A video monitoring system for feeding grass to beef cattle

Abstract

The present disclosure relates to the field of video monitoring technology, and specifically to a video monitoring system for feeding forage to beef cattle. The technical solution provided by the present disclosure can obtain a feeding video that can reflect the feeding situation of beef cattle by continuously monitoring and shooting the beef cattle feeding area and the forage storage area in the breeding area, and obtain various parameters that characterize the possibility of beef cattle eating, the amount of food eaten, and the degree of willingness to eat by performing image processing and analysis on the display screen in the eating video frame by frame, and calculate the abnormal eating factors of beef cattle in the monitoring video based on the parameters obtained by these image processing. The technical solution provided by the present disclosure supports the continuous detection of abnormal eating factors of beef cattle through monitoring videos, can assist in judging whether beef cattle have abnormal eating conditions and issue corresponding alarm information in a timely manner, effectively ensuring the safety of beef cattle breeding, and has promotional value.

Description

Background Art

[0002] In the process of beef cattle breeding, in order to help farmers better understand the health status of the herd, the use of forage, and promptly discover problems to avoid wasting forage, it is necessary to monitor the eating conditions of beef cattle in real time; because the health status of beef cattle is closely related to its eating conditions, when beef cattle have some health problems, their appetite will decrease. Therefore, in order to monitor the health of beef cattle, it is necessary to use video to monitor the eating conditions of beef cattle in real time.

[0003] In the prior art, farmers usually need to use monitoring software to view the images of each monitoring area to discover abnormal conditions of beef cattle. Since a single monitoring area is large in area and contains a large number of beef cattle, and it is impossible to achieve continuous monitoring 24 hours a day manually, abnormal eating conditions of beef cattle are often not discovered in time. Summary of the invention

[0004] In order to solve the technical problems raised in the background technology, the purpose of the present disclosure is to provide a video monitoring system for feeding forage to beef cattle, and the technical solutions adopted are as follows:

[0005] A first aspect of the present disclosure provides a video monitoring system for feeding forage to beef cattle, and the video monitoring system may specifically include:

[0006] A video monitoring unit is used to continuously monitor and obtain eating videos of beef cattle feeding on forage;

[0007] An image recognition unit, used for performing image recognition based on the eating video to obtain an image of the beef cattle eating;

[0008] A feeding analysis unit, used for performing image analysis on the feeding situation of the beef cattle according to the feeding image of the beef cattle to obtain an image analysis result, wherein the image analysis result includes one or more combinations of the feeding possibility of the beef cattle, the feeding amount of the beef cattle in a period of time, and the feeding willingness of the beef cattle;

[0009] The abnormality monitoring unit is used to perform eating abnormality analysis based on the image analysis results to obtain abnormal factors that characterize the degree of abnormal eating of the beef cattle, and continuously monitor the abnormal factors to obtain the feeding material status of the beef cattle.

[0010] In a possible implementation of the first aspect above, the image content in the eating video includes a complete beef cattle feeding area and a forage storage area.

[0011] In a possible implementation of the first aspect, the image recognition unit further includes:

[0012] Obtain an eating grayscale image corresponding to each frame in the eating video;

[0013] Edge detection is performed on the eating grayscale image to obtain a beef cattle eating image, wherein the beef cattle eating image includes an edge set in the eating grayscale image and a connected domain of each beef cattle in the eating grayscale image.

[0014] In a possible implementation of the first aspect, the image recognition unit further includes:

[0015] In the grayscale image of eating, the edge of the connected domain of each beef cattle is divided into two parts and the ellipse fitting process is performed using the least squares method;

[0016] According to the result of the ellipse fitting process, the portion corresponding to the minimum effect of the ellipse fitting process is taken as the head region of the beef cattle.

[0017] In a possible implementation of the first aspect above, the eating analysis unit includes an eating possibility analysis module;

[0018] The feeding possibility analysis module is used to obtain a first distance between the head area of ​​the beef cattle and the feeding trough, and a second distance between the connected area of ​​the beef cattle and the feeding trough according to the beef cattle feeding image;

[0019] The eating possibility of the beef cattle is calculated based on the first distance and the second distance, and the eating possibility is positively correlated with the ratio of the second distance to the first distance.

[0020] In a possible implementation of the first aspect above, the food intake analysis unit further includes a food intake analysis module;

[0021] The food intake analysis module is used to obtain the change trajectory of the centroid of the head area of ​​the beef cattle within a preset time period based on the beef cattle eating image, and obtain the average value of the beef cattle's eating probability within the preset time period;

[0022] According to the trajectory of the center of mass change of the head area and the average of the possibility of eating, the amount of food eaten by the beef cattle in a preset time period is calculated. The amount of food eaten is related to the product of the moving distance of the center of mass of the head area of ​​the beef cattle between two adjacent frames in the preset time period and the average of the possibility of eating.

[0023] In a possible implementation of the first aspect above, the abnormality monitoring unit includes an eating abnormality analysis module;

[0024] The eating abnormality analysis module is used to cluster the beef cattle based on the eating images and the body shapes of the beef cattle, and obtain the food intake of the beef cattle within a preset time period;

[0025] According to the difference between the feed intake of the beef cattle and the average feed intake of all the beef cattle in the cluster where the beef cattle is located, the abnormal feeding parameter of the beef cattle is obtained. The abnormal feeding parameter of the beef cattle is associated with the ratio between the difference between the feed intake of the beef cattle and the mean feed intake of the cluster and the feed intake variance of the cluster.

[0026] In a possible implementation of the first aspect above, the eating analysis unit includes an eating intention analysis module and a feed delivery analysis module;

[0027] The eating willingness analysis module is used to obtain the distance between the connected domain of each beef cattle and each feed trough within a preset time period according to the beef cattle eating image, so as to form a distance sequence;

[0028] According to the distribution of each minimum point in each distance sequence, the eating willingness of each beef cattle is obtained;

[0029] The feed delivery analysis module is used to calculate and obtain the feed delivery accuracy parameter based on the difference between the beef cattle's willingness to eat and the beef cattle's food intake within a preset time period. The feed delivery accuracy parameter is inversely correlated with the difference between the beef cattle's willingness to eat and the beef cattle's food intake.

[0030] In a possible implementation of the first aspect above, the abnormality monitoring unit further includes an abnormality factor acquisition unit;

[0031] The abnormal factor acquisition unit is used to calculate the abnormal factor according to the feed delivery accuracy parameter and the eating abnormality parameter of the beef cattle, and the abnormal factor is configured as the product of the feed delivery accuracy parameter and the eating abnormality parameter.

[0032] In a possible implementation of the first aspect above, the abnormality monitoring unit further includes a video marking module and an abnormality warning module;

[0033] The video marking module is used to mark the image frames at the corresponding moments in the eating video and generate marking results when the abnormal factors exceed the preset range;

[0034] The abnormal warning module is used to generate corresponding abnormal eating warning information according to the marking results when the number of consecutively marked image frames is greater than a preset threshold.

[0035] Compared with the background technology, the present invention has the following beneficial effects:

[0036] The technical solution provided by the present disclosure can obtain a feeding video that can reflect the feeding situation of beef cattle by continuously monitoring and shooting the beef cattle feeding area and the forage storage area in the breeding area, and obtain various parameters that characterize the possibility of beef cattle eating, the amount of food eaten, and the degree of willingness to eat by performing image processing and analysis on the display screen in the eating video frame by frame, and calculate the abnormal eating factors of beef cattle in the monitoring video based on the parameters obtained by these image processing. The technical solution provided by the present disclosure supports the continuous detection of abnormal eating factors of beef cattle through monitoring videos, can assist in judging whether beef cattle have abnormal eating conditions and issue corresponding alarm information in a timely manner, effectively ensuring the safety of beef cattle breeding, and has promotional value. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 According to an embodiment of the present disclosure, a schematic diagram of the structure of a video monitoring system for feeding forage to beef cattle is provided;

[0039] Figure 2 According to an embodiment of the present disclosure, a detailed structural diagram of an eating analysis unit is provided;

[0040] Figure 3 According to an embodiment of the present disclosure, a detailed structural diagram of an abnormality monitoring unit is provided;

[0041] Figure 4 A schematic structural diagram of a video monitoring device for feeding forage to beef cattle is provided according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0042] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation, structure, features and effects of a video monitoring system for feeding forage for beef cattle proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0043] 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.

[0044] In response to the problems raised in the background technology, the present disclosure provides a video monitoring system for feeding beef cattle with forage, which can monitor the feeding situation of beef cattle based on the continuously shot monitoring video, and actively issue alarms and reminders when the beef cattle have abnormal feeding, and provide targeted medical treatment to the abnormal beef cattle in time. Specifically, in some embodiments of the present disclosure, Figure 1 FIG. 4 shows a schematic diagram of a video monitoring system for feeding grass to beef cattle. Figure 1 As shown, the video surveillance system may specifically include:

[0045] The video monitoring unit 100 is used to continuously monitor and obtain the feeding video of beef cattle feeding forage. The image content in the feeding video may include the complete beef cattle feeding area and the forage storage area, so as to achieve effective monitoring of beef cattle eating and forage consumption. It is understandable that due to the need for 24-hour all-weather continuous monitoring of beef cattle feeding, the selection of video acquisition equipment has the requirements of being suitable for low-light environments and large-scale coverage. Infrared cameras and equipment with night vision functions can be selected. Those skilled in the art can also select other suitable equipment according to actual needs, which is not limited here.

[0046] The image recognition unit 200 is used to perform image recognition based on the eating video to obtain the beef cattle eating image. The specific acquisition of the beef cattle eating image will be described in detail later and will not be repeated here.

[0047] The eating analysis unit 300 is used to perform image analysis on the eating situation of the beef cattle according to the eating image of the beef cattle to obtain image analysis results. Specifically, the image analysis results may include one or more combinations of the eating possibility of the beef cattle, the amount of food consumed by the beef cattle in a period of time, and the eating willingness of the beef cattle, which are not limited here; the specific content and implementation process of the image analysis results will be specifically described later, and will not be repeated here.

[0048] The abnormal monitoring unit 400 is used to perform an abnormal eating analysis based on the image analysis results to obtain abnormal factors that characterize the degree of abnormal eating of beef cattle, and continuously monitor the abnormal factors to obtain the feeding material status of beef cattle. It can be understood that through the above-mentioned functional modules, the video monitoring unit 100 to the abnormal monitoring unit 400, it is possible to monitor the eating situation of beef cattle based on the continuously shot monitoring video, timely discover possible abnormal eating problems of beef cattle and give an alarm. The specific implementation of the above-mentioned functional modules, the video monitoring unit 100 to the abnormal monitoring unit 400, will be further explained below.

[0049] In the above embodiment, further, in a specific application scenario, the image recognition unit 200 can first generate grayscale images frame by frame according to the obtained eating video stream in the process of obtaining the beef cattle eating image, so as to obtain the eating grayscale image corresponding to each frame in the eating video. Here, grayscale conversion of the frame image of the eating video can take into account the grayscale image output by an infrared camera or a camera with night vision function at night or other special scenes, and can facilitate the subsequent use of an edge detection algorithm to identify and extract the beef cattle in the image screen. In the above embodiment, after obtaining the eating grayscale image corresponding to the frame image, the eating grayscale image can be edge detected based on the Canny edge detection algorithm to obtain the beef cattle eating image, wherein the beef cattle eating image can specifically include an edge set in the eating grayscale image and a connected domain of each beef cattle in the eating grayscale image. It can be understood that Canny edge detection is a widely used edge detection algorithm, which aims to extract edges in an image and provide high-quality edge detection results, and is particularly suitable for beef cattle identification and detection required by the present disclosure. Those skilled in the art can also select other suitable edge detection algorithms according to actual needs, which are not limited here.

[0050] In some embodiments of the present disclosure, specifically, Figure 2 FIG. 2 shows a detailed structural diagram of an eating analysis unit. Figure 2 As shown, the eating analysis unit 300 can further include an eating possibility analysis module 310, which can judge and detect whether the beef cattle want to eat at a certain moment according to the beef cattle eating image obtained by the aforementioned functional module image recognition unit 200. The specific implementation of the eating possibility analysis module 310 will be further described below.

[0051] It is understandable that when the beef cattle need to eat, they need to first put their heads close to the feed trough before they can eat through the feed trough; and in the process of eating, the beef cattle will use their mouths to grab the grass and swallow and chew, and the head of the beef cattle will shake slightly during this process. Therefore, the eating possibility of the beef cattle can be obtained according to the distance between the connected domain of the beef cattle and the feed trough obtained in the grayscale image of eating, and then the amount of food eaten by the beef cattle in a certain period of time can be obtained. Specifically, the first distance between the head area of ​​the beef cattle and the feed trough, and the second distance between the connected domain of the beef cattle and the feed trough can be obtained according to the eating image of the beef cattle. In some specific implementations of the present disclosure, the edge of the connected domain of each beef cattle can be divided into two parts in the grayscale image of eating obtained by the aforementioned functional module, and the ellipse fitting process is performed using the least squares method. The shorter part in the case where the fitting effect is the smallest is used as the head area of ​​the beef cattle, so that the first distance between the head area of ​​the beef cattle and the feed trough can be obtained, and the second distance between the connected domain of the beef cattle and the feed trough can also be obtained at the same time, which will not be repeated here.

[0052] In the above embodiment, further, the eating possibility of the beef cattle can be calculated based on the first distance and the second distance based on the following mathematical expression:

[0053] ;

[0054] in, For any beef cattle in the beef cattle feeding image The corresponding connected domain; Characterize the image The eating probability corresponding to the connected domain of beef cattle; The image of beef cattle eating The distance between the connected area of ​​beef cattle and the feed trough, i.e., the second distance; The image of beef cattle eating The distance between the head area and the feed trough in the connected domain of the beef cattle, i.e., the first distance; To obtain the maximum value function, Used to avoid the extreme case where the denominator is 0; is the hyperbolic tangent function. It can be understood that in the above mathematical expression, The larger the value of , the closer the distance between the head area of ​​the beef cattle and the feed trough in the beef cattle eating image is, and the farther the distance between the whole body of the beef cattle and the feed trough is. At this time, the picture represented in the monitoring image can be that the whole beef cattle is in the state where the head area is on the feed trough and the body part is far away from the feed trough, which is consistent with the body distribution of the beef cattle when eating. Therefore, the possibility of the beef cattle eating is greater, that is, the possibility of eating is positively correlated with the ratio of the second distance to the first distance.

[0055] In the above embodiment, further, Figure 2 As shown, the eating analysis unit 300 can further include an eating amount analysis module 320. It is understandable that, based on obtaining the eating possibility of the beef cattle at a certain moment, the eating amount of the beef cattle in a period of time can be estimated by the change of the eating possibility in a period of time and the movement trajectory of the beef cattle's head. Specifically, the eating amount analysis module 320 can obtain the change trajectory of the centroid of the head area of ​​the beef cattle in a preset time period based on the beef cattle eating image, and obtain the mean value of the eating possibility of the beef cattle in the preset time period, and calculate the eating amount of the beef cattle in the preset time period based on the following mathematical expression:

[0056] ;

[0057] in, Is the first The amount of food intake corresponding to the connected domain of beef cattle is is the starting point of the preset time period. is the end point of the preset time period; It is the sequence number of any frame image within the preset time period; is the total number of image frames within the preset time period; For the The connected area of ​​beef cattle is the first Frame image and Mean eating probability of frame images; It is The connected area of ​​beef cattle is the first Frame image and The centroid distance of the head area of ​​the beef cattle in the frame image; is a hyperbolic tangent function. That is, the amount of food consumed by the beef cattle is related to the product of the moving distance of the center of mass of the head region of the beef cattle between two adjacent frames in a preset time period and the mean value of the possibility of eating.

[0058] In some embodiments of the present disclosure, specifically, Figure 3 FIG. 1 shows a detailed structural diagram of an abnormal monitoring unit. Figure 3As shown, the abnormal monitoring unit 400 may further include an abnormal eating analysis module 410. It is understandable that, based on the acquisition of the beef cattle food intake parameters provided in the aforementioned embodiments, it is possible to judge whether the beef cattle have abnormal eating conditions based on the difference in food intake between the beef cattle and other beef cattle over a period of time. Specifically, the abnormal eating analysis module 410 pre-clustered the beef cattle based on the body shape of the beef cattle according to the beef cattle eating images, and divided the beef cattle into a plurality of clusters, and the size of the connected domains of the beef cattle in each cluster is similar. It is understandable that there are differences in food intake of beef cattle of different body sizes, and the pre-clustering can better reflect the abnormal eating conditions of the beef cattle based on the food intake.

[0059] In the above embodiment, further, the eating abnormality analysis module 410 obtains the eating amount of each beef cattle in the cluster within a preset time period based on the output result of the eating amount analysis module 320 in the above embodiment, and further calculates the average eating amount of the beef cattle in the cluster and the eating amount variance of all beef cattle in the cluster, and calculates the eating abnormality parameter of the beef cattle according to the following mathematical expression:

[0060] ;

[0061] in, Characterizing the first The abnormal eating parameters corresponding to the connected domains of beef cattle; The image of beef cattle eating The average food intake of all beef cattle in the cluster where the connected domain of beef cattle is located; The image of beef cattle eating The amount of food intake corresponding to the connected domain of beef cattle; The image of beef cattle eating The variance of the food intake of all beef cattle in the cluster where the connected domain of beef cattle is located; To obtain the maximum value function, Used to avoid the extreme case where the denominator is 0; is a hyperbolic tangent function. Based on the above mathematical expressions, it can be understood that the abnormal feeding parameter of beef cattle is related to the ratio between the difference between the feeding amount of beef cattle and the mean feeding amount of the cluster, and the feeding amount variance of the cluster; for the difference between the feeding amount of beef cattle and the mean feeding amount of the cluster to which the beef cattle belongs, the larger the value, the greater the difference in feeding amount between the beef cattle and the beef cattle of similar body size in the cluster, that is, the difference between the feeding amount of beef cattle and the mean feeding amount of the cluster to which the beef cattle belongs is positively correlated with the obtained abnormal feeding parameter; and the smaller the value of the feeding amount variance of the cluster, the more similar the feeding amount of beef cattle of similar body size in the cluster, in which case the difference between the feeding amount of beef cattle and the mean feeding amount can better reflect the degree of abnormal feeding, that is, the feeding amount variance of the cluster is negatively correlated with the obtained abnormal feeding parameter.

[0062] In some embodiments of the present disclosure, Figure 2 As shown, the eating analysis unit 300 may further include an eating willingness analysis module 330 and a feed delivery analysis module 340. It is understandable that, based on the relevant description of the aforementioned embodiment, you can reflect whether each beef cattle has abnormal eating conditions according to the obtained abnormal eating parameters; in actual application scenarios, it may also happen that the beef cattle cannot eat in time due to the failure to deliver fodder in time, which makes the abnormal eating parameters of the beef cattle abnormally high. It is necessary to further analyze the behavior of each beef cattle according to the monitoring image to avoid the deviation of the feeding accuracy of the farm on the abnormal eating of the beef cattle. It is understandable that when inaccurate feeding occurs on a farm, there will be a situation where there is no food when the beef cattle go to eat, and there will be a large difference in the amount of food eaten by different beef cattle. Since some beef cattle are closer to the feed trough, they can eat more feed, while those far away from the feed trough may not eat due to untimely feeding. Therefore, it is necessary to compare the food intake of the beef cattle with their actual eating intention to obtain the accuracy of the feed. In the embodiment of the present disclosure, the eating intention analysis module 330 and the feed delivery analysis module 340 are specifically set to perform relevant analysis.

[0063] Specifically, in the above embodiment, the eating willingness analysis module 330 is used to obtain the distance between the connected domain of each beef cattle and each feeding trough within a preset time period according to the beef cattle eating image to form a distance sequence, so as to obtain the eating willingness of each beef cattle based on the distribution of each minimum point in each distance sequence based on the following mathematical expression:

[0064] ;

[0065] in, Characterization The degree of willingness to eat in the connected area of ​​beef cattle; is any minimum point in the distance sequence of the connected domain of beef cattle; It is the minimum point closest to any minimum point in the distance sequence of the connected domain of beef cattle; is the number of minimum value points in the distance sequence of the connected domain of beef cattle; For the The first The distance from the minimum point to the feeding trough; For the The first The distance value from the nearest minimum point among the minimum points to the feeding trough; For the The first The average distance between a minimum point and its nearest minimum point; For the The first The distance between a minimum point and its nearest minimum point; is a hyperbolic tangent function. It is understandable that when the beef cattle have a higher willingness to eat, they will approach the feeding trough. Therefore, the distance sequence between the connected domain of the beef cattle and each feeding trough can be obtained based on the above mathematical expression to determine whether the beef cattle will always approach the feeding trough.

[0066] In the above embodiment, further, the feed delivery analysis module 340 is used to calculate the feed delivery accuracy parameter based on the difference between the eating willingness of the beef cattle and the food intake of the beef cattle within a preset time period based on the following mathematical expression:

[0067] ;

[0068] in, Characterizes the feed delivery accuracy parameter; I represents the number of connected domains of beef cattle; is an exponential function with e as the base. It can be understood that the greater the difference between the cattle's willingness to eat and the cattle's food intake, the greater the cattle's willingness to eat and the smaller the cattle's food intake, that is, the feed placement accuracy parameter is inversely correlated with the difference between the cattle's willingness to eat and the cattle's food intake.

[0069] The feed placement accuracy parameters were negatively correlated with the difference between the cattle's willingness to eat and the amount of food they ate.

[0070] In some embodiments of the present disclosure, Figure 3 As shown, the abnormal monitoring unit 400 may further include an abnormal factor acquisition unit 420. Specifically, the abnormal factor acquisition unit 420 is used to calculate the abnormal factor according to the feed delivery accuracy parameter and the eating abnormality parameter of the beef cattle, wherein the abnormal factor is configured as the product of the feed delivery accuracy parameter and the eating abnormality parameter, and the relevant mathematical expression is as follows:

[0071] ;

[0072] in, Characterizing the first It is understandable that when the abnormality of the beef cattle's feeding and the accuracy of feeding are higher, it can be more likely that the beef cattle's feeding is abnormal due to their own problems, so the corresponding abnormality factor value is higher.

[0073] In some embodiments of the present disclosure, Figure 3 As shown, the abnormal monitoring unit 400 may further include a video marking module 430 and an abnormal warning module 440, wherein specifically, the video marking module 430 is used to mark the image frame at the corresponding moment in the eating video and generate a marking result when the abnormal factor exceeds a preset range. For example, a preset threshold value may be set, and when the value of the abnormal factor is greater than the preset threshold value, the image frame at the corresponding moment in the eating video is marked.

[0074] In the above embodiment, the abnormal warning module 440 is used to generate corresponding abnormal eating warning information according to the marking result when the number of consecutive marked image frames is greater than a preset threshold. It can be understood that when several consecutive frames are marked, they are marked as abnormal, and it can be judged that the target beef cattle are abnormal and the breeder is alarmed; further, when the number of abnormal beef cattle exceeds a certain preset threshold, it is necessary to further analyze the feed delivery situation, analyze whether there is a situation of untimely delivery, and arrange veterinarians to check the abnormal beef cattle to ensure the health of the beef cattle.

[0075] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as "circuits", "modules" or "platforms".

[0076] Specifically, Figure 4 According to an embodiment of the present disclosure, a schematic diagram of the structure of a video monitoring device for feeding grass to beef cattle is shown. Figure 4 The video surveillance device 600 according to this embodiment of the present disclosure is described. Figure 4 The video surveillance device 600 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0077] like Figure 4 As shown, the video surveillance device 600 is in the form of a general-purpose computing device. The components of the video surveillance device 600 may include, but are not limited to: at least one processing unit 610, at least one storage unit 620, a bus 630 connecting different platform components (including the storage unit 620 and the processing unit 610), a display unit 640, etc.

[0078] The storage unit stores program codes, which can be executed by the processing unit 610, so that the processing unit 610 performs the steps according to various exemplary embodiments of the present disclosure. For example, the processing unit 610 can perform the following steps: Figure 1The relevant steps of the copy generation method shown in .

[0079] The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access memory (RAM) 6201 and / or a cache memory unit 6202 , and may further include a read-only memory unit (ROM) 6203 .

[0080] The storage unit 620 may also include a program / utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0081] Bus 630 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0082] The video surveillance device 600 may also communicate with one or more external devices 700 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the video surveillance device 600, and / or communicate with any device that enables the video surveillance device 600 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 650. In addition, the video surveillance device 600 may also communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network, such as the Internet) via a network adapter 660. The network adapter 660 may communicate with other modules of the video surveillance device 600 via a bus 630. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the video surveillance device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage platforms, etc.

[0083] In summary, the technical solution provided by the present disclosure can obtain a feeding video that can reflect the feeding situation of beef cattle by continuously monitoring and shooting the beef cattle feeding area and the forage storage area in the breeding area, and obtain various parameters that characterize the possibility of beef cattle eating, the amount of food eaten, and the degree of willingness to eat by performing image processing and analysis on the display screen in the eating video frame by frame, and calculate the abnormal eating factors of beef cattle in the monitoring video based on the parameters obtained by these image processing. The technical solution provided by the present disclosure supports the continuous detection of abnormal eating factors of beef cattle through monitoring videos, can assist in judging whether beef cattle have abnormal eating conditions and issue corresponding alarm information in a timely manner, effectively ensuring the safety of beef cattle breeding, and has promotional value.

[0084] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0085] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A video monitoring system for feeding forage to beef cattle, characterized in that: The video surveillance system comprises: A video monitoring unit is used to continuously monitor and obtain eating videos of beef cattle feeding on forage; An image recognition unit, used for performing image recognition based on the eating video to obtain an image of the beef cattle eating; A feeding analysis unit, used for performing image analysis on the feeding situation of the beef cattle according to the feeding image of the beef cattle to obtain an image analysis result, wherein the image analysis result includes one or more combinations of the feeding possibility of the beef cattle, the feeding amount of the beef cattle in a period of time, and the feeding willingness of the beef cattle; An abnormality monitoring unit is used to perform an eating abnormality analysis according to the image analysis result to obtain an abnormality factor characterizing the degree of abnormal eating of the beef cattle, and continuously monitor the abnormality factor to obtain the feeding material status of the beef cattle; The image recognition unit further comprises: Obtaining an eating grayscale image corresponding to each frame in the eating video; Performing edge detection on the eating grayscale image to obtain the beef cattle eating image, wherein the beef cattle eating image includes an edge set in the eating grayscale image and a connected domain of each beef cattle in the eating grayscale image; The image recognition unit also includes: In the eating grayscale image, the edge of the connected domain of each beef cattle is divided into two parts and an ellipse fitting process is performed using the least square method; According to the result of the ellipse fitting process, the portion corresponding to the minimum effect of the ellipse fitting process is used as the head region of the beef cattle; The eating analysis unit includes an eating possibility analysis module; The eating possibility analysis module is used to obtain a first distance between the head area of ​​the beef cattle and the feeding trough, and a second distance between the connected area of ​​the beef cattle and the feeding trough according to the beef cattle eating image; According to the first distance and the second distance, calculating to obtain the eating possibility of the beef cattle, wherein the eating possibility is positively correlated with the ratio of the second distance to the first distance; The food intake analysis unit also includes a food intake analysis module; The food intake analysis module is used to obtain, based on the cattle eating image, a trajectory of the centroid change of the head area of ​​the cattle within a preset time period, and to obtain a mean value of the eating probability of the cattle within the preset time period; According to the change trajectory of the center of mass of the head region and the average of the eating possibilities, the food intake of the beef cattle in the preset time period is calculated, and the food intake is associated with the product of the moving distance of the center of mass of the head region of the beef cattle between two adjacent frames in the preset time period and the average of the eating possibilities; The abnormal monitoring unit includes an eating abnormality analysis module; The eating abnormality analysis module is used to perform clustering processing on the beef cattle based on the eating images of the beef cattle and the body shapes of the beef cattle, and to obtain the food intake of the beef cattle within the preset time period; Obtaining an abnormal eating parameter of the beef cattle according to a difference between the food intake of the beef cattle and the average food intake of all the beef cattle in the cluster where the beef cattle is located, wherein the abnormal eating parameter of the beef cattle is associated with a ratio between a difference between the food intake of the beef cattle and the mean food intake of the cluster and a food intake variance of the cluster; The feeding analysis unit includes a feeding willingness analysis module and a feed delivery analysis module; The eating willingness analysis module is used to obtain the distance between the connected domain of each of the beef cattle and each feed trough within the preset time period according to the beef cattle eating image, so as to form a distance sequence; According to the distribution of each minimum point in each distance sequence, the eating willingness of each beef cattle is obtained; The feed delivery analysis module is used to calculate and obtain a feed delivery accuracy parameter according to the difference between the beef cattle's willingness to eat and the beef cattle's food intake within the preset time period, and the feed delivery accuracy parameter is inversely correlated with the difference between the beef cattle's willingness to eat and the beef cattle's food intake; The abnormality monitoring unit also includes an abnormality factor acquisition unit; The abnormal factor acquisition unit is used to calculate the abnormal factor according to the feed delivery accuracy parameter and the eating abnormality parameter of the beef cattle, and the abnormal factor is configured as the product of the feed delivery accuracy parameter and the eating abnormality parameter.

2. The video monitoring system for feeding forage to beef cattle according to claim 1, characterized in that: The image content in the eating video includes the complete beef cattle feeding area and the forage storage area.

3. The video monitoring system for feeding forage to beef cattle according to claim 1, characterized in that: The abnormal monitoring unit further includes a video marking module and an abnormal warning module; The video marking module is used to mark the image frame at the corresponding moment in the eating video and generate a marking result when the abnormal factor exceeds a preset range; The abnormal warning module is used to generate corresponding abnormal eating warning information according to the marking result when the number of the image frames continuously marked is greater than a preset threshold.

Images