Table 6 Summary of Euclidean distances between clusters, self-reported scores and cluster populations when L*a*b* colour pixels are clustered
From: The BELT and phenoSEED platforms: shape and colour phenotyping of seed samples
Clustering metric | Example A | Example B | Example C |
|---|---|---|---|
K-means cluster | |||
 Distance | 3.58 | 11.75 | 13.52 |
GMM cluster | |||
 Distance | 1.03 | 10.27 | 13.09 |
GMM Average | |||
  Log-likelihood | − 6.90 | − 8.50 | − 8.89 |
K-means cluster population | |||
 (Green/Blue) | 64% / 36% | 51% / 49% | 59% / 41% |
GMM cluster population | |||
 (Green/Blue) | 73% / 27% | 53% / 47% | 83% / 17% |