Methods for classification of somatic embryos comprising hyperspectral line imaging
First Claim
Patent Images
1. A method for classifying plant embryos comprising:
- (a) developing a classification model by(i) acquiring reflectance spectral raw data representative of a known embryo type by obtaining a plurality of line images of a plurality of embryos of the known embryo type, wherein the plurality of line images are taken from a plurality of positions along and substantially perpendicular to a desired axis of ones of the plurality of embryos;
(ii) performing a data analysis by applying one or more classification algorithms to the spectral raw data, the data analysis resulting in development of a classification model for classifying plant embryos by embryo type;
(b) acquiring, by a computer, reflectance spectral raw data of a plant embryo of unknown embryo type by obtaining a plurality of line images of the plant embryo of unknown embryo type, wherein the plurality of line images are taken from a plurality of positions along and substantially perpendicular to a corresponding axis of the plant embryo of unknown embryo type; and
(c) applying the developed classification model to the spectral raw data of the plant embryo of unknown embryo type in order to classify the type of plant embryo of unknown embryo type.
1 Assignment
0 Petitions
Accused Products
Abstract
The present invention relates to methods, apparatus, and imaging systems for using near-infrared spectroscopy imaging of plant embryos for classifying plant embryos. In one embodiment, a method is provided for classifying a plant embryo of an unknown type based on near infrared spectroscopy imaging.
26 Citations
13 Claims
-
1. A method for classifying plant embryos comprising:
-
(a) developing a classification model by (i) acquiring reflectance spectral raw data representative of a known embryo type by obtaining a plurality of line images of a plurality of embryos of the known embryo type, wherein the plurality of line images are taken from a plurality of positions along and substantially perpendicular to a desired axis of ones of the plurality of embryos; (ii) performing a data analysis by applying one or more classification algorithms to the spectral raw data, the data analysis resulting in development of a classification model for classifying plant embryos by embryo type; (b) acquiring, by a computer, reflectance spectral raw data of a plant embryo of unknown embryo type by obtaining a plurality of line images of the plant embryo of unknown embryo type, wherein the plurality of line images are taken from a plurality of positions along and substantially perpendicular to a corresponding axis of the plant embryo of unknown embryo type; and (c) applying the developed classification model to the spectral raw data of the plant embryo of unknown embryo type in order to classify the type of plant embryo of unknown embryo type. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
-
10. A method for classifying a plant embryo of an unknown type based on near infrared spectroscopy imaging comprising:
-
(a) illuminating an axis of a plant embryo of unknown type; (b) capturing a plurality of hyperspectral line images from the light reflected off the illuminated plant embryo, wherein the plurality of hyperspectral line images are captured from a plurality of positions along and substantially perpendicular to the axis of the plant embryo of unknown embryo type; (c) splitting the infrared portion of the captured plurality of hyperspectral line images into component wavelengths from 900 to 1680 nm; (d) detecting the split line images with a detector array; (e) outputting the detected images as wavelength-by-position data of the plant embryo of unknown type to at least one of a computer storage device, computer-readable media, or a user interface; and (f) comparing the wavelength-by-position data of the plant embryo of unknown type to wavelength-by-position data of a reference plant embryo of known type to classify the plant embryo. - View Dependent Claims (11, 12, 13)
-
Specification