Understanding Rembrandt: Directed Knowledge Improves Robustness and Evolution of Facial Phenotype Modeling

Abstract

Directed knowledge understanding is a deep knowledge service structure strategy proposed in the face of logically complex solving needs along with the increasing scale of data production. This study proposes an improved method for facial phenotype modelling based on directed knowledge information understanding, which effectively utilizes the information framework constructed by the concept of directed knowledge understanding, and uses Renaissance physiological and anatomical knowledge, medical pathology detection, artwork hyperspectral image data, Netherlandish oil painting genealogy and Rembrandt art feature study as directed regions, and transforms directed knowledge through the classification constraints of emotional understanding into behavioral laws, realize parametric extraction and then encode with coupled solution method to complete the improved embedding of facial feature extraction algorithm. The experimental analysis shows that 1) deep knowledge understanding achieves the compensation of sparse feature localization for the deficiency of expression sensitivity, 2) the calculation of surface curvature after drawing on anatomical knowledge can vividly describe the implicit features of facial phenotype, and 3) the sensitivity of implicit emotion observation of facial objects can be effectively improved in the general technique by virtue of the characteristics of facial feature region texture influenced by physiological indicators, combined with the edge recognition of highlight region. The improved facial modelling process has more humane perceptual habits and enhances the accuracy and robustness of the service domain requirements.