Author(s):  
P. G. L. Leach
Keyword(s):  
Direct Method ◽  
Dynamical Symmetry ◽  
Time Dependent ◽  
Symmetry Groups ◽  
New Approach ◽  
One Dimensional ◽  
The One ◽  
A Direct Method ◽  
Exact Invariant

AbstractAn exact invariant is found for the one-dimensional oscillator with equation of motion . The method used is that of linear canonical transformations with time-dependent coeffcients. This is a new approach to the problem and has the advantage of simplicity. When f(t) and g(t) are zero, the invariant is related to the well-known Lewis invariant. The significance of extension to higher dimension of these results is indicated, in particular for the existence of non-invariance dynamical symmetry groups.

Analysis of Thin-Walled Closed Beams With General Quadrilateral Cross Sections

1999 ◽  
Vol 66 (4) ◽  
pp. 904-912 ◽  
Author(s):  
J. H. Kim ◽  
Y. Y. Kim
Keyword(s):  
Finite Element ◽  
Cross Sections ◽  
Distortion Function ◽  
Element Formulation ◽  
Numerical Work ◽  
New Approach ◽  
One Dimensional ◽  
Static And Dynamic Analysis ◽  
Thin Walled ◽  
The One

This paper deals with the one-dimensional static and dynamic analysis of thin-walled closed beams with general quadrilateral cross sections. The coupled deformations of distortion as well as torsion and warping are investigated in this work. A new approach to determine the functions describing section deformations is proposed. In particular, the present distortion function satisfies all the necessary continuity conditions unlike Vlasov's distortion function. Based on these section deformation functions, a one-dimensional theory dealing with the coupled deformations is presented. The actual numerical work is carried out using two-node C0 finite element formulation. The present one-dimensional results for some static and free-vibration problems are compared with the existing and the plate finite element results.

A New Approach to The Design of Fan Volute Profiles

1996 ◽  
Vol 210 (3) ◽  
pp. 287-294 ◽  
Author(s):  
D T Qi ◽  
M J Pomfret ◽  
K Lam
Keyword(s):  
Design Method ◽  
Operating Conditions ◽  
Flow Angle ◽  
New Approach ◽  
One Dimensional ◽  
Distribution Form ◽  
Flow Parameters ◽  
Centrifugal Fans ◽  
Ultimate Objective ◽  
The One

In traditional volute design methods, the flow at the volute inlet is assumed to be uniform according to one-dimensional theory. However, many experimental results have shown that under the design operating conditions, the actual distributions of the flow parameters at the volute inlet are different from those predicted by the ideal assumption. This is because of the influence of the non-axisymmetrical volute geometry, especially due to the presence of the volute tongue. Based on this fact, a new method was considered whereby different volute profiles can be generated by varying the distributions of flow parameters at the volute inlet. The ultimate objective is to achieve the optimal volute profile design so that the performance of a centrifugal machine can be improved. In this paper, a two-dimensional inverse method has been used to redesign the original volute profiles of two centrifugal fans. The results show that the method presented provides effective improvement to the one-dimensional volute design method. The controlled distribution form of volute inlet flow angle α4 in the redesign is qualitatively discussed and can be used for reference in the volute design and further research.

Sign in / Sign up

Export Citation Format

Share Document