CN114782421B - Poultry veterinarian auxiliary system based on egg laying abnormality detection - Google Patents

Poultry veterinarian auxiliary system based on egg laying abnormality detection Download PDF

Info

Publication number
CN114782421B
CN114782421B CN202210683005.3A CN202210683005A CN114782421B CN 114782421 B CN114782421 B CN 114782421B CN 202210683005 A CN202210683005 A CN 202210683005A CN 114782421 B CN114782421 B CN 114782421B
Authority
CN
China
Prior art keywords
egg
image
pixel point
area
gray
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202210683005.3A
Other languages
Chinese (zh)
Other versions
CN114782421A (en
Inventor
李玉森
栾海宏
李瑞芳
尚菲
邢俊玲
丁岩岩
刘倩
谢秀云
王永涛
董建华
宋存鑫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong Yukou Poultry Co ltd
Original Assignee
Shandong Yukou Poultry Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shandong Yukou Poultry Co ltd filed Critical Shandong Yukou Poultry Co ltd
Priority to CN202210683005.3A priority Critical patent/CN114782421B/en
Publication of CN114782421A publication Critical patent/CN114782421A/en
Application granted granted Critical
Publication of CN114782421B publication Critical patent/CN114782421B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/13—Edge detection
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/40—Analysis of texture
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/90—Determination of colour characteristics
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
    • Y02A40/70—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in livestock or poultry

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Image Processing (AREA)

Abstract

The invention relates to the technical field of data processing, in particular to a poultry and veterinary auxiliary system based on egg laying abnormality detection. According to the invention, the egg-laying abnormality condition of whether the eggs to be detected are soft-shell eggs can be accurately determined by acquiring the visible light images of the eggs to be detected and carrying out corresponding data processing on the visible light images, so that the whole process does not need human participation, and the egg-laying abnormality detection cost is effectively reduced.

Description

Poultry veterinarian auxiliary system based on egg laying abnormality detection
Technical Field
The invention relates to the technical field of data processing, in particular to a poultry veterinarian auxiliary system based on egg-laying abnormality detection.
Background
The poultry often has abnormal egg laying conditions such as rapid egg laying rate reduction, abnormal egg laying (such as soft-shell eggs, thick-shell eggs, malformed eggs and the like), no egg laying peak period or continuous low egg laying rate and the like in the egg laying period, and economic losses of different degrees are caused to poultry farmers. Therefore, when abnormal egg laying occurs, the abnormal egg laying condition needs to be found in time, and then factors such as the egg laying environment of the poultry and the like are corrected and changed in time, so that the abnormal egg laying condition is reduced, and the economic loss is reduced in time.
The existing egg laying abnormality detection is usually manual detection, but because the egg laying abnormality occurrence probability is not very high and egg laying is possible at night, the cost of manual detection is greatly increased, and therefore an unsupervised egg laying abnormality automatic detection method is urgently needed to reduce the cost of egg laying abnormality detection.
Disclosure of Invention
The invention aims to provide a poultry veterinarian auxiliary system based on egg laying abnormality detection, which is used for solving the problem that the cost for manually detecting egg laying abnormality is high in the prior art.
In order to solve the technical problems, the invention provides a poultry veterinary auxiliary system based on egg laying abnormality detection, which comprises an image acquisition module, an egg laying abnormality detection module and an egg laying abnormality feedback reporting module, wherein the image acquisition module is used for acquiring visible light images of eggs to be detected and sending the visible light images to the egg laying abnormality detection module, the egg laying abnormality detection module is used for receiving the visible light images, carrying out data processing based on the visible light images, determining the abnormality detection result of the eggs to be detected and sending the abnormality detection result to the egg laying abnormality feedback reporting module, and the egg laying abnormality feedback reporting module is used for sending the abnormality detection result to a background or a far end;
the method for determining the abnormal detection result of the egg to be detected comprises the following steps:
according to a visible light image of an egg to be detected, data processing is carried out on the visible light image, and therefore an egg area image is obtained;
performing data processing on the egg area image to obtain an egg area gray image, and determining the actual gradient direction and gradient value of each pixel point in the egg area gray image according to the gray value of each pixel point in the egg area gray image;
determining a reference pixel point in the gray image of the egg area, and determining an ideal gradient direction of each pixel point in the gray image of the egg area relative to the reference pixel point according to the position of each pixel point in the gray image of the egg area and the position of the reference pixel point;
determining a gradient direction threshold corresponding to each pixel point in the gray image of the egg area according to the gray value, the gradient value and the position of each pixel point in the gray image of the egg area and the position of the reference pixel point;
determining a texture feature operator of each pixel point in the gray image of the egg area according to the actual gradient direction of each pixel point in the gray image of the egg area, the ideal gradient direction of each pixel point relative to the reference pixel point and the gradient direction threshold value corresponding to each pixel point, thereby obtaining a texture feature image;
and determining whether the egg to be detected is a soft-shell egg or not according to the texture feature operator of each pixel point in the texture feature image.
Further, the calculation formula for determining the ideal gradient direction of each pixel point in the gray level image of the egg area relative to the reference pixel point is as follows:
Figure 459988DEST_PATH_IMAGE002
wherein,
Figure DEST_PATH_IMAGE003
is the first in the grey scale image of the egg areaiThe ideal gradient direction of each pixel point relative to the reference pixel point,
Figure 580391DEST_PATH_IMAGE004
is the first in the grey scale image of the egg areaiThe vertical coordinate of each pixel point is determined,
Figure 100002_DEST_PATH_IMAGE005
is the first in the grey scale image of the egg areaiThe abscissa of each pixel point is given by its abscissa,
Figure 438756DEST_PATH_IMAGE006
is the ordinate of the reference pixel point in the gray level image of the egg area,
Figure DEST_PATH_IMAGE007
the abscissa of a reference pixel point in the gray level image of the egg area is shown.
Further, the determining a gradient direction threshold corresponding to each pixel point in the gray-scale image of the egg area includes:
determining the distance between each pixel point in the gray image of the egg area and the reference pixel point according to the position of each pixel point in the gray image of the egg area and the position of the reference pixel point, thereby obtaining a first gradient threshold value adjusting factor of each pixel point in the gray image of the egg area;
determining a second gradient threshold adjustment factor of each pixel point in the gray image of the egg area according to the gray value and the gradient value of each pixel point in the gray image of the egg area;
and calculating the gradient direction threshold corresponding to each pixel point in the gray level image of the egg area according to the first gradient threshold adjusting factor and the second gradient threshold adjusting factor of each pixel point in the gray level image of the egg area.
Further, a calculation formula corresponding to the second gradient threshold adjustment factor for determining each pixel point in the gray level image of the egg area is as follows:
Figure DEST_PATH_IMAGE009
wherein,
Figure 896282DEST_PATH_IMAGE010
is the first in the grey scale image of the egg areaiA second gradient threshold adjustment factor for each pixel,
Figure DEST_PATH_IMAGE011
is the first in the grey scale image of the egg areaiThe gray value of each pixel point is calculated,
Figure 789283DEST_PATH_IMAGE012
is the first in the grey scale image of the egg areaiGradient values of the individual pixels.
Further, a calculation formula for calculating the corresponding gradient direction threshold value of each pixel point in the gray level image of the egg area is as follows:
Figure 346166DEST_PATH_IMAGE014
wherein,
Figure DEST_PATH_IMAGE015
is the first in the grey scale image of the egg areaiThe gradient direction threshold corresponding to each pixel point,
Figure 675517DEST_PATH_IMAGE016
is a fixed value in the gradient direction,
Figure DEST_PATH_IMAGE017
is the first in the grey scale image of the egg areaiOf a pixelA first gradient threshold adjustment factor is set by the first gradient threshold adjustment factor,
Figure 218625DEST_PATH_IMAGE010
is the first in the grey scale image of the egg areaiAnd adjusting factors of the second gradient threshold of each pixel point.
Further, the determining a texture feature operator of each pixel point in the gray level image of the egg area includes:
determining the ternary codes of all pixel points in the gray level image of the egg area according to the actual gradient direction of all the pixel points in the gray level image of the egg area, the ideal gradient direction of all the pixel points relative to the reference pixel points and the gradient direction threshold value corresponding to all the pixel points;
and determining texture feature operators of all pixel points in the gray level image of the egg area according to the ternary codes of the eight neighborhood pixel points of all the pixel points in the gray level image of the egg area.
Further, the calculation formula for determining the correspondence of the three-value codes of all pixel points in the gray level image of the egg area is as follows:
Figure 356345DEST_PATH_IMAGE018
wherein,
Figure DEST_PATH_IMAGE019
is the first in the grey scale image of the egg areaiThree-value coding of each pixel point is carried out,
Figure 677605DEST_PATH_IMAGE020
is the first in the grey scale image of the egg areaiThe actual gradient direction of each pixel point is determined,
Figure 166355DEST_PATH_IMAGE003
is the first in the grey scale image of the egg areaiThe ideal gradient direction of each pixel point relative to the reference pixel point,
Figure 71793DEST_PATH_IMAGE015
is the first in the grey scale image of the egg areaiLadder corresponding to each pixel pointA degree direction threshold.
Further, the method further comprises:
and if the eggs to be detected are not soft-shell eggs, performing edge detection on the texture feature images, and if an edge area is detected, determining that the eggs to be detected are pockmarked eggs and the edge area is the pockmarked area.
Further, the method further comprises:
if the eggs to be detected are not soft-shell eggs, constructing a gray level co-occurrence matrix according to the gray level value of each pixel point in the gray level image of the egg area, and determining a contrast description operator and an energy description operator based on the gray level co-occurrence matrix;
comparing the contrast descriptor and the energy descriptor with the contrast descriptor standard value and the energy descriptor standard value respectively to obtain the eggshell roughness of the grey image of the egg area;
performing edge detection on the gray level image of the egg area to obtain an edge profile of the egg area, and performing circular fitting and rectangular fitting on the edge profile of the egg area respectively to obtain a fitted circle and a fitted rectangle;
determining a first shape index and a second shape index of the gray level image of the egg area according to the area of the edge contour of the egg area, the area of the fitting circle and the length and width of the fitting rectangle;
comparing the first shape index and the second shape index of the egg area gray level image with the first shape index standard value and the second shape index standard value respectively to obtain the eggshell malformation degree of the egg area gray level image;
according to the eggshell roughness of the egg area gray level image, whether the egg to be detected is an eggshell rough egg or not is judged, and whether the egg to be detected is an eggshell malformed egg or not is judged according to the eggshell malformation degree of the egg area gray level image.
Further, the calculation formula for determining the correspondence between the first shape index and the second shape index of the gray level image of the egg area is as follows:
Figure 329599DEST_PATH_IMAGE022
wherein,
Figure DEST_PATH_IMAGE023
and B are respectively a first shape index and a second shape index of the gray level image of the egg area,
Figure 87340DEST_PATH_IMAGE024
in order to fit the area of the circle,
Figure DEST_PATH_IMAGE025
is the area of the edge contour of the egg area,
Figure 532228DEST_PATH_IMAGE026
and
Figure DEST_PATH_IMAGE027
respectively the length and width of the fitted rectangle.
The invention has the following beneficial effects: the method comprises the steps of obtaining a visible light image of an egg to be detected, carrying out data processing on the visible light image to obtain an egg region gray image, determining the actual gradient direction of each pixel point in the egg region gray image, the ideal gradient direction of each pixel point relative to a reference pixel point and the gradient direction threshold value corresponding to each pixel point to obtain a texture feature image, and determining whether the egg to be detected is a soft-shell egg or not based on the texture feature image. According to the invention, the visible light image of the egg to be detected is obtained, and the visible light image is subjected to corresponding data processing, so that whether the egg to be detected is a soft-shell egg or not can be accurately determined, the whole process does not need human participation, and the egg laying abnormality detection cost is effectively reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
Fig. 1 is a flow chart of a method implemented by a poultry veterinary assistance system based on egg laying abnormality detection according to an embodiment of the present invention.
Detailed Description
To further explain the technical means and effects of the present invention adopted to achieve the predetermined objects, the following detailed description of the embodiments, structures, features and effects of the technical solutions according to the present invention will be given with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
Specific scenes are as follows: under the condition of breeding poultry to lay eggs, whether the laid eggs are abnormal eggs such as soft-shell eggs, pockmarked eggs, thick-shell eggs, malformed eggs and the like is detected.
To above-mentioned specific scene, this embodiment provides a poultry animal doctor auxiliary system based on egg laying abnormality detects, this system includes the supplementary body device of poultry animal doctor, still including the image acquisition module that connects gradually, egg laying abnormality detection module and egg laying abnormality feedback report module, this image acquisition module is used for acquireing the visible light image of waiting to detect the egg, and send the visible light image for egg laying abnormality detection module, this egg laying abnormality detection module is used for receiving the visible light image, and carry out data processing based on the visible light image, confirm the abnormality detection result of waiting to detect the egg, and send the abnormality detection result for egg laying abnormality feedback report module, egg laying abnormality feedback report module is used for sending the abnormality detection result for backstage or distal end.
The method for processing the data by the egg laying abnormality detection module based on the visible light image and determining the abnormality detection result of the egg to be detected is shown in fig. 1, and comprises the following steps:
step S1: and carrying out data processing on the visible light image according to the visible light image of the egg to be detected, thereby obtaining an egg area image.
At the automated poultry farm, the eggs produced by the poultry are photographed using an image capture module of the poultry veterinary assistance system, such as an industrial camera, to obtain visible light images. It is emphasized that in photographing eggs laid by poultry, an industrial camera is disposed at one side of the egg under a single natural light irradiation, and the egg is photographed from the side, and the side center position of the eggshell is opposite to the center of the camera.
After the visible light image is collected, the eggs in the visible light image are identified by utilizing a semantic segmentation neural network, so that an egg area image is obtained. Since the specific implementation process of identifying eggs in the visible light image by using the semantic segmentation neural network belongs to the prior art, the detailed description is omitted here.
The method comprises the steps of obtaining a visible light image of an egg to be detected, using ambient light, obtaining an image of an egg area, and obtaining a central highlight image of the egg area.
In combination with the illumination characteristics of normal eggs, the eggshell surface is known to be uniformly changing. And the acquired image is gradually lightened from the periphery to the middle of the eggshell, so the gradient direction of the image is intensively directed to the center of the eggshell. When the egg to be detected is a soft-shell egg, the uppermost part of the eggshell is sunken to form a pit, and the situation that the gradient direction of the image is concentrated to the center of the eggshell can be avoided. Therefore, whether the eggs to be detected are soft-shell eggs or not can be identified by analyzing the gradient condition of the pixel points of the gray level images of the egg areas, and the specific implementation process is as follows:
step S2: and performing data processing on the egg area image to obtain an egg area gray image, and determining the actual gradient direction and gradient value of each pixel point in the egg area gray image according to the gray value of each pixel point in the egg area gray image.
And carrying out gray level processing on the egg area image to obtain an egg area gray level image. According to the gray value of each pixel point in the gray image of the egg area, the Sobel operator is used for carrying out gradient calculation on the gray image of the egg area, so that the gradient direction and the gradient value of each pixel point in the gray image of the egg area are obtained, in order to distinguish the gradient direction from the ideal gradient direction of the subsequent pixel point, the gradient direction of each pixel point obtained through the gradient calculation is called as the actual gradient direction.
Step S3: and determining a reference pixel point in the gray image of the egg area, and determining the ideal gradient direction of each pixel point in the gray image of the egg area relative to the reference pixel point according to the position of each pixel point in the gray image of the egg area and the position of the reference pixel point.
For a normal egg, the gradient direction of the image is to be concentrated to the center of the eggshell, and based on the characteristic, a reference pixel point in the center of the eggshell is determined. Because the center position of the side surface of the eggshell is over against the center of the camera when the visible light image is obtained, theoretically, the reference pixel point is the center pixel point of the gray image of the egg area. However, considering that the center of the camera may slightly deviate from the center of the side surface of the egg shell, in this embodiment, the pixel point corresponding to the maximum gray value is determined according to the gray value of each pixel point in the gray image of the egg area, and the pixel point corresponding to the maximum gray value is used as the reference pixel point.
After the reference pixel point in the gray image of the egg area is determined, the ideal gradient direction of each pixel point in the gray image of the egg area relative to the reference pixel point can be determined according to the positions of other pixel points in the image relative to the reference pixel point, and the corresponding calculation formula is as follows:
Figure 885980DEST_PATH_IMAGE028
wherein,
Figure 122926DEST_PATH_IMAGE003
is the first in the grey scale image of the egg areaiThe ideal gradient direction of each pixel point relative to the reference pixel point,
Figure 192513DEST_PATH_IMAGE004
is the first in the grey scale image of the egg areaiThe vertical coordinate of each pixel point is set,
Figure 390276DEST_PATH_IMAGE005
is the first in the grey scale image of the egg areaiThe abscissa of each pixel point is given by its coordinate,
Figure 78878DEST_PATH_IMAGE006
is the ordinate of the reference pixel point in the gray scale image of the egg area,
Figure 45697DEST_PATH_IMAGE007
the abscissa of a reference pixel point in the gray level image of the egg area is shown.
Step S4: and determining a gradient direction threshold corresponding to each pixel point in the gray image of the egg area according to the gray value, the gradient value and the position of each pixel point in the gray image of the egg area and the position of the reference pixel point.
After the actual gradient direction and the ideal gradient direction of each pixel point in the gray image of the egg area are obtained through the steps S2 and S3, the texture feature operator of each pixel point in the gray image of the egg area can be determined through calculating the difference between the actual gradient direction and the ideal gradient direction, so that a texture feature image is obtained, and whether the egg to be detected is a soft-shell egg or not can be identified based on the texture feature image. However, it is considered that, because the egg body to be detected is irregular ellipsoid, even if the egg body is a normal egg, the deviation between the actual gradient direction and the ideal gradient direction of the pixel point may be large, and further, the false detection may occur during the detection. Therefore, in the embodiment, the gradient direction threshold corresponding to each pixel point is determined in a self-adaptive manner by analyzing the gray value and the gradient value of each pixel point and the position of each pixel point relative to the reference pixel point, so that the robustness to illumination change is enhanced, and an accurate texture feature image is finally determined. The specific implementation steps of adaptively determining the gradient direction threshold corresponding to each pixel point comprise:
a first substep: and determining the distance between each pixel point in the gray image of the egg area and the reference pixel point according to the position of each pixel point in the gray image of the egg area and the position of the reference pixel point, thereby obtaining a first gradient threshold value regulating factor of each pixel point in the gray image of the egg area.
By analyzing the gray level image of the egg area, the closer to the edge of the image, the more likely the actual gradient direction of the pixel point accords with the ideal gradient direction, that is, the less likely the deviation between the actual gradient direction and the ideal gradient direction occurs, so that different weights can be given according to the position of the pixel point in the image, and the closer to the edge of the image, the smaller the threshold value of the corresponding gradient direction of the pixel point is. In order to measure the positions of pixel points in an image, the distance between each pixel point in a gray image of an egg area and a reference pixel point is calculated, a first gradient threshold value adjusting factor is determined according to the distance, the first gradient threshold value adjusting factor can be regarded as a weight, and the corresponding calculation formula is as follows:
Figure 551764DEST_PATH_IMAGE030
wherein,
Figure 830299DEST_PATH_IMAGE017
is the first in the grey scale image of the egg areaiA first gradient threshold adjustment factor for each pixel,
Figure 447225DEST_PATH_IMAGE005
and
Figure 612758DEST_PATH_IMAGE004
respectively in the gray scale image of the egg areaiThe abscissa and ordinate of each pixel point,
Figure 414361DEST_PATH_IMAGE007
and
Figure 321137DEST_PATH_IMAGE006
respectively, the abscissa and the ordinate of the reference pixel point.
The second sub-step: and determining a second gradient threshold adjustment factor of each pixel point in the gray image of the egg area according to the gray value and the gradient value of each pixel point in the gray image of the egg area.
Considering that other information cannot be observed due to brightness when detection is carried out in a highlight area, and according to the characteristic that defects occur in soft-shell eggs, the egg defects mainly occur in the upper portion of an egg body, namely the upper portion of an ellipsoid, and the obtained highlight portion is mostly the middle portion of the side face of the ellipsoid when images are shot, so that when the actual gradient direction is compared with the ideal gradient direction, a larger threshold value can be given to the highlight area, namely, the gradient direction difference of the highlight area is ignored. And because the image is judged according to the gradient direction, when the gradient value of the original image is greatly changed, the gradient direction at the position needs to pay more attention, namely, the larger the gradient value of the pixel point in the image is, the smaller the subsequent corresponding threshold value is.
Based on the analysis, determining a second gradient threshold value adjusting factor of each pixel point in the gray image of the egg area according to the gray value and the gradient value of each pixel point in the gray image of the egg area, wherein the corresponding calculation formula is as follows:
Figure DEST_PATH_IMAGE031
wherein,
Figure 820383DEST_PATH_IMAGE010
is the first in the grey scale image of the egg areaiA second gradient threshold adjustment factor for each pixel,
Figure 27373DEST_PATH_IMAGE011
is the first in the grey scale image of the egg areaiEach pixel pointIs measured in a predetermined time period, and the gray value of (b),
Figure 875244DEST_PATH_IMAGE012
is the first in the grey scale image of the egg areaiGradient values of the individual pixels.
The third step: calculating a gradient direction threshold corresponding to each pixel point in the gray level image of the egg area according to the first gradient threshold regulating factor and the second gradient threshold regulating factor of each pixel point in the gray level image of the egg area, wherein the corresponding calculation formula is as follows:
Figure 393950DEST_PATH_IMAGE014
wherein,
Figure 352678DEST_PATH_IMAGE015
is the first in the grey scale image of the egg areaiThe gradient direction threshold corresponding to each pixel point,
Figure 758383DEST_PATH_IMAGE016
the fixed value of the gradient direction essentially means an error value that can be tolerated between the actual gradient direction and the ideal gradient direction without considering the influence of pixel point positions, gray scale and other factors,
Figure 777155DEST_PATH_IMAGE016
can be determined according to experience, or can be determined by calculating the absolute value of the difference between the actual gradient direction and the ideal gradient direction of each pixel point in the gray level image of the egg area and then utilizing the Otsu threshold method according to the absolute values of the differences,
Figure 658523DEST_PATH_IMAGE017
is the first in the grey scale image of the egg areaiA first gradient threshold adjustment factor for each pixel,
Figure 545576DEST_PATH_IMAGE010
is the first in the grey scale image of the egg areaiSecond of each pixel pointA gradient threshold adjustment factor.
Step S5: and determining a texture feature operator of each pixel point in the gray image of the egg area according to the actual gradient direction of each pixel point in the gray image of the egg area, the ideal gradient direction of each pixel point relative to the reference pixel point and the gradient direction threshold value corresponding to each pixel point, thereby obtaining the texture feature image.
Determining a texture characteristic operator of each pixel point by combining the actual gradient direction and the ideal gradient direction of each pixel point in the gray image of the egg area and a gradient direction threshold value determined in a self-adaptive manner, wherein the specific implementation steps comprise:
a first substep: determining the ternary codes of all the pixel points in the gray image of the egg area according to the actual gradient direction of all the pixel points in the gray image of the egg area, the ideal gradient direction of all the pixel points relative to the reference pixel point and the gradient direction threshold value corresponding to all the pixel points, wherein the corresponding calculation formula is as follows:
Figure 196001DEST_PATH_IMAGE032
wherein,
Figure 120094DEST_PATH_IMAGE019
is the first in the grey scale image of the egg areaiThe three-value coding of each pixel point,
Figure 629704DEST_PATH_IMAGE020
is the first in the grey scale image of the egg areaiThe actual gradient direction of each pixel point is determined,
Figure 930235DEST_PATH_IMAGE003
is the first in the grey scale image of the egg areaiThe ideal gradient direction of each pixel point relative to the reference pixel point,
Figure 700745DEST_PATH_IMAGE015
is the first in the grey scale image of the egg areaiAnd the corresponding gradient direction threshold value of each pixel point.
The second sub-step: and determining texture feature operators of all pixel points in the gray level image of the egg area according to the ternary codes of the eight neighborhood pixel points of all the pixel points in the gray level image of the egg area.
For each pixel point in the gray image of the egg area, eight neighborhood pixel points are numbered, namely the eight neighborhood pixel point on the right left side of each pixel point is the first eight neighborhood pixel point, and then the second eight neighborhood pixel point, the third eight neighborhood pixel point, … … and the eighth eight neighborhood pixel point are sequentially obtained according to the clockwise direction. According to the three-value coding of the eight-neighborhood pixel point of the pixel point and the serial number of the eight-neighborhood pixel point, coding is respectively carried out according to a positive value and a negative value, so that a first texture feature operator and a second texture feature operator corresponding to the pixel point are obtained, and formulas corresponding to the first texture feature operator and the second texture feature operator are obtained as follows:
Figure 920374DEST_PATH_IMAGE034
Figure 776335DEST_PATH_IMAGE036
wherein,
Figure DEST_PATH_IMAGE037
and
Figure 224765DEST_PATH_IMAGE038
respectively the first in the grey scale image of the egg areaiA first texture feature operator and a second texture feature operator of each pixel point,
Figure DEST_PATH_IMAGE039
as the first in the grey scale image of the egg areaiThe first of each pixel pointmThree-valued coding of eight neighborhood pixels, and
Figure 584202DEST_PATH_IMAGE040
，
Figure DEST_PATH_IMAGE041
as the first in the grey scale image of the egg areaiThe first of each pixel pointnThree-valued coding of eight neighborhood pixels, and
Figure 709153DEST_PATH_IMAGE042
。
in the process of respectively coding each pixel point in the gray image of the egg area according to positive, negative and positive to obtain a first texture feature operator and a second texture feature operator, three conditions of ternary codes corresponding to each eight-neighborhood pixel point of each pixel point in the gray image of the egg area are provided, namely the ternary codes are-1, 0 or +1, when the first texture feature operator is determined, the ternary codes are obtained only according to eight-neighborhood pixel points with the ternary codes of +1 in 8 eight-neighborhood pixel points, and when the second texture feature operator is determined, the ternary codes are obtained only according to eight-neighborhood pixel points with the ternary codes of-1 in 8 eight-neighborhood pixel points. As shown in Table 1 below, only eight neighborhood pixels with a ternary code of +1 are considered in the positive encoding process, and at this time
Figure 662196DEST_PATH_IMAGE037
=2 2 +2 3 = 12; when negative coding is carried out, only eight neighborhood pixel points with three-value coding as-1 are considered, and at the moment
Figure 570109DEST_PATH_IMAGE038
=2 6 +2 7 =192。
TABLE 1
1 1 0
0 0
0 -1 -1
After a first texture feature operator and a second texture feature operator of each pixel point in the gray level image of the egg area are obtained, two texture feature operator graphs can be obtained, and the two texture feature operator graphs respectively reflect the gradient direction features in the gray level image of the egg area from different angles. And overlapping the pixel values of the same pixel point in the two texture feature operator images, namely overlapping a first texture feature operator and a second texture feature operator of the same pixel point, so as to obtain the texture feature operator of each pixel point in the egg region gray level image, and further obtain the texture feature image of the egg region gray level image.
According to the determination mode of the texture feature operator of each pixel point in the texture feature image, for each pixel point in the texture feature image, when the difference between the actual gradient direction and the ideal gradient direction of any eight-neighborhood pixel point of the pixel point is larger than the gradient direction threshold corresponding to any eight-neighborhood pixel point, the gradient direction difference of any eight-neighborhood pixel point of the pixel point is considered to be larger, and when the number of the gradient direction differences in the eight-neighborhood pixel points of the pixel point is larger, the larger the texture feature operator corresponding to the pixel point is, the brighter the texture feature image is.
Step S6: and determining whether the egg to be detected is a soft-shell egg or not according to the texture feature operator of each pixel point in the texture feature image.
After the texture feature image corresponding to the egg area gray level image is obtained, the texture feature image is analyzed, so that the texture of the dark area part in the image is in accordance with the prediction direction, namely the texture of the area part with the smaller texture feature operator in the image is in accordance with the prediction direction, the gradient direction of the bright area part is in accordance with the texture change of a normal egg, namely the area part with the larger texture feature operator in the image is in accordance with the texture change of the normal egg.
Therefore, the average textural feature operator can be obtained by calculating the average value of the textural feature operators of all the pixel points in the textural feature image
Figure DEST_PATH_IMAGE043
The average texture feature operator may be regarded as an average gray value of the texture feature image, the average gray value of the texture feature image may represent a uniformity degree of the eggshell gray, and when the average gray value is lower, it is regarded that the eggshell gray changes more uniformly. Setting texture feature operator threshold
Figure 908687DEST_PATH_IMAGE044
In this embodiment, the threshold of the texture feature operator is set
Figure DEST_PATH_IMAGE045
Judging the average gray value
Figure 79905DEST_PATH_IMAGE043
And texture feature operator threshold
Figure 785824DEST_PATH_IMAGE044
The size between, if
Figure 966270DEST_PATH_IMAGE046
And judging that the egg to be detected is a soft-shell egg.
Step S7: and if the egg to be detected is not the soft-shell egg, performing data processing on the texture characteristic image, and judging whether the egg to be detected is the spotted egg.
And if the eggs to be detected are not soft-shell eggs, performing edge detection on the texture feature images, and if an edge area is detected, determining that the eggs to be detected are pockmarked eggs and the edge area is the pockmarked area.
As a result of analyzing the texture feature image obtained in step S5, when a pock occurs on the eggshell, the change in the direction of the gray scale gradient in the vicinity of the pock is not uniform, and is reflected in the texture feature image, that is, some areas with high brightness exist. Therefore, the edge detection is performed on the texture feature image, and the obtained edge contour is the pockmark area, which indicates that the egg to be detected is the pockmark egg.
Step S8: if the to-be-detected eggs are not soft-shell eggs, carrying out data processing on the gray level images of the egg areas, and judging whether the to-be-detected eggs are rough eggshells and deformed eggshells.
A first substep: and constructing a gray level co-occurrence matrix according to the gray level value of each pixel point in the gray level image of the egg area, and determining a contrast description operator and an energy description operator based on the gray level co-occurrence matrix.
And constructing a gray level co-occurrence matrix of the gray level image of the egg area, and further determining a contrast descriptor Con and an energy descriptor Asm of the gray level co-occurrence matrix. The contrast description operator Con reflects the definition of the image and the depth of the grooves of the texture, and the energy description operator Asm reflects the uniformity of the gray level distribution of the image and the thickness of the texture. Since the specific process of obtaining the contrast descriptor Con and the energy descriptor Asm of the gray level co-occurrence matrix belongs to the prior art, the detailed description is omitted here.
The second sub-step: comparing the contrast descriptor and the energy descriptor with the contrast descriptor standard value and the energy descriptor standard value respectively to obtain the eggshell roughness of the grey image of the egg area, wherein the corresponding calculation formula is as follows:
Figure 565878DEST_PATH_IMAGE048
wherein,
Figure DEST_PATH_IMAGE049
the roughness of the eggshell of the gray level image of the egg area,
Figure 32632DEST_PATH_IMAGE050
and
Figure DEST_PATH_IMAGE051
respectively a contrast descriptor and an energy descriptor,
Figure 694688DEST_PATH_IMAGE052
and
Figure DEST_PATH_IMAGE053
respectively a contrast descriptor standard value and an energy descriptor standard value,
Figure 413246DEST_PATH_IMAGE052
and
Figure 991995DEST_PATH_IMAGE053
is determined according to the gray level image of the egg area of the normal standard egg of the same type as the egg to be detected.
The third step: and performing edge detection on the gray level image of the egg area to obtain an edge profile of the egg area, and performing circular fitting and rectangular fitting on the edge profile of the egg area respectively to obtain a fitted circle and a fitted rectangle.
And (3) carrying out edge detection on the gray level image of the egg area by using a canny operator to obtain an image edge, and fitting the image edge to obtain an edge profile of the egg area. Carry out circular fitting and rectangle fitting respectively to this egg region edge profile to obtain fitting circular and fitting rectangle, because the concrete realization process of carrying out circular fitting and rectangle fitting belongs to prior art, no longer give unnecessary details here.
The fourth step: determining a first shape index and a second shape index of the gray level image of the egg area according to the area of the edge contour of the egg area, the area of the fitting circle and the length and width of the fitting rectangle, wherein the corresponding calculation formula is as follows:
Figure 770595DEST_PATH_IMAGE054
wherein,
Figure 451106DEST_PATH_IMAGE023
and B are respectively a first shape index and a second shape index of the gray level image of the egg area,
Figure 973354DEST_PATH_IMAGE024
in order to fit the area of the circle,
Figure 547555DEST_PATH_IMAGE025
is the area of the edge contour of the egg area,
Figure 887269DEST_PATH_IMAGE026
and
Figure 648552DEST_PATH_IMAGE027
respectively the length and width of the fitted rectangle.
A fifth step of: comparing the first shape index and the second shape index of the egg area gray level image with the first shape index standard value and the second shape index standard value respectively to obtain the eggshell malformation degree of the egg area gray level image, wherein the corresponding calculation formula is as follows:
Figure 240070DEST_PATH_IMAGE056
wherein,Kthe degree of eggshell deformity of the gray-scale image of the egg area,
Figure DEST_PATH_IMAGE057
and
Figure 747406DEST_PATH_IMAGE058
respectively a first shape index and a second shape index,
Figure DEST_PATH_IMAGE059
and
Figure 726864DEST_PATH_IMAGE060
a first shape index standard value and a second shape index standard value,
Figure 506601DEST_PATH_IMAGE059
and
Figure 246018DEST_PATH_IMAGE060
is determined according to the gray image of the egg area of the normal standard egg of the same type as the egg to be detected.
Degree of eggshell deformity according to gray level image of egg areaKThe calculation formula shows that the egg shell malformation degreeKIs obtained by comparing the difference between the first shape index and the second shape index corresponding to the egg to be detected and the first shape index standard value and the second shape index standard value determined according to the normal standard egg, when the difference is smaller, the degree of the malformation of the eggshell is more obviousKThe smaller the size, the better the shape of the egg to be detected.
The sixth substep: according to the eggshell roughness of the egg area gray level image, whether the egg to be detected is an eggshell rough egg or not is judged, and whether the egg to be detected is an eggshell malformed egg or not is judged according to the eggshell malformation degree of the egg area gray level image.
Presetting a threshold value of the roughness degree of the eggshell
Figure DEST_PATH_IMAGE061
And threshold value of degree of abnormality of eggshell
Figure 122707DEST_PATH_IMAGE062
In the present embodiment arrangement
Figure DEST_PATH_IMAGE063
And is
Figure 679590DEST_PATH_IMAGE062
And = 0.3. The roughness degree of the eggshell of the gray level image of the egg areaCAnd egg shell roughness threshold
Figure 290831DEST_PATH_IMAGE061
Make a comparison if
Figure 224152DEST_PATH_IMAGE064
It can be considered as the examinationAnd if the egg is detected to have the abnormal situation of the rough eggshell, judging that the egg to be detected is the egg with the rough eggshell. The degree of eggshell deformity of the gray level image of the egg areaKAnd threshold of egg shell malformation degree
Figure 96293DEST_PATH_IMAGE062
Make a comparison if
Figure DEST_PATH_IMAGE065
If the egg to be detected has the abnormal situation of the eggshell malformation, the egg to be detected is judged to be the eggshell malformed egg.
When the egg to be detected has one or more egg laying abnormality conditions at the same time through the steps, the egg laying abnormality detection module sends the egg laying abnormality detection result to the egg laying abnormality feedback reporting module, and the egg laying abnormality feedback reporting module reports the egg laying abnormality detection result to the background or the far end in time. And (4) giving out referential abnormal reasons and solutions by combining the specific egg laying abnormal conditions in the background or the far end. If soft shell eggs and thin shell eggs appear, the deficiency of vitamin D3 or the deficiency of calcium and phosphorus in the feed and improper proportion mostly result; the pockmarked egg shell indicates that the poultry oviduct or cloaca is inflamed. According to the abnormal reasons of the abnormal egg laying conditions, the egg laying environment of the poultry is changed in time, and the abnormal egg laying conditions are reduced or eliminated, so that the economic loss caused by the abnormal egg laying conditions can be reduced.
According to the invention, the visible light image of the egg to be detected is obtained, and the data identification and processing are carried out on the visible light image, so that whether the egg to be detected is an abnormal condition such as a soft-shell egg, a pockmarked egg, a rough-eggshell egg, an eggshell deformed egg and the like can be automatically and accurately determined, no human participation is required, and the cost of egg laying abnormality detection is effectively reduced.
It should be noted that: the above-mentioned embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.

Claims (5)

1. The poultry veterinarian auxiliary system based on egg laying abnormality detection is characterized by comprising an image acquisition module, an egg laying abnormality detection module and an egg laying abnormality feedback reporting module, wherein the image acquisition module is used for acquiring visible light images of eggs to be detected and sending the visible light images to the egg laying abnormality detection module, the egg laying abnormality detection module is used for receiving the visible light images, performing data processing based on the visible light images, determining abnormality detection results of the eggs to be detected and sending the abnormality detection results to the egg laying abnormality feedback reporting module, and the egg laying abnormality feedback reporting module is used for sending the abnormality detection results to a background or a far end;
the method for determining the abnormal detection result of the egg to be detected comprises the following steps:
according to the visible light image of the egg to be detected, carrying out data processing on the visible light image so as to obtain an egg area image;
performing data processing on the egg area image to obtain an egg area gray image, and determining the actual gradient direction and gradient value of each pixel point in the egg area gray image according to the gray value of each pixel point in the egg area gray image;
determining reference pixel points in the gray image of the egg area, and determining the ideal gradient direction of each pixel point in the gray image of the egg area relative to the reference pixel points according to the position of each pixel point in the gray image of the egg area and the position of the reference pixel points;
determining a gradient direction threshold corresponding to each pixel point in the gray image of the egg area according to the gray value, the gradient value and the position of each pixel point in the gray image of the egg area and the position of the reference pixel point;
determining a texture feature operator of each pixel point in the gray image of the egg area according to the actual gradient direction of each pixel point in the gray image of the egg area, the ideal gradient direction of each pixel point relative to the reference pixel point and the gradient direction threshold value corresponding to each pixel point, thereby obtaining a texture feature image;
determining whether the egg to be detected is a soft-shell egg or not according to the texture feature operator of each pixel point in the texture feature image;
the determining of the gradient direction threshold corresponding to each pixel point in the gray level image of the egg area comprises the following steps:
determining the distance between each pixel point in the gray image of the egg area and the reference pixel point according to the position of each pixel point in the gray image of the egg area and the position of the reference pixel point, thereby obtaining a first gradient threshold value adjusting factor of each pixel point in the gray image of the egg area;
determining a second gradient threshold adjustment factor of each pixel point in the gray image of the egg area according to the gray value and the gradient value of each pixel point in the gray image of the egg area;
calculating a gradient direction threshold corresponding to each pixel point in the gray level image of the egg area according to the first gradient threshold adjustment factor and the second gradient threshold adjustment factor of each pixel point in the gray level image of the egg area;
determining a calculation formula corresponding to the second gradient threshold adjustment factor of each pixel point in the gray level image of the egg area as follows:
Figure DEST_PATH_IMAGE001
wherein,
Figure 368538DEST_PATH_IMAGE002
is the first in the grey scale image of the egg areaiA second gradient threshold adjustment factor for each pixel,
Figure 438738DEST_PATH_IMAGE003
is the first in the grey scale image of the egg areaiThe gray value of each pixel point is the first gray value in the gray image of the egg areaiGradient values of the individual pixel points;
the calculation formula for calculating the corresponding gradient direction threshold value of each pixel point in the gray level image of the egg area is as follows:
Figure 861760DEST_PATH_IMAGE004
wherein,
Figure DEST_PATH_IMAGE005
is the first in the grey scale image of the egg areaiThe gradient direction threshold corresponding to each pixel point,
Figure 51083DEST_PATH_IMAGE006
is a fixed value in the gradient direction,
Figure 698096DEST_PATH_IMAGE007
is the first in the grey scale image of the egg areaiA first gradient threshold adjustment factor for each pixel,
Figure 992942DEST_PATH_IMAGE002
is the first in the grey scale image of the egg areaiA second gradient threshold adjustment factor for each pixel point;
the determining of the texture feature operator of each pixel point in the gray level image of the egg area comprises the following steps:
determining the ternary codes of all pixel points in the gray level image of the egg area according to the actual gradient direction of all the pixel points in the gray level image of the egg area, the ideal gradient direction of all the pixel points relative to the reference pixel points and the gradient direction threshold value corresponding to all the pixel points;
determining texture feature operators of all pixel points in the gray level image of the egg area according to the ternary codes of the eight neighborhood pixel points of all the pixel points in the gray level image of the egg area;
determining a calculation formula corresponding to the three-value codes of all pixel points in the gray level image of the egg area as follows:
Figure 750814DEST_PATH_IMAGE008
wherein,
Figure 241314DEST_PATH_IMAGE009
is the first in the grey scale image of the egg areaiThe three-value coding of each pixel point,
Figure 324807DEST_PATH_IMAGE010
is the first in the grey scale image of the egg areaiThe actual gradient direction of each pixel point is determined,
Figure 434846DEST_PATH_IMAGE011
is the first in the grey scale image of the egg areaiThe ideal gradient direction of each pixel point relative to the reference pixel point,
Figure 527567DEST_PATH_IMAGE005
is the first in the grey scale image of the egg areaiAnd the corresponding gradient direction threshold value of each pixel point.
2. The poultry veterinary auxiliary system based on egg-laying abnormality detection according to claim 1, wherein the calculation formula for determining the ideal gradient direction of each pixel point in the gray image of the egg area relative to the reference pixel point is:
Figure 672240DEST_PATH_IMAGE012
wherein,
Figure 988952DEST_PATH_IMAGE011
is the first in the grey scale image of the egg areaiThe ideal gradient direction of each pixel point relative to the reference pixel point,
Figure 117445DEST_PATH_IMAGE013
is the first in the grey scale image of the egg areaiThe vertical coordinate of each pixel point is determined,
Figure 294085DEST_PATH_IMAGE014
is the first in the grey scale image of the egg areaiThe abscissa of each pixel point is given by its coordinate,
Figure 558844DEST_PATH_IMAGE015
is the ordinate of the reference pixel point in the gray level image of the egg area,
Figure 780878DEST_PATH_IMAGE016
the abscissa of a reference pixel point in the gray level image of the egg area is shown.
3. The poultry veterinary assistance system based on egg abnormality detection according to claim 1, characterized in that the method further comprises:
and if the eggs to be detected are not soft-shell eggs, performing edge detection on the texture feature images, and if an edge area is detected, determining that the eggs to be detected are pockmarked eggs and the edge area is the pockmarked area.
4. The poultry veterinary assistance system based on egg abnormality detection according to claim 1, characterized in that the method further comprises:
if the eggs to be detected are not soft-shell eggs, constructing a gray level co-occurrence matrix according to the gray level value of each pixel point in the gray level image of the egg area, and determining a contrast description operator and an energy description operator based on the gray level co-occurrence matrix;
comparing the contrast descriptor and the energy descriptor with the contrast descriptor standard value and the energy descriptor standard value respectively to obtain the eggshell roughness of the grey image of the egg area;
performing edge detection on the gray level image of the egg area to obtain an edge profile of the egg area, and performing circular fitting and rectangular fitting on the edge profile of the egg area respectively to obtain a fitted circle and a fitted rectangle;
determining a first shape index and a second shape index of the gray level image of the egg area according to the area of the edge contour of the egg area, the area of the fitting circle and the length and width of the fitting rectangle;
comparing the first shape index and the second shape index of the egg area gray level image with the first shape index standard value and the second shape index standard value respectively to obtain the eggshell malformation degree of the egg area gray level image;
according to the eggshell roughness of the egg area gray level image, whether the egg to be detected is an eggshell rough egg or not is judged, and whether the egg to be detected is an eggshell malformed egg or not is judged according to the eggshell malformation degree of the egg area gray level image.
5. The poultry veterinary assistance system based on egg abnormality detection according to claim 4, wherein the calculation formula for determining the correspondence of the first shape index and the second shape index of the gray scale image of the egg area is:
Figure 865508DEST_PATH_IMAGE017
wherein,
Figure 300032DEST_PATH_IMAGE018
and B are respectively a first shape index and a second shape index of the gray level image of the egg area,
Figure 684877DEST_PATH_IMAGE019
in order to fit the area of the circle,
Figure 343391DEST_PATH_IMAGE020
is the area of the edge contour of the egg area,
Figure 446477DEST_PATH_IMAGE021
and
Figure 583495DEST_PATH_IMAGE022
respectively the length and width of the fitted rectangle.
CN202210683005.3A 2022-06-17 2022-06-17 Poultry veterinarian auxiliary system based on egg laying abnormality detection Active CN114782421B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210683005.3A CN114782421B (en) 2022-06-17 2022-06-17 Poultry veterinarian auxiliary system based on egg laying abnormality detection

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210683005.3A CN114782421B (en) 2022-06-17 2022-06-17 Poultry veterinarian auxiliary system based on egg laying abnormality detection

Publications (2)

Publication Number Publication Date
CN114782421A CN114782421A (en) 2022-07-22
CN114782421B true CN114782421B (en) 2022-08-26

Family

ID=82420924

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210683005.3A Active CN114782421B (en) 2022-06-17 2022-06-17 Poultry veterinarian auxiliary system based on egg laying abnormality detection

Country Status (1)

Country Link
CN (1) CN114782421B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115760884B (en) * 2023-01-06 2023-04-14 山东恩信特种车辆制造有限公司 Semitrailer surface welding slag optimization segmentation method based on image processing
JP7349218B1 (en) * 2023-05-29 2023-09-22 株式会社日本選別化工 Egg surface inspection device
CN117390089B (en) * 2023-11-14 2024-03-19 河北玖兴农牧发展有限公司 Pre-hatching egg fertilization information statistical method and system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000235005A (en) * 1999-02-15 2000-08-29 Nidec Tosok Corp Egg inspection device
CN108776143A (en) * 2018-05-28 2018-11-09 湖北工业大学 A kind of online vision inspection apparatus and method of the small stain of egg eggshell surface
CN110991220A (en) * 2019-10-15 2020-04-10 北京海益同展信息科技有限公司 Egg detection method, egg image processing method, egg detection device, egg image processing device, electronic equipment and storage medium
CN114529802A (en) * 2022-01-26 2022-05-24 扬州大学 Goose egg identification and positioning method and system based on machine vision

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2000235005A (en) * 1999-02-15 2000-08-29 Nidec Tosok Corp Egg inspection device
CN108776143A (en) * 2018-05-28 2018-11-09 湖北工业大学 A kind of online vision inspection apparatus and method of the small stain of egg eggshell surface
CN110991220A (en) * 2019-10-15 2020-04-10 北京海益同展信息科技有限公司 Egg detection method, egg image processing method, egg detection device, egg image processing device, electronic equipment and storage medium
CN114529802A (en) * 2022-01-26 2022-05-24 扬州大学 Goose egg identification and positioning method and system based on machine vision

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
基于Adaboosting_SVM算法的多特征蛋壳裂纹识别;熊利荣等;《华中农业大学学报》;20150315(第02期);全文 *
基于可见/近红外高光谱成像技术的鸡蛋新鲜度无损检测;杨晓玉等;《食品与机械》;20171128(第11期);全文 *
基于短时傅立叶变换的鸡蛋破损检测技术的研究;平建峰等;《传感技术学报》;20090720(第07期);全文 *

Also Published As

Publication number Publication date
CN114782421A (en) 2022-07-22

Similar Documents

Publication Publication Date Title
CN114782421B (en) Poultry veterinarian auxiliary system based on egg laying abnormality detection
CN116109644A (en) Surface defect detection method for copper-aluminum transfer bar
CN116740070A (en) Plastic pipeline appearance defect detection method based on machine vision
CN116310845B (en) Intelligent monitoring system for sewage treatment
CN116337879B (en) Rapid detection method for abrasion defect of cable insulation skin
CN116229335A (en) Livestock and poultry farm environment recognition method based on image data
CN116071363B (en) Automatic change shaped steel intelligent production monitoring system
CN116843692B (en) Regenerated active carbon state detection method based on artificial intelligence
CN116485801A (en) Rubber tube quality online detection method and system based on computer vision
CN115170567A (en) Method for detecting defects of waterproof steel plate for ship
CN117934460B (en) Intelligent detection method for surface defects of insulating plate based on visual detection
CN117893457B (en) PCB intelligent detection method, device and computer equipment
CN117830300B (en) Visual-based gas pipeline appearance quality detection method
CN117218115B (en) Auto part paint surface abnormality detection method
CN117705815B (en) Printing defect detection method based on machine vision
CN116577345B (en) Method and system for detecting number of tabs of lithium battery
CN117078678B (en) Waste silicon wafer shape detection method based on image recognition
CN115861318B (en) Cotton processing production quality detection method
CN117388263A (en) Hardware terminal quality detection method for charging gun
CN117274405A (en) LED lamp working color detection method based on machine vision
CN111126371B (en) Coarse pointer dial reading method based on image processing
CN116205923A (en) Nondestructive testing method for internal defects of automobile hub based on X-RAY
CN116993717A (en) Visual detection method for production quality of heating wire of electronic cigarette
CN115205317B (en) Bridge monitoring photoelectric target image light spot center point extraction method
CN116958134B (en) Plastic film extrusion quality evaluation method based on image processing

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant