Tie-dyeing pattern fast-generation method based on deep-learning and digital-image-processing technology

Abstract Contingency and uniqueness are regarded as typical artistic characteristics. To accomplish the realistic effect of each tie-dyeing pattern artwork, we propose a digital fast-generation algorithm based on auxiliary-classifier deep-convolution generative adversarial network (AC-DCGAN) image-processing technology. apply this algorithm, designer first draws planar layout diagram patterns. The consists white background polychrome circles, regional-connectivity is used to extract information all circle positions well categories in diagram. Then AC-DCGAN-generated image color-corrected stitch complete its construction. also then segmented copied area. Mean filtering creates final Results show no obvious color difference generated patterns, splicing edges uniform transition, unique patterns exhibit tie-dye characteristics, achieving effects.