|
| 基于概率神经网络的尿样颜色识别 |
| Research on color recognition of urine test paper based on probabilistic neural network |
| |
| DOI: |
| 中文关键词: 概率神经网络(PNN) 颜色识别 生化分析 |
| 英文关键词:probabilistic neural network(PNN) color recognition biochemical analysis |
| 基金项目: |
|
| 摘要点击次数: 3680 |
| 全文下载次数: 582 |
| 中文摘要: |
| 基于概率神经网络,提出一种与尿液反应后尿试纸的颜色识别方法。针对颜色色空间转换的非线性复杂关系,获取标准阈值颜色色度值,进行归一化处理后,建立基于概率神经网络的尿样颜色识别模型。实验结果表明,用概率神经网络进行尿样颜色识别是可行而有效的。与颜色色差评价方法作比较,该方法无须进行色空间转换,只利用设备原有RGB颜色空间的RGB值即可实现,更易于操作。 |
| 英文摘要: |
| A color recognition method of urine test paper based on probabilistic neural networks (PNN) is proposed. According to the complicated nonlinear relationship of the color space transformation, the standard threshold value is obtained after normalization. PNN is established for color recognition of urine test paper. The results indicate that this method has high convergence speed and forecast precision as well as high feasibility. A comparison with the color difference evaluation method indicates that the proposed method here does not need color space conversion, using only original RGB color value, which is easy to understand and operate. |
| HTML 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|