Research on Multi-scale Geometric Analysis and Compositional Optimization Methods in the Creation of Haikai Ink Figure Painting

Haikai ink figure paintings are often equipped with localized deformation of the painting form, which brings inconvenience to the feature extraction of the paintings. Based on this, this paper proposes to utilize Contourlet transform to process ink figure images, and the low-frequency and high-frequency energy features in the ink figure paintings are classified by the Contourlet transform. Color space conversion and preprocessing of ink figure paintings, geometric reconstruction of ink figure paintings, expressed as a new composition of ink figure painting creation. Filter the LP filter in Contourlet transform, and test the image fusion effect using objective evaluation indexes, such as information entropy, joint information, mutual information, edge strength, and average gradient. Analyze the advantages of reconstructing images using Contourlet transforms. Apply the image aesthetic evaluation index to aesthetically analyze the ink figure painting based on Contourlet transform. The ink figure paintings processed by Contourlet transform have 0.0661~0.7124, 0.0668~0.5659, 0.0384~0.5657, and 0.0465~0.1782 enhancements in saturation E_S, contrast E_C, sharpness E_R, and composition E_M, respectively.