Cage rearing dead chicken position detection method, system and equipment and storage medium

By installing an image acquisition device on the manure cleaning belt and identifying abnormal areas, combined with the speed data of the manure cleaning belt, the problem of missing dead chickens being blocked is solved, and efficient and accurate detection and processing of dead chickens is achieved. .

CN118968406APending Publication Date: 2024-11-15CHINA AGRI UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411011618.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-26
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

In large-scale laying hen farms, dead chickens are not easily discovered because they are blocked, resulting in omissions and environmental pollution. It is difficult for the existing technology to achieve efficient and accurate detection of dead chicken locations.

Method used

By installing an image acquisition device on the feces cleaning tape, the images of the feces cleaning tape are obtained and preprocessed, abnormal areas are identified, and the location of the dead chicken is traced based on the speed data of the feces cleaning tape, and the staff are notified to pick it up in time through the alarm system.

Benefits of technology

The accurate, rapid positioning and timely processing of dead chickens is achieved, and the omission of dead chickens is avoided due to being blocked is improved, the accuracy and efficiency of detection is improved, and the labor intensity of manual inspection is reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118968406A_ABST
    Figure CN118968406A_ABST
Patent Text Reader

Abstract

The invention relates to a cage rearing dead chicken position detection method, system and device, and a storage medium. The method comprises the following steps: obtaining an excrement cleaning zone image; preprocessing the excrement cleaning zone image to obtain a to-be-recognized image sequence; identifying an abnormal region in the to-be-identified image sequence through an abnormal region identification algorithm; based on the abnormal area, tracing the dead chicken position by combining the rotating speed data of the manure cleaning belt; and giving an alarm based on a dead chicken position traceability result, so that a worker can pick up dead chickens at fixed points in time. According to the invention, the dead chicken position is traced by identifying the abnormal area below the dead chicken on the manure cleaning belt image, so that long-time dead chicken leaving caused by chicken shielding and artificial omission during daily dead chicken inspection is avoided, the artificial inspection time is shortened, the labor intensity is reduced, and the automatic management level of a chicken house is improved. The method can be widely applied to the technical field of poultry breeding.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a method, system, equipment and storage medium for detecting positions of dead chickens in cages, belonging to the technical field of poultry breeding. Background Art

[0002] At present, the large-scale laying hen farming in my country is still dominated by traditional cage farming. According to statistics, in large-scale farms, about one to two ten thousandths of chickens die every day. In the dense chicken farming environment, if the dead chickens are not handled in time, they will rot or even autolyze, which may cause the growth of bacteria, the spread of diseases, and air pollution in the house, becoming the source of disease transmission, posing a serious threat to the health of the entire flock, and even causing infectious diseases in the flock, causing major economic losses and threatening human health. Therefore, timely removal of dead chickens is an important daily task in large-scale farms. At present, most domestic poultry farms still adopt manual methods in the inspection of dead chickens. Especially for some large chicken houses with a single breeding capacity of more than 100,000 chickens, breeders need to spend several hours every day to inspect dead chickens, and the top chicken cages need to be climbed for inspection, which not only consumes a lot of time and physical strength, but also easily misses dead chickens due to chickens blocking and inspection omissions, resulting in dead chickens left in the chicken cages for a long time, causing environmental safety hazards. In order to solve the problem of picking up dead chickens in actual production, it is necessary to develop an efficient and accurate dead chicken location detection method.

[0003] In the prior art, there are already a large number of methods for detecting the location of dead chickens. One method is to identify the dead chickens by using binocular visible light cameras and thermal infrared cameras carried by patrol robots to identify the movement state and body temperature of each chicken, which reduces labor costs, detects each chicken by double decision, and improves the recognition accuracy. However, due to the high density of caged chickens and the feeding trough on one side of the aisle, the chickens will gather at the feeding trough to feed or lie down to lay eggs most of the time, so the dead chickens are easily blocked by other chickens and are not easily discovered by the patrol cameras in the aisle, resulting in omissions, which in turn causes the dead chickens to stay and rot in the chicken cages for a long time, polluting the environment in the house and affecting the health of live chickens. Another method is to reversely deduce the location of the chicken cage by identifying abnormal chicken manure, but this method needs to make distance marks on each layer of the manure cleaning belt for camera calibration before calculating the chicken cage. The manual marks may be covered by the chicken feces and the process is cumbersome. Another method is to use a CCD camera to collect images of chicken feces on the conveyor belt in sequence, compare the color and consistency of the feces with the known feces-disease relationship to obtain the type of feces, and monitor the health of the chickens in each chicken cage. This technology can accurately determine the health status of poultry, but because the cage traceability method is not precise enough, it is not easy to determine the specific cages where abnormal poultry are located in order to implement targeted solutions. Summary of the invention

[0004] In view of the problems that dead chicken inspection in caged chicken houses requires a lot of manual labor and dead chickens are easily obscured, the purpose of the present invention is to provide a dead chicken location detection method, system, equipment and storage medium to solve the problem that dead chickens are obscured and difficult to pick out, and ensure that the location of chicken cages can be traced conveniently and accurately.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] In a first aspect, the present invention provides a method for detecting the position of dead chickens in cages, comprising the following steps:

[0007] Obtain images of the manure cleaning belt;

[0008] Preprocessing the manure cleaning belt image to obtain an image sequence to be identified;

[0009] Identify abnormal areas in the image sequence to be identified;

[0010] Based on the identified abnormal areas, the dead chicken locations were traced in combination with the manure cleaning belt speed data;

[0011] An alarm is issued based on the tracing results of the dead chickens’ location so that staff can pick up the dead chickens at a designated location in a timely manner.

[0012] Further, the obtaining of the feces cleaning belt image includes:

[0013] An image acquisition device is installed on the tail frame of the cage near the feces outlet on each layer;

[0014] The image acquisition device is started synchronously with the feces cleaning process, and images of the feces cleaning belt are collected in sequence as the feces cleaning belt moves horizontally.

[0015] Furthermore, the preprocessing of the manure cleaning belt image to obtain the image sequence to be identified includes:

[0016] The manure cleaning belt image is segmented and cropped so that each group of chicken cages corresponds to a corresponding cropped image with the same length and width. Each cropped image is numbered according to the sequence number of the chicken cage to obtain a numbered image sequence;

[0017] The time corresponding to each numbered image in the numbered image sequence is obtained through a timing system, and a time label of each numbered image is marked;

[0018] Eliminate the white chicken feathers that fall on the dark chicken manure in the numbered image sequence to obtain the numbered image sequence to be processed;

[0019] Binarization is performed on the numbered image sequence to be processed to obtain a binary image sequence;

[0020] The dilation and erosion operation method is used to fill the small white holes in the binary image sequence to obtain the image sequence to be recognized.

[0021] Further, the step of eliminating the white chicken feathers falling on the dark chicken manure in the numbered image sequence to obtain the numbered image sequence to be processed includes the following steps:

[0022] Convert the numbered image sequence into HSV color space;

[0023] Identify the white feather area and its outline based on the preset white threshold of the target rejection area, while retaining the color of other areas of the manure cleaning belt;

[0024] The color space mean of the preset range outside the white feather area contour is calculated, and the color space of the white feather area and its contour is converted into the mean color to obtain a sequence of numbered images to be processed.

[0025] Further, the binary processing is performed on the sequence of numbered images to be processed to obtain a binary image sequence, including:

[0026] Counting the grayscale histogram of the numbered image sequence to be processed, and obtaining the number of pixels of the grayscale level of the numbered image sequence to be processed;

[0027] A threshold is randomly selected from a preset threshold range, and the intra-class variance of dividing the sequence of numbered images to be processed into foreground and background according to the threshold is calculated based on the number of pixels;

[0028] Repeat the previous step, traverse all thresholds, and determine the optimal threshold based on the minimum intra-class variance;

[0029] The sequence of numbered images to be processed is binarized based on the optimal threshold to obtain a binary image sequence.

[0030] Furthermore, the identification of abnormal areas in the image sequence to be identified includes the following two paths:

[0031] Path 1: Divide the white area in the image sequence to be identified into several white fitting connected domains, and fit each white fitting connected domain into a closed ellipse; set the normal threshold interval of the pixel area of ​​the white fitting connected domain and the normal threshold interval of the major and minor axes of the closed fitting ellipse, identify and calculate the pixel area of ​​the white fitting connected domain in the image sequence to be identified and the major and minor axes of its closed fitting ellipse, and compare them with the threshold interval to screen out abnormal images;

[0032] Path 2: Segment the image sequence to be identified into 10*10 grids, calculate the pixel average, standard deviation, and threshold range in each grid of the image to be identified corresponding to the previous several consecutive manure cleanings in the same chicken coop, compare the pixel average of each grid of the image to be identified this time with the normal threshold range, and screen out abnormal images.

[0033] In a second aspect, the present invention provides a caged dead chicken position detection system, comprising:

[0034] An image acquisition device, used for acquiring an image of the manure cleaning belt;

[0035] An image preprocessing module is used to preprocess the manure cleaning belt image to obtain an image sequence to be identified;

[0036] An abnormal region recognition module is used to recognize abnormal regions in the image sequence to be recognized;

[0037] The dead chicken location tracing module is used to trace the dead chicken location based on the identified abnormal area and the manure cleaning belt speed data;

[0038] The display terminal is used to issue an alarm based on the tracing results of the dead chickens’ location, so that the staff can pick up the dead chickens in a timely manner.

[0039] Furthermore, the image acquisition device includes a camera, a fill light module and a control module;

[0040] The camera is arranged on the tail frame of the cage at the manure inlet and outlet end of each layer, the fill light module is arranged beside the camera, and a photosensitive module is integrated in the fill light module, the photosensitive module is used to sense the actual light environment brightness at the manure outlet of the chicken cage and send it to the control module, and the control module adjusts the light brightness of the fill light module according to the actual light environment brightness;

[0041] The control module is also used to monitor the start and stop of the manure cleaning belt in real time through a motion detection algorithm. When the manure cleaning belt starts to move horizontally, the camera is synchronously turned on to collect images. When the manure cleaning belt stops moving horizontally, the camera is synchronously turned off to obtain an image of the manure cleaning belt.

[0042] In a third aspect, the present invention provides a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions, which, when executed by a computing device, cause the computing device to perform any method.

[0043] In a fourth aspect, the present invention provides a computing device, comprising: one or more processors and a memory, wherein the memory stores one or more programs and is configured to be executed by the one or more processors, and the one or more programs include instructions for executing any method.

[0044] The present invention adopts the above technical solution, which has the following advantages:

[0045] 1. The present invention collects and identifies the distribution of chicken manure on the manure cleaning belt of each layer of chicken cages and traces the chicken cage where the dead chicken is located according to the time label of the abnormal image and the running speed of the manure cleaning belt collected by the speed sensor, thereby avoiding the situation where the dead chicken is missed due to being blocked, and at the same time, the chicken cage where the dead chicken is located can be accurately and quickly located.

[0046] 2. Since the image acquisition system of the present invention is equipped with a supplementary light device, the brightness between the feeding layers in a poor light environment is improved, and the collected manure cleaning belt image has less noise and higher definition.

[0047] 3. Since the fill light device of the present invention is equipped with a photosensitive module, in a chicken house with different lighting environments, it avoids the problem of overexposure or underexposure of the manure belt image caused by too high or too low fill light brightness, which affects the subsequent image processing and recognition effects.

[0048] 4. Since the present invention is provided with a motion detection algorithm, the image acquisition system is turned on when the feces are cleaned, and the image acquisition system is turned off when the feces are cleaned, thereby saving power energy consumption and storage space.

[0049] 5. Since the present invention is provided with a white feather elimination algorithm, the white chicken feathers falling on the chicken manure of the manure cleaning belt in actual production are further eliminated on the basis of image preprocessing operations such as expansion corrosion, thereby reducing the abnormal image recognition error and improving the accuracy of subsequent recognition.

[0050] 6. Since the abnormal area recognition algorithm of the present invention includes two paths: abnormal white fitting connected domain recognition and gridded local recognition, corresponding to the two situations in actual production where dead chickens are located above a dense chicken manure area and above a sparse chicken manure area, the two algorithm paths are recognized simultaneously, thereby improving the recognition range and recognition accuracy.

[0051] 7. Since the present invention calculates the abnormal image offset distance by obtaining the real-time rotation speed of the manure cleaning belt, the calculation error of the abnormal image offset distance caused by uneven speed and abnormal stoppage of the manure cleaning belt during actual operation is reduced, and the actual offset distance of the abnormal image is calculated by definite integral, and the calculation result is more accurate.

[0052] 8. Since the present invention adopts the compiled chicken cage position numbering rules to perform dead chicken position alarm, the alarm information is more concise and clear, which is convenient for the staff to pick up the dead chickens in time according to the dead chicken position alarm information.

[0053] Therefore, the present invention can be widely applied to the technical field of poultry breeding. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Throughout the accompanying drawings, the same reference numerals are used to represent the same components. In the accompanying drawings:

[0055] Figure 1 A flow chart of a method for detecting the position of dead chickens in cages provided by an embodiment of the present invention;

[0056] Figure 2 A schematic diagram of the installation position of the image acquisition system provided by an embodiment of the present invention;

[0057] Figure 3 A flow chart of the steps of preprocessing the feces cleaning belt image provided by an embodiment of the present invention;

[0058] Figure 4 A flow chart of an abnormal region identification algorithm provided in an embodiment of the present invention.

[0059] The reference numerals in the figures are as follows:

[0060] 1. Camera; 2. Cage tail frame; 3. Chicken cage; 4. Manure cleaning belt; 5. Filling light device. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0062] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0063] In some embodiments of the present invention, a method for detecting the position of dead chickens in cages is provided, the method comprising: collecting a manure belt image at the rear end of the manure cleaning frame by an image acquisition device during manure cleaning, dividing the manure belt image into images corresponding to each group of chicken cages, identifying the abnormal area below the dead chicken in the manure belt image, and calculating the position of the chicken cage where the dead chicken is located by using a dead chicken position reverse formula through a speed sensor and a motion detection algorithm installed at the rotating shaft. The present invention traces the position of the dead chicken by identifying the abnormal area below the dead chicken on the manure belt image, avoiding the long-term retention of dead chickens due to chicken occlusion and human omissions during daily inspections of dead chickens, reducing manual inspection time, reducing labor intensity, and improving the level of automated management of chicken houses.

[0064] Correspondingly, in other embodiments of the present invention, a system, device and storage medium for detecting the location of dead chickens in cages are provided.

[0065] Example 1

[0066] like Figure 1 As shown, the present invention provides a method for detecting the position of dead chickens in cages, which comprises the following steps:

[0067] 1) Obtaining an image of the feces cleaning belt;

[0068] 2) Preprocessing the feces cleaning belt image to obtain an image sequence to be identified;

[0069] 3) Identify abnormal regions in the image sequence to be identified through an abnormal region recognition algorithm;

[0070] 4) Based on the abnormal area, the dead chicken location is traced in combination with the speed data of the manure cleaning belt;

[0071] 5) Based on the tracing results of the dead chicken location, an alarm is issued so that the staff can pick up the dead chickens in time.

[0072] Furthermore, the above step 1) includes the following steps:

[0073] 1.1) Install an image acquisition device on the tail frame of the cage near the feces outlet on each floor;

[0074] 1.2) The image acquisition device is synchronously turned on according to the actual manure cleaning process, and the images of the manure cleaning belt are collected in sequence as the manure cleaning belt moves horizontally.

[0075] Further, in the above step 1.1), if Figure 2 As shown, in this embodiment, the image acquisition device includes a camera 01, a fill light module 05 and a control module (not shown in the figure). Among them, the camera 01 is set on the cage tail frame 02 at the manure inlet and outlet end of each layer, and is used to collect the image of the manure cleaning belt under the control of the control module; the fill light module 05 is set on the side of the camera 01, and is used to fill light for the camera 01; the fill light module 05 is also integrated with a photosensitive module, which is used to sense the actual light environment brightness at the manure outlet of the chicken cage and send it to the control module, and the control module adjusts the light brightness of the fill light module 05 according to the actual light environment brightness; the control module is used to monitor the start and stop of the manure cleaning belt 04 in real time through a motion detection algorithm. When the manure cleaning process starts, that is, the manure cleaning belt starts to move horizontally, the camera 01 is turned on synchronously for image acquisition. When the manure cleaning process stops, that is, the manure cleaning belt stops moving horizontally, the camera 01 is turned off synchronously to generate a set of image sequences with time tags.

[0076] Further, in the above step 2), if Figure 3 As shown, preprocessing is performed to obtain a sequence of images to be identified, including the following steps:

[0077] 2.1) Segment and crop the manure cleaning belt image so that each group of chicken cages corresponds to a corresponding cropped image with the same length and width, and label each cropped image according to the sequence number of the chicken cage to obtain a numbered image sequence;

[0078] 2.2) Obtain the time corresponding to each numbered image in the numbered image sequence through the timing system, and mark the time label of each numbered image;

[0079] 2.3) Eliminate the white chicken feathers that fall on the dark chicken manure in each numbered image to reduce the error of subsequent binarization processing and obtain a sequence of numbered images to be processed;

[0080] 2.4) Binarizing the image sequence to be processed to obtain a binary image sequence;

[0081] 2.5) Use the expansion and corrosion operation method to fill the small white holes in the binary image sequence to obtain the image sequence to be identified.

[0082] Furthermore, in the above step 2.3), when the white chicken feathers falling on the dark chicken manure in the numbered image sequence are eliminated, the white feather elimination algorithm is used, which specifically includes the following steps:

[0083] 2.3.1) Convert the numbered image sequence into HSV color space;

[0084] 2.3.2) Identify the white feather area and its outline based on the white threshold of the target rejection area, while retaining the color of other areas of the manure cleaning belt;

[0085] In this embodiment, when the pixel value is greater than the preset white threshold, it is considered to be a white feather area and then removed, and other areas are retained;

[0086] 2.3.4) Calculate the color space mean of a preset range (for example, within 0-1 cm) outside the outline of the white feather area, convert the color space of the white feather area and its outline into the mean color, and obtain a sequence of numbered images to be processed.

[0087] Furthermore, in the above step 2.4), when the binary processing of the numbered image sequence to be processed is performed, it is implemented by the Otsu algorithm, which specifically includes the following steps:

[0088] 2.4.1) Counting the grayscale histogram of the numbered image sequence to be processed, and obtaining the number of pixels of the grayscale level of the numbered image sequence to be processed;

[0089] 2.4.2) randomly selecting a threshold value from a preset threshold range (e.g., 0-255), and calculating the intra-class variance of dividing the sequence of numbered images to be processed into foreground and background according to the threshold value based on the number of pixels;

[0090] The calculation method of intra-class variance is:

[0091] S 2 =W a S2 a +W 2 b S b

[0092] where W a is the proportion of the background in the whole picture under the threshold T, and S 2 a is the background variance, and W 2 b is the proportion of the background in the whole picture under the threshold T, and S 2 b is the foreground variance;

[0093] 2.4.3) Repeat step 2.4.2), traverse all possible thresholds (0 - 255), and determine the optimal threshold according to the minimum within-class variance;

[0094] 2.4.4) Binarize the image sequence to be processed based on the optimal threshold to obtain a binary image sequence.

[0095] Furthermore, in the above step 2.5), when using the dilation and erosion operation processing method to fill the small white holes in the binary image sequence, the dilation and erosion operations can be performed through a kernel array of dilation 1*1 and erosion 3*3 to fill the small white holes and obtain the image sequence to be recognized. Among them, using the kernel array for dilation and erosion operations is a well-known technology to those skilled in the art, and the present invention will not elaborate on this.

[0096] More preferably, it is also possible to perform a process of first dilation and then erosion through a kernel array of dilation 3*3 and erosion 3*3. The obtained image sequence to be processed can better eliminate the small holes in the manure cleaning belt image, reducing the subsequent recognition calculation amount and error of the abnormal area.

[0097] Furthermore, in the above step 3), the abnormal area recognition algorithm screens out abnormal images by recognizing the abnormal blank area under the dead chickens on the manure cleaning belt. Specifically, as Figure 4 shown, it includes the following two paths:

[0098] Path 1: Divide the white areas in the image sequence to be recognized into several white fitting connected domains, and fit each white fitting connected domain into a closed ellipse; set the normal threshold interval of the pixel area of the white fitting connected domain and the normal threshold interval of the major and minor axes of the closed fitting ellipse. Identify and calculate the pixel area s of the white fitting connected domain in the image sequence to be recognized and its major axis a and minor axis b of the closed fitting ellipse, and compare them with the threshold interval (3000 < s < 18000; 20 < a < 80; 5 < b < 50) to screen out abnormal images;

[0099] Path 2: Segment the image sequence to be identified into 10*10 grids, calculate the pixel average, standard deviation, and threshold range in each grid of the image to be identified for the previous several times (for example, 5 times) of manure cleaning in the same chicken coop, compare the pixel average of each grid of the image to be identified with the normal threshold range, and screen out abnormal images.

[0100] More preferably, the path 2 algorithm of the abnormal area identification step uses the grid identification result to compare with the average pixel value, standard deviation, and threshold interval of each grid of the image to be identified corresponding to the previous two consecutive manure cleanings of the same chicken cage. Although this embodiment can reduce the storage space and the amount of calculation, the occasional error of the average pixel value in the grid of the two manure cleanings is relatively large.

[0101] Further, the above step 4) includes the following steps:

[0102] 4.1) Based on the rotation speed of the manure cleaning belt, the cage number Z of the chicken cage where the dead chicken is located closest to the manure outlet is calculated using the following formula:

[0103]

[0104] Where S is the actual offset distance of the abnormal image; D is the horizontal distance between the camera and the chicken coop closest to the manure outlet; k is the longitudinal length of the chicken coop;

[0105] The calculation method of the actual offset distance S of the abnormal image is:

[0106] S=ωRT

[0107] Where, ω is the average speed of the manure cleaning belt shaft, which can be obtained by the speed sensor; T is the time label corresponding to the abnormal image; R is the radius of the manure cleaning shaft;

[0108] Preferably, since the manure cleaning belt may run at a non-uniform speed in actual production, and there may be a period of time when the speed stops at 0 during the manure cleaning process, errors may occur when calculating the offset distance S using the average speed ω and the running time T. Therefore, in this embodiment, the abnormal image offset distance S is calculated using the vt curve to make the calculation more accurate. At this time, the calculation formula is:

[0109]

[0110] In the formula, v is the rotation speed of the manure cleaning belt shaft. ;

[0111] 4.2) Based on the cage number Z of the chicken cage where the dead chicken is located closest to the manure outlet, calculate the number of chicken cage groups X where the dead chicken is located. The calculation formula is:

[0112] X=N-Z+1

[0113] Among them, N is the total number of chicken cages in each row of cages.

[0114] Further, in the above step 4.1),

[0115] The calculation formula for the number of cages Z of the chicken cages where the dead chickens are located closest to the manure outlet is:

[0116]

[0117] Where S is; D is the horizontal distance between the camera and the chicken coop closest to the manure outlet; k is the longitudinal length of the chicken coop.

[0118] Furthermore, in the above step 5), when the location information of the chicken cage where the dead chickens are located in each image acquisition module is transmitted to the display terminal in real time, the dead chicken location information is encoded in the abcd-e format and output to the display terminal, wherein a is the chicken house number, b is the aisle number, c is the column number, d is the layer number, and e is the chicken cage group number.

[0119] Example 2

[0120] The above-mentioned embodiment 1 provides a method for detecting the position of dead chickens in cages. Correspondingly, this embodiment provides a system for detecting the position of dead chickens in cages. The system provided in this embodiment can implement the method for detecting the position of dead chickens in cages of embodiment 1, and the system can be implemented by software, hardware, or a combination of software and hardware. For example, the system may include integrated or separate functional modules or functional units to execute the corresponding steps in each method of embodiment 1. Since the system of this embodiment is basically similar to the method embodiment, the process described in this embodiment is relatively simple, and the relevant parts can refer to the partial description of embodiment 1. The embodiment of the system provided in this embodiment is only schematic.

[0121] The caged dead chicken position detection system provided in this embodiment includes:

[0122] An image acquisition device, used for acquiring an image of the manure cleaning belt;

[0123] An image preprocessing module is used to preprocess the manure cleaning belt image to obtain an image sequence to be identified;

[0124] An abnormal region recognition module is used to recognize abnormal regions in the image sequence to be recognized;

[0125] The dead chicken location tracing module is used to trace the dead chicken location based on the identified abnormal area and the manure cleaning belt speed data;

[0126] The display terminal is used to issue an alarm based on the tracing results of the dead chickens’ location, so that the staff can pick up the dead chickens in a timely manner.

[0127] Example 3

[0128] This embodiment provides a processing device corresponding to the caged dead chicken location detection method provided in this embodiment 1. The processing device can be a processing device for a client, such as a mobile phone, a laptop computer, a tablet computer, a desktop computer, etc., to execute the method of embodiment 1.

[0129] The processing device includes a processor, a memory, a communication interface and a bus, and the processor, the memory and the communication interface are connected through the bus to complete mutual communication. The memory stores a computer program that can be run on the processor, and the processor executes the caged dead chicken location detection method provided in this embodiment 1 when running the computer program.

[0130] Preferably, the memory may be a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory.

[0131] Preferably, the processor may be a central processing unit (CPU), a digital signal processor (DSP) or other general-purpose processors of various types, which are not limited here.

[0132] Example 4

[0133] The method for detecting the location of dead chickens in cages of this embodiment 1 can be specifically implemented as a computer program product, which may include a computer-readable storage medium carrying computer-readable program instructions for executing the method for detecting the location of dead chickens in cages of this embodiment 1.

[0134] Computer readable storage media can be tangible devices that hold and store instructions used by instruction execution devices. Computer readable storage media can be, for example, but not limited to, electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any combination thereof.

[0135] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0136] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0137] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0138] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for detecting the position of dead chickens in cages, characterized in that: The following steps are involved: Obtain images of the manure cleaning belt; Preprocessing the manure cleaning belt image to obtain an image sequence to be identified; Identify abnormal areas in the image sequence to be identified; Based on the identified abnormal areas, the dead chicken locations were traced in combination with the manure cleaning belt speed data; An alarm is issued based on the tracing results of the dead chickens’ location so that staff can pick up the dead chickens at a designated location in a timely manner.

2. A method for detecting the position of dead chickens in cages as claimed in claim 1, characterized in that: The step of obtaining the feces cleaning belt image comprises: An image acquisition device is installed on the tail frame of the cage near the feces outlet on each layer; The image acquisition device is started synchronously with the feces cleaning process, and images of the feces cleaning belt are collected in sequence as the feces cleaning belt moves horizontally.

3. A method for detecting the position of dead chickens in cages as claimed in claim 1, characterized in that: The method of preprocessing the feces cleaning belt image to obtain a sequence of images to be identified includes: The manure cleaning belt image is segmented and cropped so that each group of chicken cages corresponds to a corresponding cropped image with the same length and width. Each cropped image is numbered according to the sequence number of the chicken cage to obtain a numbered image sequence; The time corresponding to each numbered image in the numbered image sequence is obtained through a timing system, and a time label of each numbered image is marked; Eliminate the white chicken feathers that fall on the dark chicken manure in the numbered image sequence to obtain the numbered image sequence to be processed; Binarization is performed on the numbered image sequence to be processed to obtain a binary image sequence; The dilation and erosion operation method is used to fill the small white holes in the binary image sequence to obtain the image sequence to be recognized.

4. A method for detecting the position of dead chickens in cages as claimed in claim 3, characterized in that: The method of eliminating white chicken feathers falling on dark chicken manure in the numbered image sequence to obtain the numbered image sequence to be processed comprises the following steps: Convert the numbered image sequence into HSV color space; Identify the white feather area and its outline based on the preset white threshold of the target rejection area, while retaining the color of other areas of the manure cleaning belt; The color space mean of the preset range outside the white feather area contour is calculated, and the color space of the white feather area and its contour is converted into the mean color to obtain a sequence of numbered images to be processed.

5. A method for detecting the position of dead chickens in cages as claimed in claim 3, characterized in that: The step of performing binarization processing on the sequence of numbered images to be processed to obtain a binary image sequence comprises: Counting the grayscale histogram of the numbered image sequence to be processed, and obtaining the number of pixels of the grayscale level of the numbered image sequence to be processed; A threshold is randomly selected from a preset threshold range, and the intra-class variance of dividing the sequence of numbered images to be processed into foreground and background according to the threshold is calculated based on the number of pixels; Repeat the previous step, traverse all thresholds, and determine the optimal threshold based on the minimum intra-class variance; The sequence of numbered images to be processed is binarized based on the optimal threshold to obtain a binary image sequence.

6. A method for detecting the position of dead chickens in cages as claimed in claim 1, characterized in that: The identification of abnormal regions in the image sequence to be identified includes the following two paths: Path 1: Divide the white area in the image sequence to be identified into several white fitting connected domains, and fit each white fitting connected domain into a closed ellipse; set the normal threshold interval of the pixel area of ​​the white fitting connected domain and the normal threshold interval of the major and minor axes of the closed fitting ellipse, identify and calculate the pixel area of ​​the white fitting connected domain in the image sequence to be identified and the major and minor axes of its closed fitting ellipse, and compare them with the threshold interval to screen out abnormal images; Path 2: Segment the image sequence to be identified into 10*10 grids, calculate the pixel average, standard deviation, and threshold range in each grid of the image to be identified corresponding to the previous several consecutive manure cleanings in the same chicken coop, compare the pixel average of each grid of the image to be identified this time with the normal threshold range, and screen out abnormal images.

7. A caged dead chicken location detection system, characterized in that: include: An image acquisition device, used for acquiring an image of the manure cleaning belt; An image preprocessing module is used to preprocess the manure cleaning belt image to obtain an image sequence to be identified; An abnormal region recognition module is used to recognize abnormal regions in the image sequence to be recognized; The dead chicken location tracing module is used to trace the dead chicken location based on the identified abnormal area and the manure cleaning belt speed data; The display terminal is used to issue an alarm based on the tracing results of the dead chickens’ location, so that the staff can pick up the dead chickens in a timely manner.

8. A caged dead chicken location detection system as claimed in claim 7, characterized in that: The image acquisition device includes a camera, a fill light module and a control module; The camera is arranged on the tail frame of the cage at the manure inlet and outlet end of each layer, the fill light module is arranged beside the camera, and a photosensitive module is integrated in the fill light module, the photosensitive module is used to sense the actual light environment brightness at the manure outlet of the chicken cage and send it to the control module, and the control module adjusts the light brightness of the fill light module according to the actual light environment brightness; The control module is also used to monitor the start and stop of the manure cleaning belt in real time through a motion detection algorithm. When the manure cleaning belt starts to move horizontally, the camera is synchronously turned on to collect images. When the manure cleaning belt stops moving horizontally, the camera is synchronously turned off to obtain an image of the manure cleaning belt.

9. A computer-readable storage medium storing one or more programs, characterized in that: The one or more programs include instructions, which, when executed by a computing device, cause the computing device to perform any one of the methods of claims 1 to 6.

10. A computing device, characterized in that include: One or more processors and a memory, wherein the memory stores one or more programs and is configured to be executed by the one or more processors, wherein the one or more programs include instructions for executing any one of the methods described in claims 1 to 6.