基于人工免疫的彩色图像颜色约减算法

Color reduction algorithm of color image based on artificial immune

针对真彩色图像中颜色种类多、信息冗余量大的问题,通过分析生物免疫与自学习分类的相似性,提出一种基于人工免疫的彩色图像颜色约减算法。通过人工免疫的自学习颜色分类约减,将原彩色图像的颜色约减到较少的数目,便于后续的目标定位、图像分割和识别。实验结果表明,本方法具有良好的颜色约减性能,且与传统的颜色量化方法相比,本方法具有较低的量化误差。

A self-learning color reduction algorithm of color image is proposed for the large amount of color and redundant information in true color image through similarity analysis of biological immunity and self-learning classification. Through self-learning color reduction, the color of original image can be reduced for the applications, such as object localization, image segmentation and recognition. Experiment results demonstrate that the proposed algorithm has good performance on color reduction and low quantization error compared with traditional color quantization algo-rithm.