| 摘要: |
| 针对L*a*b*值-CMYK值转换的非线性特性，提出了一种基于改进BP神经网络的非线性转换方法。运用神经网络特有的非线性映射能力和不需要精确数学模型的分析能力，实现了这一非线性转换。实验结果表明，建立的网络模型具有较好的精度，能够满足色彩空间的转换要求。 |
| 关键词: 色彩空间 神经网络 非线性 |
| DOI： |
| 分类号:TS801.3 |
| 基金项目:山东省高校科技计划项目（J10LD09） |
|
| Study of L*a*b*-CMYK Color Space Conversion Based on Neural Network |
|
GAO Jun, FAN Zeng-hua
|
|
Shandong University of Technology, Zibo 255049, China
|
| Abstract: |
| A new method of non-linear conversion based on improved BP neural network was presented in view of the non-inear characteristic of L*a*b*-CMYK value. The non-linear conversion was achieved with the non-linear mapping ability of neural network and the analysis ability not requiring precise mathematical model. The result of experiment showed that the network model has good accuracy and can meet the requirements of color space conversion. |
| Key words: color space neural network non-linear |
