CLASSIFICATION OF PLANT SOMATIC EMBRYOS BY USING NEURAL-NETWORK CLASSIFIERS

The development of somatic embryos is characterized by a series of mor phological changes. Quantitative kinetic studies have been hampered by the difficulties in enumerating and characterizing embryo populations . By employing neural networks, we have developed a pattern-recognitio n system for characterizing the morphological features of carrot somat ic embryos. This pattern-recognition system employs a hierarchical dec ision tree to achieve optimal classification. It successfully classifi ed carrot somatic embryos into normal and abnormal embryo classes. For normal embryo classes (globular, oblong, heart, and torpedo embryos), an accuracy of 90% or higher was achieved. The features identified by the neural network classifiers as most important for embryo classific ation are almost identical to those obtained by the branch-and-bound s earching algorithm used previously. However, employing the neural netw orks shortens the system developing time greatly. Coupled with an imag e analysis system, this neural-network-based pattern-recognition syste m shows great potential in embryo sorting and automation of synthetic seed production.