Optimized color transfer method for textile fabrics based on image statistical features

High-fidelity color transfer is of great significance for digital textile design. However, existing methods struggle to maintain both color and texture consistency with the visual perception of real fabrics. This paper introduces an optimized color transfer method for textile fabrics based on image statistical features. The method adjusts the luminance of the target image using the mean and standard deviation to match the texture characteristics of the reference image, effectively preserving the texture details of the fabric. Experiments were evaluated using a combination of visual perception and objective evaluation metrics. The results demonstrate that the proposed method achieves a CIEDE2000 color difference of 0.3663, a cosine similarity of 0.9948, a chi-square distance of 0.2609, contrast and homogeneity of the gray-level co-occurrence matrix of 0.0099 and 0.2609, and an LPIPS score of 0.2609. Furthermore, the method exhibits realistic and natural visual perception. Compared with existing color transfer methods, the proposed method performs better in both quantitative metrics and subjective perception. This method offers a new perspective for digital color transfer technology in textiles.