Developing a color discrimination model for heritage apparel and accessories based on computational algorithms

Heritage apparel and accessories embody the culture, history and aesthetics of certain region or nation, and their colors reflect profound social level, religious belief and lifestyles of the wearers. However, existing studies on their colors are fragmented and inefficient, leading to a limited understanding of their color characteristics and hindering deeper insight into the culture behind. To address these challenges, this study has developed an advanced color discrimination model for heritage apparel and accessories based on computational algorithms and introduce Bonpo apparel and accessories as case study to validate the model. By applying the Fuzzy C-Means (FCM) algorithm, the Elbow method, and Gaussian Mixture Model (GMM), this model systematically captures and discriminates key color characteristics, including color ratios, LAB color space, Euclidean distance, and color co-occurrence frequency. This systematic color discrimination model for heritage apparel and accessories that can be employed for both further studies and new designs.