Appeal 2007-2681
Application 10/680,676
14. The method according to Claim 1 wherein the
quantifiable characteristics comprise conversion potential, resistance to
pathogens, drought resistance, heat resistance, cold resistance, salt tolerance,
preference for light quality, or suitability for long-term storage.
Thus, claim 1 is directed to an automated method for classifying plant
embryos. Digital images of plant embryos having known characteristics are
analyzed and used to develop a “single metric classification model” for
classifying other plant embryos. That is, the known samples are used to
determine a “metric value” of known characteristics. The metric values are
then used to calculate a Lorenz curve, which is used to determine a threshold
value “to arrive at a single metric classification model.”
After a single metric classification model has been generated, it is
applied to digital images of a plant embryo of unknown characteristics, in
order to classify the unknown embryo “according to its presumed
quantifiable characteristics”; i.e., the unknown embryo is classified as
similar to or dissimilar from the embryos in the training set. (Spec. 8: 17-18
(“Unclassified embryos are classified as acceptable or not based on how
close images of the unclassified embryos fit to the classification model
developed from the training set groups.”)).
Claim 14 specifies that the “quantifiable characteristics” can be
conversion potential (i.e., likelihood to germinate; Spec. 7: 13-15);
resistance to pathogens, drought, heat, or cold; salt tolerance; preference for
light quality; or suitability to long-term storage (i.e., storage of the embryos
themselves; Spec. 7: 18). These characteristics are referred to generically as
“plant embryo quality” in the Specification (id. at 7: 15-18).
4
Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 Next
Last modified: September 9, 2013