Image classification using a new set of separable two-dimensional discrete orthogonal invariant moments

2014 ◽  
Vol 23 (1) ◽  
pp. 013026 ◽  
Author(s):  
Abdeslam Hmimid ◽  
Mhamed Sayyouri ◽  
Hassan Qjidaa
Keyword(s):  
Image Classification ◽  
Two Dimensional ◽  
Invariant Moments ◽  
Discrete Orthogonal

Fault Diagnosis of Rolling Bearing Based on S Transform and Image of Invariant Moments

2013 ◽  
Vol 706-708 ◽  
pp. 798-802 ◽  
Author(s):  
Jun Feng Guo ◽  
Bao Zeng Wang ◽  
Peng Wei ◽  
Xing Chun Wei
Keyword(s):  
Fault Diagnosis ◽  
Rolling Bearing ◽  
Frequency Function ◽  
Time Signal ◽  
Two Dimensional ◽  
Moment Invariant ◽  
Time Frequency ◽  
Bearing Fault ◽  
Two Dimensional Image ◽  
Invariant Moments

According the characteristics of rolling bearing fault information are nonlinear and non-stationary, the method of time-frequency is often used to make the one dimensional time signal map into two dimensional time and frequency function, and describe the energy density of signal at different times and frequency simultaneously. A method of fault diagnosis based on S transformation and image Hu of invariant moments was put forward in this paper. First of all the measured rolling bearing signals have been S transformed, and time-frequency spectrum which is got is expressed as two dimensional image, then Hu geometric moment invariant of the S transformation spectrum is calculated and the simulation research is carried out using invariant moment principle in image processing. The results show that this method can distinguish the inner ring, outer ring and bearing roller fault intuitively and accurately and measure rolling bearing fault diagnosis efficiently.

Sign in / Sign up

Export Citation Format

Share Document