Rapid-transform based rotation invariant descriptor for texture classification under non-ideal conditions

作者：

Highlights：

• A rotation invariant descriptor is proposed for adapting non-ideal conditions.

• The proposed descriptor is based on Rapid-transform which is shift invariant.

• Feature selection approach is designed to improve the performance of the descriptor.

• More robust descriptor is proposed under noise condition.

摘要

•A rotation invariant descriptor is proposed for adapting non-ideal conditions.•The proposed descriptor is based on Rapid-transform which is shift invariant.•Feature selection approach is designed to improve the performance of the descriptor.•More robust descriptor is proposed under noise condition.

论文关键词：RTRID,Rapid-transform based rotation invariant descriptor,OSF-RTRID,Optimal feature subset of RTRID,V2-RTRID,Version 2 of RTRID,LBP,Local binary patterns,SIFT,Scale-invariant feature transform,RIFT,Rotation invariant feature transform,GMRF,Gaussian Markov random field,CWT,Complex wavelet transform,DWT,Discrete wavelet transform,DFT,Discrete Fourier transform,Texture classification,Descriptor,Rotation invariance,Rapid-transform,Random noise,Spatial multiscale

论文评审过程：Received 2 May 2012, Revised 28 February 2013, Accepted 4 May 2013, Available online 16 May 2013.

论文官网地址：https://doi.org/10.1016/j.patcog.2013.05.003