Method and system for monitoring egg-laying performance of caged laying-hens at laying-hen positions

Patent metadata

Field	Value
Publication number	US-11523590-B2
Application number	US-202017607422-A
Country	US
Kind code	B2
Filing date	May 22, 2020
Priority date	Jan 19, 2020
Publication date	Dec 13, 2022
Grant date	Dec 13, 2022

Abstract

A method and system for monitoring the egg-laying performance of caged laying-hens at laying-hen positions are provided. A plurality of coops are disposed side by side, a conveyor belt is disposed below the plurality of coops, and a machine-vision inspection device having a CCD camera and a lighting device is provided on an output end of the conveyor belt. Eggs laid by the hens fall onto the conveyor belt from the bottom of the coops. After all eggs are laid, the conveyor belt is activated to convey the eggs to the machine-vision inspection device, and the CCD camera is triggered to capture an image of a surface of the conveyor belt carrying the eggs.

First claim

What is claimed is:

1. A method for monitoring an egg-laying performance of caged laying-hens at laying-hen positions, comprising:
disposing a plurality of coops side by side, and disposing a conveyor belt below the plurality of coops, wherein a conveying direction of the conveyor belt is along a direction of disposing the plurality of coops side by side, a machine-vision inspection device is secured on an output end of the conveyor belt, a charge-coupled device (CCD) camera and a lighting device are installed in the machine-vision inspection device, both the CCD camera and the lighting device are located above the conveyor belt and face the conveyor belt, a motor drives the conveyor belt to run, and the motor and the CCD camera are connected to a computer;
allowing eggs laid by hens in the plurality of coops to fall onto the conveyor belt from a bottom of the plurality of coops, and
after the hens in the plurality of coops have laid all eggs, activating the conveyor belt, recording a time of activating the conveyor belt, and conveying, by the conveyor belt, the eggs to the machine-vision inspection device; and
after the eggs enter the machine-vision inspection device, triggering the CCD camera to capture an image of a surface of the conveyor belt carrying the eggs, obtaining contours of the eggs in the image through image processing, statistically counting a quantity of the eggs, and obtaining a coop position, a color, a size, and a weight for each egg through further analysis and processing, to implement real-time monitoring;
wherein the method further comprises:
when the conveyor belt is activated, starting the CCD camera to collect the image, detecting the contours of the eggs through real-time image analysis and processing, numbering an egg when it is detected that an egg reaches a middle position of the image, and recording a current time t n as an egg arrival time; and
when it is detected that an n th egg passes by, obtaining a coop position x of the n th egg based on the egg arrival time to according to the following formula:

x
=

⌈

v
·

(

t
n

-

t
0

)

L

⌉

wherein v represents a running speed of the conveyor belt, L represents a length of a single coop, t 0 represents an initial running time of the conveyor belt, and Π represents rounding up.

2. The method for monitoring the egg-laying performance of the caged laying-hens at the laying-hen positions according to claim 1 , further comprising:
using the CCD camera to obtain a video stream, extracting each frame as an original image for processing, converting RGB color space of the image into a V component in YUV color space, performing noise reduction on the image, and then performing background segmentation and binarization by an Otsu's method, to determine whether there is an egg: if the image is completely black, determining that there is no egg, or if the image is not completely black, determining that there is an egg;
extracting an Otsu image containing an egg, and performing erosion for b times using an a*a elliptical kernel;
performing a judgement on the eroded image by the following step: if each gray value of a 3*3 central pixel region of the image is not 0, selecting and extracting a bounding rectangle whose gray value is not 0 as a local image region; and performing dilation on the local image region for b times by using the same a*a elliptical kernel, to obtain an intermediate image E; and
obtaining shape and size information of the intermediate image E by the following steps:
processing the intermediate image E by using a minimum bounding rectangle (MBR) method, to establish an MBR of a contour of an egg, using a major axis and a minor axis of the MBR as a major axis and a minor axis of the egg, respectively, to obtain lengths of the major axis and the minor axis of the egg, and then inputting the lengths of the major axis and the minor axis, a perimeter, and an area of the egg into a following egg shape and size-weight estimation model, to obtain a weight of the egg:

W
=

ρ
⁢

⁢

h
2

⁢
l

+

μ
⁡

(

C
-

2
⁢
l

)

⁢
h
⁢
l

+

τ
⁢
S
⁢
h

+
ψ

wherein W represents an estimated weight of the egg; h
represents the length of the minor axis of the egg; 1 represents the length of the major axis of the egg; C represents the obtained perimeter of the egg, S represents the obtained area of the egg; and ρ, μ, τ, and ψ represent model parameters of a first term, a second term, a third term, and a fourth term, respectively.

3. The method for monitoring the egg-laying performance of the caged laying-hens at the laying-hen positions according to claim 2 , comprising: performing pixel-wise multiplication on the intermediate image E and the original image Frame, extracting a region of the egg in the image, obtaining color information of the egg, and then performing color recognition based on the color information of the egg in HSV color space to implement color grading.

4. The method for monitoring the egg-laying performance of the caged laying-hens at the laying-hen positions according to claim 1 , further comprising:
determining normal egg-laying parameters based on a known egg-laying condition, comparing a quantity of eggs at each coop position x, a weight of each egg, and a color of each egg with the normal egg-laying parameters, and then monitoring-an egg-laying performance of laying-hens at the each position, to determine the egg-laying performance and a health status of the laying-hens at the each coop position.

5. The method for monitoring the egg-laying performance of the caged laying-hens at the laying-hen positions according to claim 4 , wherein using the CCD camera to obtain a video stream, extracting each frame as an original image for processing, converting RGB color space of the image into a V component in YUV color space, performing noise reduction on the image, and then performing background segmentation and binarization by an Otsu's method, to determine whether there is an egg: if the imag

Assignees

• Univ Zhejiang

Inventors

• Pan Jinming
• Liu Yifan
• Pan Chenghao

Classifications

• G06M1/272

  using photoelectric means · CPC title

• G01B11/24

  for measuring contours or curvatures · CPC title

• A01K45/00 Primary

  Other aviculture appliances, e.g. devices for determining whether a bird is about to lay · CPC title

• A01K29/005 Primary

  Monitoring or measuring activity · CPC title

• G06M7/00

  Counting of objects carried by a conveyor · CPC title