影像科学与光化学 ›› 2018, Vol. 36 ›› Issue (4): 340-349.DOI: 10.7517/j.issn.1674-0475.180109

• 论文 • 上一篇    下一篇

尿液试纸条的手机图像比色分析新方法的研究

杨任兵1, 程文播2, 钱庆2, 章强2, 白鹏利2, 潘宇骏2, 段鹏2   

  1. 1. 上海大学 通信与信息工程学院, 上海 200444;
    2. 中国科学院 苏州生物医学工程技术研究所, 江苏 苏州 215163
  • 收稿日期:2018-01-23 修回日期:2018-03-20 出版日期:2018-07-15 发布日期:2018-07-15
  • 通讯作者: 程文播
  • 基金资助:
    国家重点研发计划(2016YFC0903502)、中国科学院科技服务网络计划(KFJ-STS-SCYD-007)、吉林省与中国科学院科技合作高技术产业化专项资金项目(2018SYHZ0007)、中国科学院仪器设备研制项目(YJKYYQ20170067)资助

Novel Colorimetric Method for Urine Test Strips Based on Smartphone Image

YANG Renbing1, CHENG Wenbo2, QIAN Qing2, ZHANG Qiang2, BAI Pengli2, PAN Yujun2, DUAN Peng2   

  1. 1. School of Communication & Information Engineering, Shanghai University, Shanghai 200444, P. R. China;
    2. Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou 215163, Jiangsu, P. R. China
  • Received:2018-01-23 Revised:2018-03-20 Online:2018-07-15 Published:2018-07-15

摘要: 针对目前基于手机图像比色算法存在的色差计算不可靠、计算量大等问题,本文提出一种符合人眼视觉的比色分析方法。该方法是将原始的非线性RGB颜色空间的色彩信息转换到符合人眼视觉的CIELab颜色空间,在此空间计算测试颜色和参考颜色之间的色差,可以在参考颜色中准确找到与测试颜色最相近的颜色,并在此基础上创新地提出"垂点法",对颜色的相似程度信息进一步定量,实现了对尿液试纸条的定量分析。实验通过手机拍照获取密闭实验箱中尿液试纸条的图像,定量分析人工尿液中的pH(4.5~9.5)、葡萄糖含量(0~60 mmol/L)和蛋白质含量(0~4 g/L),并使用目测分析法作为对照。结果表明,本文所提出的比色算法不仅准确可靠、重复性好,而且计算量也比较小,非常适合应用在基于手机图像的尿液试纸比色分析中。

关键词: CIELab颜色空间, 比色分析法, 尿液试纸条, 智能手机

Abstract: This paper proposes a new colorimetric analysis algorithm based on smartphone images. Conforming to human eye vision, it has the advantages of minor calculation and best calculating stability. This method converts the color information of the original nonlinear RGB color space to the CIELab color space that conforms to the human vision, therefore the reference color closest to the test color can be accurately picked. Meantime, "vertical point method" is proposed innovatively to further quantify the similarity of color information and realize the quantitative analysis of urine test strips. As for the experiment, the pH (4.59.5), glucose content (060 mmol/L) and protein content (04 g/L) in artificial urine were quantitatively analyzed by taking photos of urine test strips in an ambient light-isolated box, and visual colorimetry was used as a control. The results show that the proposed colorimetric algorithm is not only accurate, reliable, reproducible and computable, but also suitable for application in urine test strip colorimetric analysis based on mobile phone images.

Key words: CIELab color space, colorimetric analysis, urine test strips, smart phone