| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 5136562 | Journal of Chromatography B | 2017 | 6 Pages |
•Traditional Chinese medicines (TCM) are a complex system.•Moments are the important type of shape descriptors for the images.•Tchebichef moment was used for the similarity analysis of TCM.•As the comparison, traditional PCA method was also carried out.•The results obtained from Tchebichef moments were more reliable and accurate.
Traditional Chinese medicines have gotten growing attention and the product quality also became uneven. In this paper, Tchebichef image moments coupled with high performance liquid chromatography with photodiode array detector were proposed to the similarity analysis of the fourteen batches of Pudilan Xiaoyan tablets (Chinese medicine complex), which were based on the grayscale images of three dimensional (3D) fingerprint spectra. Hierarchical cluster analysis was used for displaying the classification results. As a comparison, traditional principal component analysis (PCA) method based on the chromatograms under single-fixed wavelength was carried out on the same samples. The analytical results indicated that the proposed approach could provide more accurate and reasonable results owing to its powerful invariant and multi-resolution capability of image moments.
