Hyperspectral identification of egg fertility and gender

• US 9,435,732 B2
• Filed: 06/27/2010
• Issued: 09/06/2016
• Est. Priority Date: 06/25/2009
• Status: Active Grant

• Current Assignee: Yissum Research Development Company of the Hebrew University of Jerusalem Ltd. (Hebrew University of Jerusalem), Ramot at Tel Aviv University Limited (Tel Aviv University)
• Original Assignee: Yissum Research Development Company of the Hebrew University of Jerusalem Ltd. (Hebrew University of Jerusalem), Ramot at Tel Aviv University Limited (Tel Aviv University)
• Inventors: Rozenboim, Israel, Ben Dor, Eyal
• Primary Examiner(s): Beck, Leron
• Application Number:
  US13/379,443
• Publication Number:
  US 20130044210A1
• Time in Patent Office: 2,263 Days
• Field of Search: 348/135
• US Class Current: 1/1
• CPC Class Codes: G01N 2021/3155 Measuring in two spectral ranges, e.g. UV and visible
• CPC Class Codes: G01N 21/31 Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry G01N21/72 takes precedence
• CPC Class Codes: G01N 21/3563 for analysing solids; Preparation of samples therefor
• CPC Class Codes: G01N 2201/1296 using neural networks
• CPC Class Codes: G01N 33/08 Eggs, e.g. by candling

Abstract

A hyperspectral method for detecting the present condition of an avian egg is disclosed in which a neural network algorithm is used to compare the spectrum of a test egg against a spectral library. The method can detect fertility with greater than 90% reliability on the day of laying and the gender of the chick with greater than 75% reliability on the 12th day after laying.

Claims

1. A non-invasive method for detecting the present condition of an avian egg, comprising the steps of:

obtaining a test measurement with a hyperspectral camera by measuring, at a predefined portion of a field of view of the hyperspectral camera, an amount of light of at least one predetermined wavelength corresponding to a reflectance or transmittance feature of an avian egg;

comparing the test measurement to a reference measurement at substantially the same predetermined wavelength;

confirming presence of an avian egg within the field of view of said camera when a difference between the test measurement and the reference measurement exceeds a predetermined threshold;

obtaining, with the hyperspectral camera, at least one spectrum of the confirmed avian egg over a predetermined wavelength range;

using a neural network algorithm to compare said spectrum with a predefined database of spectra defining possible values of said present condition; and

,

using the results of said comparison to assess the present condition of said avian egg.

2. The method according to claim 1, further comprising a step of discarding any egg for which said present condition does not match a predefined condition.

3. The method according to claim 1, wherein said step of using a neural network algorithm comprises steps of:

using principal component analysis to transform a data set;

selecting the first m scores from each sample; and

,

inputting said m scores into a neural network comprising n neurons, a transfer function in a hidden layer, and a linear transfer function for the output.

4. The method according to claim 3, wherein m=5, n=3, and said transfer function in the hidden layer is the “

tan sig”

function.

5. The method according to claim 1, wherein said step of using a neural network algorithm to compare said spectrum with a predefined database of spectra defining possible values of said present condition further comprises an additional step of preprocessing the spectral data.

6. The method according to claim 5, wherein said additional step of preprocessing the spectral data further comprises a step chosen from the group consisting of (a) performing dimension reduction and (b) performing spectral enhancement.

7. The method of claim 1, further comprising the additional steps of

obtaining at least one reference spectrum;

obtaining at least one calibration spectrum;

obtaining at least one test spectrum;

using said at least one reference spectrum to correct said at least one test spectrum; and

,

using said calibration spectrum to calibrate at least one spectral feature.

8. The method according to claim 7, wherein said step of obtaining a reference spectrum further comprises an additional step chosen from the group consisting of (a) constructing a reference spectrum according to a predetermined physical model and (b) obtaining a reference spectrum by means of physical measurement.

9. The method according to claim 1, wherein said at least one predetermined wavelength is within the calcite absorption band centered at about 2340 nm.

10. The method according to claim 1, further comprising the additional step of placing said avian egg in a sample holder during spectral analysis, said sample holder chosen from the group consisting of:

a sample holder comprising means for excluding ambient light;

means for permitting light from a light source to illuminate said egg; and

, means for permitting at least a portion of the light reflected from said egg to impinge on a detector; and

,

a sample holder comprising a casing;

two substantially parallel rods rotatably mounted within said casing and arranged to support an avian egg placed thereupon;

at least one motor in mechanical connection with said rods, said at least one motor adapted to rotate each of said rods about its longitudinal axis;

light admitting means for admitting light emitted by a light source into said casing, said light admitting means disposed such that at least a portion of said light illuminates an avian egg placed upon said rods; and

means for admitting one end of a fiber optic cable into said casing such that at least a portion of said light admitted via said light admitting means and reflected from said avian egg enters said fiber optic cable.

11. The method according to claim 1, wherein said present condition is gender, and further wherein said determination is made not more than two weeks after the time said egg is settled in the incubator.

12. The method according to claim 1, wherein said at least one spectrum is chosen from the group consisting of (a) a reflectance spectrum and (b) a transmittance spectrum.

13. The method according to claim 1, wherein said spectrum extends from the near-UV to the mid-IR.

14. The method according to claim 1, wherein said spectrum is a derivative spectrum.

15. An apparatus for non-invasively determining the present condition of at least one avian egg, said apparatus comprising:

a sample holder comprising;

an outer casing operable to exclude ambient light;

an inner casing that defines a test volume;

means for introducing at least one fiber optic cable into the interior of said sample holder; and

,

means for separating at least part of the upper portion of said sample holder from the lower portion of said sample holder;

a hyperspectral camera configured to capturing a spectrum from an avian egg placed within said sample holder;

data transfer means for transferring said at least one spectrum to a computing device;

computing means installed on said computing device, adapted to analyze said at least one spectrum according to a predetermined algorithm, the algorithm comprising;

obtaining a test measurement captured by the hyperspectral camera, at a predefined portion of a field of view of the hyperspectral camera, of an amount of light of at least one predetermined wavelength corresponding to a reflectance or transmittance feature of an avian egg;

comparing the test measurement to a reference measurement at substantially the same predetermined wavelength;

confirming presence of an avian egg within the field of view when a difference between the test measurement and the reference measurement exceeds a predetermined threshold;

obtaining from the hyperspectral camera at least one spectrum of the confirmed avian egg over a predetermined wavelength range;

using a neural network algorithm to compare said spectrum with a predefined database of spectra defining possible values of said present condition; and

,

using the results of said comparison to assess the present condition of said avian egg.

16. An apparatus for non-invasively determining the present condition of at least one avian egg, said apparatus comprising:

a sample holder comprising;

a casing;

two substantially parallel rods rotatably mounted within said casing and arranged to support an avian egg placed thereupon;

at least one motor in mechanical connection with said rods, said at least one motor adapted to rotate each of said rods about its longitudinal axis;

light admitting means for admitting light emitted by a light source into said casing, said light admitting means disposed such that at least a portion of said light illuminates an avian egg placed upon said rods; and

,

means for admitting one end of a fiber optic cable into said casing such that at least a portion of said light admitted via said light admitting means and reflected from said avian egg enters said fiber optic cable, wherein rotation of said avian egg enables collection of spectral data from the entire circumference of said avian egg;

a hyperspectral camera configured to capturing a spectrum from an avian egg placed within said sample holder;

data transfer means for transferring said at least one spectrum to a computing device; and

,

computing means installed on said computing device, adapted to analyze said at least one spectrum according to a predetermined algorithm, the algorithm comprising;

obtaining a test measurement captured by the hyperspectral camera, at a predefined portion of a field of view of the hyperspectral camera, of an amount of light of at least one predetermined wavelength corresponding to a reflectance or transmittance feature of an avian egg;

comparing the test measurement to a reference measurement at substantially the same predetermined wavelength;

confirming presence of an avian egg within the field of view when a difference between the test measurement and the reference measurement exceeds a predetermined threshold;

obtaining from the hyperspectral camera at least one spectrum of the confirmed avian egg over a predetermined wavelength range;

using a neural network algorithm to compare said spectrum with a predefined database of spectra defining possible values of said present condition; and

,

using the results of said comparison to assess the present condition of said avian egg.