Machine Vision Techniques for Somatic Coffee Embryo
Morphological Feature Extraction

Z.Cheng and P.P.Ling

Abstract: Machine vision algorithms were developed to examine somatic coffee embryo
morphological features between maturation and germination stages. A skeleton morphological
feature was extracted and recognition rules were established to identify "Y" shaped skeletons.
And improved thinning algorithm was developed to obtain single-pixel-width skeleton that
simplified the recognition task. The "Y" shaped skeleton was found to be a promising feature
to represent torpedo stage embryo. For the 127 embryos tested, the success rate was 73%
in identifying torpedo-stage somatic coffee embryos from the selected morphological feature.