基于超像素的图像语义区域引导的色彩迁移

针对图像色彩处理技术，提出基于超像素的图像色彩迁移方法，其以图像语义区域进行引导，以LAB色彩空间进行映射。首先，采用K-means和SLIC算法对输入图像进行分割；其次，对每一子区域块进行区域协方差处理，获得其二阶语义特征并生成超像素，并利用相似度测量函数构造相似矩阵，对区域块聚类可生成图像超像素；最后，再对图像内语义信息相似的像素基于LAB空间映射，完成色彩迁移。结果显示，该方法具有处理复杂图像能力较高及颜色迁移效果准确的优点。

Color transfer guided by semantic region of image based on super pixel

Aiming at image color processing technology, this paper proposed an image color migration method based on super-pixel. Semantic region of image guided the algorithm, and the algorithm mapped in LAB space. Firstly, the K-means algorithm and the SLIC algorithm segmented the input image, and regional covariance processed each sub region block. In this way, the method obtained the second order semantic feature and generated the super-pixel. Secondly, similarity measurement function constructed similarity matrix. Then, the algorithm generated image super pixels by clustering blocks. Finally, this paper mapped the pixels with similar semantic information in the image based on LAB space and completed color transfer. The results show that the proposed algorithm is very efficient in processing complex images, and achieves accurate color transfer effect.