Marginal distribution covariance model in the multiple wavelet domain for texture representation

作者：

Highlights：

• We proposed MDCM in CDF spaceto model the wavelet coefficients.

• We combine multiple wavelets with MDCM to improve the performance of the model.

• A novel inter-filter dependence in DTCWT domain is developed for image representation.

摘要

•We proposed MDCM in CDF spaceto model the wavelet coefficients.•We combine multiple wavelets with MDCM to improve the performance of the model.•A novel inter-filter dependence in DTCWT domain is developed for image representation.

论文关键词：Texture representation,Covariance matrix,Gabor wavelet,Dual tree complex wavelet,Orthogonal wavelet,Convolutional neural network

论文评审过程：Received 9 September 2018, Revised 7 March 2019, Accepted 4 April 2019, Available online 5 April 2019, Version of Record 8 April 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.04.003