Enhancement of user mode prediction through texture similarity

Abstract EN

Identifying user mode is one of the difficult psychological problems.

A general framework for predicting user mode is presented in this paper.

It depends on using content based image retrieval (CBIR) that is dependent on texture similarity instead of color coherent similarity.

Theoretically, color coherence vectors (CCVs) will retrieve images with overall similar color distribution.

Mode or attitude change may not affect the color distribution, but any change in background, clothes, makeup and hair style will do.

Systems based on color coherent feature extraction require tremendous offline changes in the image database to compile necessary changes for accurate prediction of user mode.

Consequently, the system presented in this paper is a promising step towards using CBIR based on texture features rather than color features to enhance the prediction of user attitude.

The results exhibit better improvements in defining user mode in comparison with the previous work.