Research on seedling grading of Platycladus orientalis based on fuzzy C-means clustering algorithm improved by genetic sparrow family search optimization

Platycladus orientalis has high ecological and economic value, but the quality of Platycladus orientalis seedlings varies greatly. The original quality grading standard is relatively simple and the grading boundary is not clear. It is of great significance to use the in-situ Electrical Impedance Spectroscopy (EIS) technique for nondestructive testing of Platycladus orientalis roots and the improved fuzzy C-means clustering algorithm for scientific and accurate grading of Platycladus orientalis seedlings combined with morphological parameters.As a result, five parameters were found reflecting the quality of Platycladus orientalis seedlings, including phase angle at 4000Hz and tangent value at 400kHz, which could reflect the quality of roots. An Fuzzy C-means (FCM) clustering algorithm improved by genetic sparrow family search optimization (GSFS-FCM) was proposed. Through the test of standard functions and typical datasets, it was found that the number of clusters of the algorithm could be determined automatically. Although the running time is increased, the number of iterations is reduced by more than 12%, and the clustering error rate is reduced by more than 8%. Finally, eight months old Platycladus orientalis seedlings were divided into six grades with the GSFS-FCM algorithm.From the present study it can be concluded that among the Platycladus orientalis seedlings root EIS parameters measured by the instrument in situ, phase angle at 4000Hz and tangent value at 400kHz can be used to represent the quality of the roots. GSFS-FCM algorithm can effectively and automatically grading the seedlings of Platycladus orientalis without specifying the number of clusters. The algorithm has fewer iterations and higher accuracy.