2026-10-15T06:41:12Z https://data.mendeley.com/oai

oai:data.mendeley.com/4r2m7ynrt2.2 2026-08-27T15:42:37Z

Boukdir , Abdelbasset PatternChroma-41K Mendeley Data PatternChroma-41K is a comprehensive benchmark dataset containing 41,015 annotated pattern images with detailed color and emotional analysis. Each image includes dominant color extraction, 6-color palettes, dimensional metadata (resolution, aspect ratio, orientation), and three-level classification: color names (12 categories), emotions (12 categories), and contextual tags (111 descriptors). The dataset provides quantitative metrics including brightness (μ=0.501, σ=0.200), saturation (μ=0.479, σ=0.233), and contrast (μ=0.158, σ=0.078), enabling systematic analysis of color-emotion relationships in visual patterns. Color distribution is led by BLUE (17.94%), PURPLE (14.20%), and ORANGE (12.57%), while emotion classification shows prevalence of Comfort (16.91%), Imagination (16.28%), and Calm (12.30%). Contextual tags are headed by Dreamy (10.58%), Melancholic (6.65%), and Warm (5.37%). This benchmark supports multiple research domains: computational design, color psychology, affective computing, pattern recognition, textile and fashion analysis, and digital humanities. PatternChroma-41K addresses the gap in large-scale, multi-dimensional pattern datasets and provides a foundation for developing color-aware design systems, emotion recognition models, and pattern generation algorithms. All images are provided with complete metadata including RGB/Hex values, dimensional properties, and contextual descriptors. Computer Science Artificial Intelligence Computer Vision Image Processing Multimedia Machine Learning Pattern Recognition Deep Learning Dataset doi:10.17632/4r2m7ynrt2.2 oai:data.mendeley.com/4r2m7ynrt2.2 Creative Commons Attribution 4.0 International http://creativecommons.org/licenses/by/4.0 https://data.mendeley.com/datasets/4r2m7ynrt2 2026-08-27T15:42:37Z