CN111523446A - Image recognition method and device, and storage medium - Google Patents
Image recognition method and device, and storage medium Download PDFInfo
- Publication number
- CN111523446A CN111523446A CN202010319299.2A CN202010319299A CN111523446A CN 111523446 A CN111523446 A CN 111523446A CN 202010319299 A CN202010319299 A CN 202010319299A CN 111523446 A CN111523446 A CN 111523446A
- Authority
- CN
- China
- Prior art keywords
- livestock
- behavior
- milk
- target detection
- detection model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention provides an image recognition method, an image recognition device and a storage medium, wherein the method comprises the steps of photographing livestock in a fence through a camera in the fence to obtain image information; determining livestock behaviors in the image information by using a target detection model, wherein the target detection model is obtained by using a plurality of groups of data through machine learning training, and each group of data of the plurality of groups of data comprises: livestock behavior tags and image information; feeding back alarm information under the condition that the livestock behavior is determined to belong to milk eating abnormity, wherein the milk eating abnormity at least comprises one of the following steps: the young animals in the livestock can not eat the milk, and the young animals in the livestock can not eat the milk. According to the invention, the problem that livestock cannot be nursed delicately and efficiently is solved, and the effects of accurate detection and timely reporting of abnormal behaviors of livestock are further achieved.
Description
Technical Field
The invention relates to the field of image recognition and livestock care, in particular to an image recognition method and device and a storage medium.
Background
Livestock cubs are mainly monitored through manual work when primary nursing, such as listening to a clatter and the like, whether the cubs have no milk or whether the cubs have enough milk to eat or not is preliminarily judged, and then the livestock cubs are manually moved to a fence to be checked. However, if the livestock cubs with huge numbers are monitored manually, a large amount of manpower resources are occupied. In addition, the abnormal findings are not timely, resulting in livestock cub malnutrition.
Aiming at the problem that fine and efficient nursing of livestock cannot be carried out in the related technology, an effective solution does not exist at present.
Disclosure of Invention
The embodiment of the invention provides an image identification method, an image identification device and a storage medium, which at least solve the problem that livestock cannot be carefully and efficiently nursed in the related technology.
According to an embodiment of the present invention, there is provided an image recognition method including: shooting livestock in the fence through a camera in the fence to obtain image information; determining livestock behaviors in the image information by using a target detection model, wherein the target detection model is obtained by using a plurality of groups of data through machine learning training, and each group of data of the plurality of groups of data comprises: livestock behavior tags and image information; feeding back alarm information under the condition that the livestock behavior is determined to belong to milk eating abnormity, wherein the milk eating abnormity at least comprises one of the following steps: the young animals in the livestock can not eat the milk, and the young animals in the livestock can not eat the milk.
According to another embodiment of the present invention, there is provided an image recognition apparatus including: the acquisition module is used for photographing livestock in the fence through a camera in the fence to obtain image information; the recognition module is used for determining livestock behaviors in the image information by using a target detection model, wherein the target detection model is obtained by using a plurality of groups of data through machine learning training, and each group of data of the plurality of groups of data comprises: livestock behavior tags and image information; the feedback module is used for feeding back alarm information under the condition that the livestock behavior is determined to belong to milk eating abnormity, wherein the milk eating abnormity at least comprises one of the following components: the young animals in the livestock can not eat the milk, and the young animals in the livestock can not eat the milk.
According to a further embodiment of the present invention, there is also provided a storage medium having a computer program stored therein, wherein the computer program is arranged to perform the steps of any of the above method embodiments when executed.
According to yet another embodiment of the present invention, there is also provided an electronic device, including a memory in which a computer program is stored and a processor configured to execute the computer program to perform the steps in any of the above method embodiments.
According to the invention, as the target detection model is used for determining the livestock behaviors in the image information, the livestock behaviors in the fence can be fed back by only photographing the livestock in the fence through the camera in the fence to obtain the image information, and the alarm information can be fed back under the condition that the livestock behaviors are determined to be abnormal in milk intake. Therefore, the problems that manpower is consumed during livestock breeding and abnormal livestock behaviors are not found timely enough can be solved, and the effects of accurately detecting and timely reporting the abnormal livestock behaviors are achieved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiment(s) of the invention and together with the description serve to explain the invention without limiting the invention. In the drawings:
fig. 1 is a block diagram of a hardware configuration of a mobile terminal of an image recognition method according to an embodiment of the present invention;
FIG. 2 is a flow chart of an image recognition method according to an embodiment of the present invention;
FIG. 3 is a block diagram of an image recognition apparatus according to an alternative embodiment of the present invention;
fig. 4 is a block diagram of an image recognition apparatus according to an alternative embodiment of the present invention.
Detailed Description
The invention will be described in detail hereinafter with reference to the accompanying drawings in conjunction with embodiments. It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.
Example 1
The method provided by the first embodiment of the present application may be executed in a mobile terminal, a computer terminal, or a similar computing device. Taking an example of the method performed in a mobile terminal, fig. 1 is a block diagram of a hardware structure of the mobile terminal according to an embodiment of the present invention. As shown in fig. 1, the mobile terminal 10 may include one or more (only one shown in fig. 1) processors 102 (the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, and optionally may also include a transmission device 106 for communication functions and an input-output device 108. It will be understood by those skilled in the art that the structure shown in fig. 1 is only an illustration, and does not limit the structure of the mobile terminal. For example, the mobile terminal 10 may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.
The memory 104 may be used to store a computer program, for example, a software program and a module of application software, such as a computer program corresponding to the image recognition method in the embodiment of the present invention, and the processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, so as to implement the method described above. The memory 104 may include high speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory located remotely from the processor 102, which may be connected to the mobile terminal 10 via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmission device 106 is used for receiving or transmitting data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of the mobile terminal 10. In one example, the transmission device 106 includes a Network adapter (NIC), which can be connected to other Network devices through a base station so as to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is used for communicating with the internet in a wireless manner.
In the present embodiment, an image recognition method operating in the mobile terminal is provided, and fig. 2 is a flowchart of the image recognition method according to an embodiment of the present invention, as shown in fig. 2, the flowchart includes the following steps:
step S202, shooting livestock in the fence through a camera in the fence to obtain image information;
it should be noted that the pen may accommodate a plurality of animals as well as animal litters, and the specific number is not specifically limited in this application.
Step S204, determining livestock behaviors in the image information by using a target detection model, wherein the target detection model is obtained by using multiple groups of data through machine learning training, and each group of data of the multiple groups of data comprises: livestock behavior tags and image information;
each group of data comprises the marked livestock behavior tags and the original image information, namely, the livestock behavior tags are arranged on each image.
Step S206, under the condition that the livestock behavior is determined to belong to milk eating abnormity, feeding back alarm information, wherein the milk eating abnormity at least comprises one of the following conditions: the young animals in the livestock can not eat the milk, and the young animals in the livestock can not eat the milk.
Image information that obtains is shot through the camera promptly, through image information discerns the action of livestock cub, and then judges the condition of eating inadequately to the livestock cub milk. The condition where the livestock pup is not enough may be that the livestock pup has milk but not enough milk or that the livestock pup is not. And training the model after the target frame is marked through the target detection model. And the final output result of the target detection model is to judge whether the behavior of the livestock cubs is abnormal or not, and if the behavior of the livestock cubs is abnormal, the abnormal reporting is further carried out, and alarm information is fed back.
Through the steps, as the target detection model is used for determining the livestock behaviors in the image information, the livestock behaviors in the fence can be fed back with alarm information only by photographing the livestock in the fence through the camera in the fence to obtain the image information under the condition that the livestock behaviors are determined to be abnormal in milk intake. Therefore, the problems that manpower is consumed during livestock breeding and abnormal livestock behaviors are not found timely enough can be solved, and the effects of accurately detecting and timely reporting the abnormal livestock behaviors are achieved.
In order to guarantee that every cub can both have sufficient breast milk, improve the automatic detection of computer through artificial intelligence technique, still include: before determining the livestock behaviors in the image information by using a target detection model, obtaining the target detection model through machine learning training for the first livestock behaviors, and determining a first behavior area of the livestock in the image information by using the target detection model, wherein the first behavior refers to the milk eating behavior of the cubs in the livestock, and the first behavior area refers to an area where the cubs eat milk. The first action area can be detected through the target detection model obtained by training the first action of the livestock, namely the area where the cubs eat the milk is detected through the target detection model. Marking the position of the milk eating area of the cub in the image before training, wherein the marking content can comprise: and obtaining a sample set for model training by using the coordinates of the upper left corner of the region, the length of the region and the width of the region.
Or before determining the livestock behavior in the image information by using a target detection model, obtaining the target detection model by machine learning training for a second livestock behavior, and determining a second behavior area of the livestock in the image information by using the target detection model, wherein the second behavior is the breast-feeding behavior of the female livestock in the livestock, and the second behavior area is the area where the female livestock breast-feed is located. Namely, the area where the female animals are suckling is detected through an object detection model. Marking the position of the area where the female animal suckles in the image before training, wherein the marking content can comprise: and obtaining a sample set for model training by using the coordinates of the upper left corner of the region, the length of the region and the width of the region. In detecting the area where the dam is lactating, for example, each teat area of the dam may be detected by the target detection model, Faster R-CNN.
It should be noted that, the female animals are described by taking the sows as an example, the nipples of the sows have different milk yields, some have more milk and some have less milk, and some have no milk. When some weak pig cubs and newly-fostered pig cubs cannot occupy the teats with high milk yield, malnutrition and diseases are prone to be caused for a long time.
Several conditions can be manifested in the case of inadequate milk consumption by piglets: one condition is long term arching of the sow's teat, emaciation, and abdominal collapse. Another situation is when other pig litters are sleeping in the incubator, they are turning outside the incubator. In another case, when the piglets eat the milk, the piglets tend to roll back and forth. A further situation is when two piglet pups fight a teat, each of which holds the teat up to milk. If two piglets fight for one nipple, one piglet can be weakened due to insufficient milk, and different abnormal behaviors are explained in detail below.
And aiming at different abnormal livestock behaviors, training the abnormal livestock behaviors in scenes to obtain corresponding target detection models, and detecting the abnormal livestock behaviors by adopting the target detection models respectively. Determining livestock behavior in the image information using a target detection model, comprising: determining the area where the cubs are suckling in the image information by using the target detection model; identifying whether the cubs have preset behaviors in the region where the cubs eat milk or not according to the identification model; and under the condition that the cubs are identified to have preset behaviors in the region where the cubs eat the milk according to the identification model, judging whether the cubs are thin or weak.
Taking the piglet as an example for specific explanation, when finding that a certain piglet (piglet) is arched by the cow teat for a long time, the regional image information is intercepted, and then the target detection model is utilized to judge whether the piglet is weak in body, whether the abdomen is collapsed, and the like. The method comprises the steps of firstly detecting the area of each piglet through a target detection model such as fast R-CNN, then identifying whether the behavior of the piglet arching the nipple exists in each piglet area through an I3D behavior identification model, carrying out example segmentation on the piglet area in the piglet milk eating area if the behavior exists, then inputting the extracted characteristics into a full connection layer for classification, and judging whether the piglet is thin or weak.
And aiming at different abnormal livestock behaviors, training the abnormal livestock behaviors in scenes to obtain corresponding target detection models, and detecting the abnormal livestock behaviors by adopting the target detection models respectively. Determining livestock behavior in the image information using a target detection model, comprising: determining the region where the female animals suckle in the image information by using the target detection model; identifying whether the female animal has a preset behavior in the area where the female animal suckles according to an identification model; and under the condition that the livestock is identified to have a preset behavior in the area where the female livestock suckles according to the identification model, judging whether the cubs have back-and-forth movement behaviors or not.
Taking the piglet as an example for specific explanation, when other piglet (piglet) sleeps in the incubator, the piglet is found to turn outside the incubator or turn around when the piglet eats milk. And for the image to be detected and identified, detecting whether the cubs eat the milk in the area by using the target detection model. For example, an area where each cub eats milk can be detected through the Faster R-CNN target detection model, and then whether a pig cub yo-yo behavior exists in the area where each cub milk is identified through the behavior identification model I3D.
And aiming at different abnormal livestock behaviors, training the abnormal livestock behaviors in scenes to obtain corresponding target detection models, and detecting the abnormal livestock behaviors by adopting the target detection models respectively. Before determining livestock behavior in the image information using a target detection model, comprising: obtaining a target detection model for the third behavior of the livestock through machine learning training, and determining the third behavior area of the livestock in the image information through the target detection model, wherein the third behavior refers to the behavior of suckling the piglets of at least two livestock simultaneously, and the third behavior area refers to the milk eating area of a plurality of piglets; after determining the livestock behavior in the image information by using the target detection model, the method further comprises the following steps: and in the case that the number of livestock which are used for feeding and the number of suckling which are provided are not equal in the multi-cub milk feeding area, judging that the at least two cubs have the behavior of feeding milk at the same time.
Considering that if two piglets fight for one teat, one piglet is inevitably weakened due to insufficient milk intake and finally becomes a dead pig. Taking the piglet as an example for specific explanation, when the condition that two piglets contend for one teat is judged, a piglet baby feeding area can be detected by using the target detection model, and then whether the number of the piglets in the piglet baby feeding area is equal to the number of the cows is judged.
Optionally, there are also situations where the piglet baby does not eat the teat. When the number of the pig cubs in the pig cub milk eating area is more than the number of the female animal teats, the pig cubs cannot eat the teats, so that the physical weakness is caused.
Further, for feeding back the early warning information in real time, under the condition that the livestock behavior is determined to belong to abnormal milk eating, the feeding back of the early warning information comprises the following steps: feeding back first warning information in case it is determined that the livestock behavior belongs to the fact that the young animals in the livestock cannot eat milk, wherein the first warning information comprises: fence information of the cubs and image information that the cubs can not eat milk; feeding back second warning information in case it is determined that the livestock behavior belongs to the fact that the young milk in the livestock is not sufficient to eat, wherein the second warning information comprises: fence information of the cubs and image information of the cubs in the livestock which are not enough to eat. The alarm information is different, and the fence information, the image information that the cubs cannot eat the milk and the image information that the cubs in the livestock cannot eat the milk can be fed back and reported in the alarm information, so that the fence information can be timely positioned, and the cubs with abnormal behaviors can be found.
Through the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but the former is a better implementation mode in many cases. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (e.g., ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal device (e.g., a mobile phone, a computer, a server, or a network device) to execute the method according to the embodiments of the present invention.
Example 2
In this embodiment, an image recognition apparatus is further provided, and the apparatus is used to implement the foregoing embodiments and preferred embodiments, and the description of the apparatus is omitted for brevity. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. Although the means described in the embodiments below are preferably implemented in software, an implementation in hardware, or a combination of software and hardware is also possible and contemplated.
Fig. 3 is a block diagram of an image recognition apparatus according to an embodiment of the present invention, as shown in fig. 3, the apparatus including:
the acquisition module 30 is used for photographing livestock in the fence through a camera in the fence to obtain image information;
an identification module 32, configured to determine livestock behaviors in the image information by using a target detection model, where the target detection model is obtained by machine learning training using multiple sets of data, and each set of data of the multiple sets of data includes: livestock behavior tags and image information;
a feedback module 34, configured to feed back alarm information in case that it is determined that the livestock behavior belongs to a milk intake anomaly, where the milk intake anomaly includes at least one of: the young animals in the livestock can not eat the milk, and the young animals in the livestock can not eat the milk.
In the harvesting module 30, it should be noted that the pen can accommodate a plurality of livestock and livestock litters, and the specific number is not specifically limited in this application.
In the identification module 32, each set of data includes the marked livestock behavior tag and the original image information, that is, each image has the livestock behavior tag.
The feedback module 34 is provided with image information obtained by camera shooting, and the behavior of the livestock cubs is identified through the image information, so that the condition that the livestock cubs are not eaten enough is judged. The condition where the livestock pup is not enough may be that the livestock pup has milk but not enough milk or that the livestock pup is not. And training the model after the target frame is marked through the target detection model. And the final output result of the target detection model is to judge whether the behavior of the livestock cubs is abnormal or not, and if the behavior of the livestock cubs is abnormal, the abnormal reporting is further carried out, and alarm information is fed back.
Through the modules, as the target detection model is used for determining the livestock behaviors in the image information, the livestock behaviors in the fence can be fed back with alarm information only by photographing the livestock in the fence through the camera in the fence to obtain the image information under the condition that the livestock behaviors are determined to be abnormal in milk eating. Therefore, the problems that manpower is consumed during livestock breeding and abnormal livestock behaviors are not found timely enough can be solved, and the effects of accurately detecting and timely reporting the abnormal livestock behaviors are achieved.
Fig. 4 is a block diagram of an image recognition apparatus according to an alternative embodiment of the present invention, as shown in fig. 4, the apparatus further includes a first training module 40, in addition to all modules shown in fig. 4, for training a first action of the livestock to obtain a target detection model through machine learning before determining the action of the livestock in the image information by using the target detection model, and determining a first action area of the livestock in the image information through the target detection model, wherein the first action is a milking action of a young animal in the livestock, and the first action area is an area where the young animal eats milk.
The first action area can be detected through the target detection model obtained by training the first action of the livestock in the first training module 40, namely, the area where the cubs eat the milk is detected through the target detection model. Marking the position of the milk eating area of the cub in the image before training, wherein the marking content can comprise: and obtaining a sample set for model training by using the coordinates of the upper left corner of the region, the length of the region and the width of the region.
The second training module 42 is configured to, before determining the livestock behavior in the image information by using the target detection model, perform machine learning training on the second livestock behavior to obtain a target detection model, and determine a second behavior region of the livestock in the image information by using the target detection model, where the second behavior is a breast-feeding behavior of the livestock, and the second behavior region is a region where the livestock is breast-feeding.
The second training module 42 detects the area where the female animal is lactating through the target detection model. Marking the position of the area where the female animal suckles in the image before training, wherein the marking content can comprise: and obtaining a sample set for model training by using the coordinates of the upper left corner of the region, the length of the region and the width of the region. In detecting the area where the dam is lactating, for example, each teat area of the dam may be detected by the target detection model, Faster R-CNN.
It should be noted that, the above modules may be implemented by software or hardware, and for the latter, the following may be implemented, but not limited to: the modules are all positioned in the same processor; alternatively, the modules are respectively located in different processors in any combination.
Embodiments of the present invention also provide a storage medium having a computer program stored therein, wherein the computer program is arranged to perform the steps of any of the above method embodiments when executed.
Alternatively, in the present embodiment, the storage medium may be configured to store a computer program for executing the steps of:
s1, shooting the livestock in the fence through a camera in the fence to obtain image information;
s2, determining livestock behaviors in the image information by using a target detection model, wherein the target detection model is obtained by using multiple groups of data through machine learning training, and each group of data of the multiple groups of data comprises: livestock behavior tags and image information;
s3, feeding back alarm information under the condition that the livestock behavior is determined to belong to milk eating abnormity, wherein the milk eating abnormity at least comprises one of the following conditions: the young animals in the livestock can not eat the milk, and the young animals in the livestock can not eat the milk.
Optionally, the storage medium is further arranged to store a computer program for performing the steps of:
s21, before determining the livestock behaviors in the image information by using a target detection model, obtaining the target detection model by machine learning training for the first livestock behaviors, and determining a first behavior area of the livestock in the image information by using the target detection model, wherein the first behavior is the milk eating behavior of cubs in the livestock, and the first behavior area is the area where the cubs eat milk;
s22, before determining the livestock behavior in the image information by using the target detection model, obtaining the target detection model by machine learning training for the second livestock behavior, and determining a second behavior area of the livestock in the image information by using the target detection model, wherein the second behavior is the breast-feeding behavior of the female livestock in the livestock, and the second behavior area is the area where the female livestock breast-feed is located.
Optionally, in this embodiment, the storage medium may include, but is not limited to: various media capable of storing computer programs, such as a usb disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a removable hard disk, a magnetic disk, or an optical disk.
Embodiments of the present invention also provide an electronic device comprising a memory having a computer program stored therein and a processor arranged to run the computer program to perform the steps of any of the above method embodiments.
Optionally, the electronic apparatus may further include a transmission device and an input/output device, wherein the transmission device is connected to the processor, and the input/output device is connected to the processor.
Optionally, in this embodiment, the processor may be configured to execute the following steps by a computer program:
s1, shooting the livestock in the fence through a camera in the fence to obtain image information;
s2, determining livestock behaviors in the image information by using a target detection model, wherein the target detection model is obtained by using multiple groups of data through machine learning training, and each group of data of the multiple groups of data comprises: livestock behavior tags and image information;
s3, feeding back alarm information under the condition that the livestock behavior is determined to belong to milk eating abnormity, wherein the milk eating abnormity at least comprises one of the following conditions: the young animals in the livestock can not eat the milk, and the young animals in the livestock can not eat the milk.
Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation manners, and this embodiment is not described herein again.
It will be apparent to those skilled in the art that the modules or steps of the present invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and alternatively, they may be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, the present invention is not limited to any specific combination of hardware and software.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. An image recognition method, comprising:
shooting livestock in the fence through a camera in the fence to obtain image information;
determining livestock behaviors in the image information by using a target detection model, wherein the target detection model is obtained by using a plurality of groups of data through machine learning training, and each group of data of the plurality of groups of data comprises: livestock behavior tags and image information;
feeding back alarm information under the condition that the livestock behavior is determined to belong to milk eating abnormity, wherein the milk eating abnormity at least comprises one of the following steps: the young animals in the livestock can not eat the milk, and the young animals in the livestock can not eat the milk.
2. The image recognition method according to claim 1, further comprising:
before determining the livestock behaviors in the image information by using a target detection model, obtaining the target detection model by machine learning training for the first livestock behaviors, and determining a first behavior area of the livestock in the image information by using the target detection model, wherein the first behaviors refer to the milk eating behaviors of cubs in the livestock, and the first behavior area refers to an area where the cubs eat milk; or
Before determining the livestock behaviors in the image information by using a target detection model, obtaining the target detection model by machine learning training for second livestock behaviors, and determining a second behavior area of the livestock in the image information by using the target detection model, wherein the second behaviors refer to the lactation behaviors of the female livestock in the livestock, and the second behavior area refers to the area where the female livestock is lactated.
3. The image recognition method of claim 1, wherein determining livestock behavior in the image information using a target detection model comprises:
determining the area where the cubs are suckling in the image information by using the target detection model;
identifying whether the cubs have preset behaviors in the region where the cubs eat milk or not according to the identification model;
and under the condition that the cubs are identified to have preset behaviors in the region where the cubs eat the milk according to the identification model, judging whether the cubs are thin or weak.
4. The image recognition method of claim 1, wherein determining livestock behavior in the image information using a target detection model comprises:
determining the region where the female animals suckle in the image information by using the target detection model;
identifying whether the female animal has a preset behavior in the area where the female animal suckles according to an identification model;
and under the condition that the livestock is identified to have a preset behavior in the area where the female livestock suckles according to the identification model, judging whether the cubs have back-and-forth movement behaviors or not.
5. The image recognition method according to claim 1,
before determining livestock behavior in the image information using a target detection model, comprising:
obtaining a target detection model for the third behavior of the livestock through machine learning training, and determining the third behavior area of the livestock in the image information through the target detection model, wherein the third behavior refers to the behavior of suckling the piglets of at least two livestock simultaneously, and the third behavior area refers to the milk eating area of a plurality of piglets;
after determining the livestock behavior in the image information by using the target detection model, the method further comprises the following steps:
and in the case that the number of livestock which are used for feeding and the number of suckling which are provided are not equal in the multi-cub milk feeding area, judging that the at least two cubs have the behavior of feeding milk at the same time.
6. The image recognition method of claim 1, wherein in the case where it is determined that the livestock behavior belongs to a milk eating abnormality, feeding back alarm information includes:
feeding back first warning information in case it is determined that the livestock behavior belongs to the fact that the young animals in the livestock cannot eat milk, wherein the first warning information comprises: fence information of the cubs and image information that the cubs can not eat milk;
feeding back second warning information in case it is determined that the livestock behavior belongs to the fact that the young milk in the livestock is not sufficient to eat, wherein the second warning information comprises: fence information of the cubs and image information of the cubs in the livestock which are not enough to eat.
7. An image recognition apparatus, comprising:
the acquisition module is used for photographing livestock in the fence through a camera in the fence to obtain image information;
the recognition module is used for determining livestock behaviors in the image information by using a target detection model, wherein the target detection model is obtained by using a plurality of groups of data through machine learning training, and each group of data of the plurality of groups of data comprises: livestock behavior tags and image information;
the feedback module is used for feeding back alarm information under the condition that the livestock behavior is determined to belong to milk eating abnormity, wherein the milk eating abnormity at least comprises one of the following components: the young animals in the livestock can not eat the milk, and the young animals in the livestock can not eat the milk.
8. The apparatus of claim 7, further comprising:
the first training module is used for obtaining a target detection model through machine learning training for the first livestock behavior before the target detection model is used for determining the livestock behavior in the image information, and determining a first behavior area of the livestock in the image information through the target detection model, wherein the first behavior is the milk eating behavior of the cubs in the livestock, and the first behavior area is the area where the cubs eat the milk.
9. The apparatus of claim 7, further comprising:
and the second training module is used for obtaining a target detection model through machine learning training for second livestock behaviors before determining the livestock behaviors in the image information by using the target detection model, and determining a second behavior area of the livestock in the image information through the target detection model, wherein the second behaviors refer to the lactation behaviors of the female livestock in the livestock, and the second behavior area refers to the area where the female livestock suckle.
10. A storage medium, in which a computer program is stored, wherein the computer program is arranged to perform the method of any of claims 1 to 6 when executed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010319299.2A CN111523446A (en) | 2020-04-21 | 2020-04-21 | Image recognition method and device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010319299.2A CN111523446A (en) | 2020-04-21 | 2020-04-21 | Image recognition method and device, and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111523446A true CN111523446A (en) | 2020-08-11 |
Family
ID=71903955
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010319299.2A Withdrawn CN111523446A (en) | 2020-04-21 | 2020-04-21 | Image recognition method and device, and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111523446A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112101291A (en) * | 2020-09-27 | 2020-12-18 | 成都睿畜电子科技有限公司 | Livestock nursing method, device, medium and electronic equipment |
-
2020
- 2020-04-21 CN CN202010319299.2A patent/CN111523446A/en not_active Withdrawn
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112101291A (en) * | 2020-09-27 | 2020-12-18 | 成都睿畜电子科技有限公司 | Livestock nursing method, device, medium and electronic equipment |
| CN112101291B (en) * | 2020-09-27 | 2024-01-30 | 成都睿畜电子科技有限公司 | Livestock nursing method, device, medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109618961B (en) | Intelligent feeding system and method for livestock | |
| Tedeschi et al. | Advancements in sensor technology and decision support intelligent tools to assist smart livestock farming | |
| Groher et al. | Digital technology adoption in livestock production with a special focus on ruminant farming | |
| Wang et al. | ASAS-NANP SYMPOSIUM: Applications of machine learning for livestock body weight prediction from digital images | |
| Van der Zande et al. | Individual detection and tracking of group housed pigs in their home pen using computer vision | |
| CN113191655B (en) | Feeding method, device and computer readable storage medium | |
| CN111260895A (en) | Alarm information sending method and device, storage medium and electronic device | |
| KR102141582B1 (en) | Prediction method and the apparatus for onset time of sow farrowing by image analysis | |
| KR20200117610A (en) | Electronic Sow Management Apparatus | |
| de Mol | Automated detection of oestrus and mastitis in dairy cows | |
| JP6635968B2 (en) | Behavior identification device, behavior identification method, and program | |
| CN111523446A (en) | Image recognition method and device, and storage medium | |
| Gu et al. | A two-stage recognition method based on deep learning for sheep behavior | |
| WO2021015683A1 (en) | A herd management system and a tracking sensor | |
| JP7089098B2 (en) | Behavior identification device, behavior identification method, and program | |
| KR20230033068A (en) | Smart farm management system based on behavior pattern and method thereof | |
| Vickery et al. | A preliminary investigation of the feeding behaviour of dairy goat kids reared away from their dams on a computerised ad libitum milk feeding system | |
| Ronghua et al. | Cow behavioral recognition using dynamic analysis | |
| Alammar et al. | An Intelligent Approach of the Fish Feeding System | |
| Hu et al. | Novel cattle farm management system based on IoT and RFID | |
| JP6636204B2 (en) | Behavior identification device, behavior identification method, and program | |
| WO2023190024A1 (en) | Free-range livestock management server device, system, method, and program | |
| CN115830488A (en) | Cow parturition detection method and system based on video understanding | |
| Belluz et al. | Rearing calves with nurse cows: experiences of French organic dairy farmers | |
| Deepak et al. | Advancements in Automated Livestock Monitoring: A Concise Review of Deep Learning-Based Cattle Activity Recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20200811 |