| 摘要: |
| 利用神经网络算法的非线性转换优势,构建了基于 BP 神经网络的颜色空间转换正向和反向仿真模型,提出了数据仿真饱和度优先的方法。 通过训练样本的选取、仿真实验和数据分析,得到了较好的训练效率和转换效果。 仿真结果表明,BP 神经网络适合于颜色空间转换,转换精度较高。 |
| 关键词: 颜色空间转换 神经网络 算法模型 |
| DOI： |
| 分类号:TS801. 3; TS807 |
| 基金项目:国家新闻出版总署数字印刷研究中心开放基金项目资助 |
|
| Research on Color Space Conversion Model Based on Saturation Priority BP Neural Network |
|
LYU Zhao-feng, WANG Xiao-hong
|
|
University of Shanghai for Science and Technology, Shanghai 200093, China
|
| Abstract: |
| Saturation priority method for data simulation was put forward. Forward and reverse simulation model of color space conversion based on BP neural network was constructed using the advantages of neural network in nonlinear conversion. Better train efficiency and conversion effects were obtained through test sample selection, simulation experiment, and data analysis. The simulation results showed that the BP neural network is suitable for color space conversion, and accuracy is higher. |
| Key words: color conversion neural network algorithm model |
