.

if you are looking for scale / rotation / translation invavriant features, you might find the following interesting :

http://www.whoi.edu/science/AOPE/dept/Publications/050.pdf

.

you can use these links

https://www.researchgate.net/publication/259052445_Automated_detection_of_diabetes_using_higher_order_spectral_features_extracted_from_heart_rate_signals

http://dl.acm.org/citation.cfm?id=2595563

http://content.iospress.com/articles/intelligent-data-analysis/ida00580

Article Automated detection of diabetes using higher order spectral ...

Dear Khamael

the reference [1] is a basic one and gives useful information about Higher Order Spectral . also, reference [2] is an applicable one in 2D domain (image).

I hope be useful. 

[1] Chandran, Vinod, et al. "Pattern recognition using invariants defined from higher order spectra: 2-D image inputs." Image Processing, IEEE Transactions on 6.5 (1997): 703-712.

[2] Landgrebe, David. "Hyperspectral image data analysis." Signal Processing Magazine, IEEE 19.1 (2002): 17-28.