Rabbit farm intelligent management system and method based on Internet of Things

Abstract

The invention belongs to the field of rabbit farm intelligent management, and particularly relates to a rabbit farm intelligent management system and method based on the Internet of Things. The rabbit farm intelligent management system based on the Internet of Things comprises an acquisition terminal, a processing terminal, a first terminal and a second terminal. The collection terminal collects image information of the young rabbit as first information, and the first information comprises the overall appearance and facial details of the young rabbit; and the processing terminal identifies and segments the rabbit contour in the first information, performs feature identification on the segmented rabbit contour, extracts the rabbit flower color corresponding to the rabbit contour, queries the rabbit number corresponding to the rabbit flower color by using the rabbit flower color combined with the stored rabbit database, and obtains the rabbit face flower color according to the rabbit contour. Extracting illness characteristics in the face flower color of the rabbit, and matching the illness type of the rabbit from the stored rabbit illness knowledge graph according to the extracted illness characteristics. According to the scheme, the problem of low efficiency of a traditional rabbit farm breeding mode is solved.

Description

Technical Field

[0001] This solution belongs to the field of intelligent management of rabbit farms, and specifically relates to an intelligent management system and method for rabbit farms based on the Internet of Things. Background Art

[0002] With the rapid development of modern aquaculture, the construction scale of farms is constantly expanding. As an important type of aquaculture, rabbit farms have an increasingly urgent need for intelligent information management. In the traditional process of large-scale rabbit breeding, the determination of the time for various operations mainly depends on the experience of the breeders, and various precise feeding operations are completed. Since each breeder is responsible for managing a large number of rabbits, the breeder needs to shuttle between numerous rabbit cages to complete various feeding tasks, which is time-consuming and laborious. Especially in large-scale rabbit farms, feeding all the rabbits often takes a lot of time, and as the number of rabbits increases, it is easier for breeders to miss feeding tasks, which not only further reduces the breeding efficiency but also makes it difficult to ensure that each rabbit can be fed in a standardized manner.

[0003] To solve the above problems, Chinese Patent CN107563730A proposed an intelligent management system for large-scale rabbit breeding. This system combines a background management terminal and a breeder user terminal. Through a batch management module, a task management module, and a task assignment module, it dynamically gives the breeder's breeding management tasks for the day based on the breeding batch of breeding rabbits and breeding tasks, and realizes intelligent real-time reminders. This system organizes and clarifies complex rabbit farm affairs, effectively avoiding the situation of breeders making mistakes and omissions due to complicated tasks.

[0004] However, although this system has made remarkable progress in task planning and reminders, greatly optimizing the task management process, breeders still need to shuttle back and forth in the rabbit farm to manually complete most of the repetitive tasks such as feeding, inspection (rabbit health status), and cleaning, which greatly consumes the time and energy of breeders. In addition, this mode does not fundamentally solve the problem of low efficiency in traditional breeding, nor can it meet the requirements of modern breeding for high automation and intelligence.

[0005] Therefore, there is an urgent need for an intelligent management system and method for rabbit farms based on the Internet of Things to further improve the work efficiency of rabbit farms, optimize the feeding process, reduce manual intervention, and realize the full-process automation and intelligent management from task assignment to execution. Summary of the Invention

[0006] The purpose of this solution is to provide an intelligent management system and method for rabbit farms based on the Internet of Things to solve the problem of low efficiency in the traditional breeding method of rabbit farms.

[0007] To achieve the above purpose, this solution provides an intelligent management system for rabbit farms based on the Internet of Things, including:

[0008] A collection terminal that collects image information of young rabbits as the first information, where the first information includes the overall appearance and facial details of the young rabbits;

[0009] A processing terminal that identifies and segments the rabbit contour in the first information, performs feature recognition on the segmented rabbit contour, extracts the rabbit coat color corresponding to the rabbit contour, uses the rabbit coat color to query the rabbit number corresponding to the rabbit coat color in the stored rabbit database, then obtains the rabbit facial coat color based on the rabbit contour, extracts the sick features in the rabbit facial coat color, matches the sick type of the rabbit from the stored rabbit sick knowledge graph according to the extracted sick features, generates a sick mark with the sick type and the rabbit number, and adds the corresponding sick mark to the rabbit contour in the first information;

[0010] A first terminal that obtains and displays the first information containing the sick mark, then receives the entered diagnosis and treatment opinions according to the rabbit number, and transmits the diagnosis and treatment opinions and the rabbit number to the processing terminal;

[0011] The processing terminal is also used to obtain the diagnosis and treatment opinions and the rabbit number. When the processing terminal is used to obtain the first information, it obtains the rabbit contour in the first information, identifies and marks the water outlet pipe contour in the first information, and obtains the positional relationship between the rabbit contour and the water outlet pipe contour as the first positional relationship. The processing terminal is used to establish an association relationship between the diagnosis and treatment opinions and the sick mark;

[0012] A second terminal that includes a water outlet pipe and a medicine outlet pipe. One end of the water outlet pipe is connected to a water outlet device, the medicine outlet pipe is connected to a medicine outlet device, and the water outlet pipe is wrapped outside the medicine outlet pipe. The second terminal is used to obtain the first positional relationship and control the time for the water outlet device to discharge drinking water into the water outlet pipe according to the first positional relationship; it is also used to obtain the sick mark in the rabbit contour when obtaining the first positional relationship and control the dosage of the medicine discharged by the medicine outlet device into the medicine outlet pipe according to the diagnosis and treatment opinions associated with the sick mark.

[0013] And, an Internet of Things-based intelligent management method for a rabbit farm using an Internet of Things-based rabbit farm intelligent management system.

[0014] The principle and technical effect of this solution are as follows: First, the breeder can remotely obtain the appearance status of the rabbits by viewing the displayed first information (the overall appearance and facial details of the rabbits) through the first terminal, reducing the time and energy of the breeder shuttling in the rabbit farm for inspections. Moreover, this solution uses image recognition and deep learning technologies to monitor the health status of the rabbits in real time, promptly discovers sick individuals and notifies the breeder. Compared with the traditional method that completely relies on the breeder's inspection and judgment, it reduces the workload of the breeder (no need to check rabbits without diseases anymore, only need to check the rabbits that may be sick screened by this solution), thereby further reducing the time and energy required by the breeder during the process of checking sick rabbits.

[0015] Secondly, this solution can monitor the health status of all rabbits in real time. Compared with the traditional method that requires breeders to check one by one, it can not only improve the work efficiency of checking rabbits, but also more accurately obtain the health status of rabbits through the clear overall appearance and facial details of rabbits, avoiding the situation of missed or incorrect inspections caused by personal factors of breeders and improving the accuracy rate of this solution. Moreover, this meticulous monitoring method can timely detect the sick characteristics of rabbits, realize early warning and rapid response to diseases, and further reduce the cost of this solution in rabbit disease treatment.

[0016] Furthermore, this solution can also judge whether a rabbit is drinking water based on the relative position relationship between the rabbit's contour and the water outlet pipe contour, and determine which rabbit (rabbit number) is drinking water by the rabbit's coat color. Then, it queries the treatment plan that the drinking rabbit is using, and discharges the corresponding medicine into the water the rabbit drinks according to the treatment plan, realizing precise control of rabbit medication. Such a method can not only reduce the time and energy of breeders running around the rabbit farm, preparing medicine, and feeding medicine, reducing labor costs, but also accurately record the dosage and time of rabbit medication in an automated manner, reducing the mortality rate caused by diseases and improving the economic benefits of breeding. Moreover, after the rabbit takes the medicine, the drinking water in the water outlet pipe can flush the medicine outlet pipe clean, which not only avoids the next drinking rabbit being affected by the medicine, reduces the waste of water resources and medicines, but also can clean the residual secretions on the water outlet pipe to a certain extent, reducing the spread of diseases and improving the sanitary quality of the breeding environment.

[0017] Finally, in the process of automated precise medication and management, this solution can store a large amount of breeding data and generate a breeding report through data analysis, providing a scientific basis for breeding management.

[0018] In summary, this solution solves the problem of low efficiency in the traditional breeding method of rabbit farms.

[0019] Further, the processing terminal is communicatively connected to multiple collection terminals. The collection positions of the collection terminals in the rabbit farm are fixed. The processing terminal stores the unique device number of each collection terminal and the rabbit number corresponding to the rabbit contour in the first information. The processing terminal generates a number that does not repeat the stored rabbit number as the rabbit number according to the device number of the collection terminal and the rabbit coat color.

[0020] The coat colors of rabbits are not easily distinguishable. Rabbits with similar coat colors can be separately raised to avoid misjudging rabbits with similar coat colors and ensure that each rabbit has a unique identifier. This method not only avoids management chaos caused by duplicate numbering but also enables precise positioning of the location and identity information of each rabbit through the rabbit number, and reduces the cumbersome process of manual numbering. It further improves the automation and intelligence of this solution, thus enhancing the working efficiency of this solution. Moreover, the method of adding rabbit numbers according to the collection location can adapt to the expansion of the rabbit farm scale and the increase in the number of rabbits. Furthermore, when a rabbit is sick, the first terminal can more conveniently observe and count the new cases of sick rabbits in the collection area, timely detect the spread of the rabbit disease, provide comprehensive breeding information for the breeding manager, and help them make more scientific decisions.

[0021] Furthermore, it includes multiple processing terminals that are communicatively connected to each other. When the processing terminal extracts the rabbit coat color corresponding to the rabbit contour, if there are different rabbit contours with the same rabbit coat color in the first information, it queries the rabbit number in the rabbit database according to the rabbit coat color, and generates an identification request including the rabbit coat color, rabbit number, and the first information and sends it to the first terminal. After receiving and displaying the identification request, the first terminal receives the input difference confirmation or cage separation request and sends the difference confirmation or cage separation request to the processing terminal. If the processing terminal receives the difference confirmation, it learns the rabbit contour and the rabbit coat color corresponding to the rabbit contour in the difference confirmation, then queries the rabbit number in the rabbit database according to the rabbit coat color, and sends the query result to the first terminal. If the processing terminal receives the cage separation request, it generates a cage separation inquiry according to the rabbit coat color and sends it to the communicatively connected processing terminal. After receiving the cage separation inquiry, the processing terminal queries the rabbit number in the rabbit database using the rabbit coat color in the cage separation inquiry. If the rabbit number is not found, it sends the cage separation reception as a response feedback to the processing terminal that sent the cage separation inquiry; otherwise, it sends the rejection reception as a response feedback. After receiving the response feedback, the processing terminal selects a processing terminal with a response feedback of cage separation reception as a cage separation recommendation and sends it to the first terminal.

[0022] When there are rabbits with the same coat color but different outlines in the first piece of information, an identification request is generated and sent to the first terminal for manual confirmation by the breeder, which can avoid recognition errors caused by the same coat color. It also enables the processing terminal to learn from the received discrimination confirmation and optimize the subsequent recognition process, further improving the recognition accuracy of the system. This intelligent learning and collaborative processing method reduces the manual intervention of the breeder in the recognition and cage-dividing operations, improving the working efficiency of this solution. When cage-dividing operations need to be performed on rabbits, this solution collaborates with other processing terminals through a cage-dividing inquiry and response feedback mechanism to ensure the correct cage-dividing of rabbits. Moreover, the cooperation among processing terminals during cage-dividing inquiries can automatically and quickly respond without manual participation. This method improves the response speed of breeding management and can also dynamically update the rabbit information in the database according to the actual situation to ensure the timeliness and accuracy of the data.

[0023] Furthermore, when the processing terminal controls the time for the water outlet device to discharge drinking water into the water outlet pipe according to the first positional relationship, the time for the water outlet device to discharge drinking water into the water outlet pipe is regarded as the drinking time. The drinking time and the corresponding rabbit number are used to generate a drinking record, and the drinking record is used to train the model established by the deep learning algorithm. Then, the rabbit number is put into the model established by the deep learning algorithm to predict the rabbit's drinking time as the drinking prediction time. If the gap between the drinking time obtained from the first piece of information and the drinking prediction time is greater than the preset drinking threshold, the drinking record corresponding to the rabbit number is sent to the first terminal as a drinking anomaly. When the processing terminal is used to record the drinking time, it is also used to record the dosage of the medicine discharged by the medicine outlet device into the medicine outlet pipe.

[0024] Through the detection of drinking anomalies, the system can promptly discover potential health problems of rabbits, such as diseases or stress responses, thus enabling early intervention. Moreover, based on the drinking records and prediction models of each rabbit, the system can provide personalized drinking management solutions for each rabbit, further improving the quality of health management. This solution automatically records and analyzes the drinking behavior of rabbits, reducing the workload of breeders in drinking water monitoring and improving the overall breeding efficiency.

[0025] Further, the processing terminal is also used to compare the shape differences of the rabbit outlines in the first information collected twice adjacent to each other according to the rabbit color pattern. When the compared shape differences are greater than the preset rest differences, the collection time is recorded as the exercise time. The number of exercise time corresponding to the exercise interval time is regularly obtained as the exercise frequency according to the preset exercise interval time, and the exercise frequency and the corresponding rabbit number are used to generate an exercise record; the processing terminal is used to obtain the rabbit number according to the illness mark, obtain the drinking time containing the medicine dosage in the drinking record as the medicine-taking time and the diagnosis and treatment opinion according to the rabbit number, obtain the medicine-taking duration according to the medicine-taking time and the diagnosis and treatment opinion, obtain the exercise frequency in the exercise record according to the medicine-taking duration. If the exercise frequency gradually increases during the medicine-taking duration, a medicine dosage reduction prompt is generated and sent to the first terminal; if the exercise frequency gradually decreases during the medicine-taking duration, a medicine dosage increase prompt is generated and sent to the first terminal; if the exercise frequency gradually decreases three times in a row, an abnormal alarm is generated and sent to the first terminal.

[0026] This solution monitors the exercise frequency of the rabbit through the rabbit outline. During the period when the rabbit is taking medicine due to illness, the use effect of the medicine will be evaluated in combination with the medicine-taking duration and the exercise frequency of the rabbit, and corresponding prompts will be sent to the first terminal of the breeder, reporting the recovery situation of the rabbit after taking medicine to the breeder in a timely manner, reducing the time and effort spent by the breeder in repeatedly checking the recovery situation of the rabbit after taking medicine, and thus improving the working efficiency of this solution.

[0027] Further, the processing terminal compares the rabbit outlines and rabbit color patterns collected twice adjacent to each other according to the first information collected by the collection terminal at a preset specific time, and identifies the change characteristics of the rabbit outlines and rabbit color patterns. The change characteristics and the collection time of the first information are recorded as the growth record, and the rabbit outlines and rabbit color patterns collected most recently in the growth record are used to replace the corresponding data in the rabbit database; the processing terminal also uses the growth record as the learning and training data for identifying and extracting the rabbit outlines and rabbit color patterns.

[0028] By comparing the rabbit outlines and color patterns collected twice adjacent to each other, identifying their change characteristics, and updating the latest characteristic data to the rabbit database, it is ensured that this solution can reflect the growth changes of the rabbit in real time. Moreover, using the growth record as the learning and training data can continuously optimize the processing terminal's recognition of the rabbit outlines and rabbit color patterns, improving the accuracy and adaptability of this solution.

[0029] Further, after the processing terminal sends the cage separation recommendation to the first terminal, it takes the processing terminal corresponding to the cage separation recommendation as the response terminal and sends reservation information to the response terminal; after receiving the reservation information, the response terminal assigns a rabbit number and sends the assigned rabbit number to the first terminal as a virtual number; the first terminal plans a running path for the breeder according to the received virtual number and the collection position corresponding to the virtual number.

[0030] By planning the optimal running path for the breeder, the moving time and physical exertion of the breeder in the rabbit farm are reduced, and the efficiency of the breeder's cage separation operation is improved. Moreover, the first terminal guides the breeder to complete the cage separation operation in real time according to the virtual number and the collection position, which can prevent the breeder from wasting time by searching for the rabbit position. Furthermore, the path planning and real-time guidance provided by the first terminal reduce the human errors in the cage separation operation of the breeder and improve the operation accuracy.

[0031] Further, the processing terminal is also used to encrypt the data stored locally and send it to other processing terminals; the processing terminal is also used to receive the encrypted data sent by other processing terminals and store the encrypted data together with the address of the processing terminal that sent the encrypted data.

[0032] This way can effectively prevent data loss caused by single-point failures and enhance the fault tolerance of this solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of the relevant structures of the first terminal, the processing terminal and the collection terminal in the embodiment of the present invention.

[0034] Figure 2 It is a schematic diagram of the relevant structures of the processing terminal and the second terminal in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] The following will clearly and completely describe the concept of the present invention and the technical effects produced in combination with the embodiments to fully understand the purpose, features and effects of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention:

[0036] Embodiment

[0037] Such as Figure 1 and Figure 2As shown, the intelligent management system for rabbit farms based on the Internet of Things includes: a collection terminal, a processing terminal, a first terminal, and a second terminal. The collection terminal selects a high-definition camera. The first terminal is the terminal used by the breeder, which can be the breeder's mobile phone, tablet, computer, etc. The second terminal is the rabbit cage terminal including a water supply device.

[0038] Among them, the rabbit cage terminal can also be equipped with a feeding device, a temperature monitoring device, a humidity monitoring device, etc., to provide a more comfortable breeding environment for the rabbits. The data monitored by the devices installed on the second terminal can all be sent by the processing terminal to the second terminal.

[0039] The collection terminal collects the image information of the young rabbits as the first information, and the first information includes the overall appearance and facial details of the young rabbits.

[0040] The processing terminal identifies and segments the rabbit contour in the first information, performs feature recognition on the segmented rabbit contour, extracts the rabbit coat color corresponding to the rabbit contour, uses the rabbit coat color to query the rabbit number corresponding to the rabbit coat color in the stored rabbit database, then obtains the rabbit facial coat color according to the rabbit contour, extracts the sick features in the rabbit facial coat color, matches the sick types of the rabbits from the stored rabbit sick knowledge graph according to the extracted sick features, generates a sick mark with the sick type and the rabbit number, and adds the corresponding sick mark to the rabbit contour in the first information.

[0041] The first terminal obtains and displays the first information containing the sick mark, then receives the entered diagnosis and treatment opinions according to the rabbit number, and transmits the diagnosis and treatment opinions and the rabbit number to the processing terminal.

[0042] The processing terminal is also used to obtain the diagnosis and treatment opinions and the rabbit number. When the processing terminal is used to obtain the first information, it obtains the rabbit contour in the first information, identifies and marks the water outlet pipe contour in the first information, and obtains the positional relationship between the rabbit contour and the water outlet pipe contour as the first positional relationship. The processing terminal is used to establish an association relationship between the diagnosis and treatment opinions and the sick mark.

[0043] The second terminal includes a water outlet pipe and a medicine outlet pipe. One end of the water outlet pipe is connected to a water outlet device, and the medicine outlet pipe is connected to a medicine outlet device. The water outlet pipe is wrapped outside the medicine outlet pipe. The second terminal is used to obtain the first positional relationship and control the time for the water outlet device to discharge drinking water into the water outlet pipe according to the first positional relationship; it is also used to obtain the sick mark in the rabbit contour when obtaining the first positional relationship, and control the dosage of the medicine discharged by the medicine outlet device into the medicine outlet pipe according to the diagnosis and treatment opinions associated with the sick mark.

[0044] Among them, the processing terminal is communicatively connected to multiple collection terminals. The collection positions of the collection terminals in the rabbit farm are fixed. The processing terminal stores the unique device numbers of each collection terminal and the rabbit numbers corresponding to the rabbit outlines in the first information. The processing terminal generates a number that does not duplicate the stored rabbit numbers as the rabbit number according to the device number of the collection terminal and the rabbit coloration.

[0045] Among them, as Figure 1 or Figure 2 shown, the processing terminals are communicatively connected to each other. When the processing terminal extracts the rabbit coloration corresponding to the rabbit outline, if there are different rabbit outlines with the same rabbit coloration in the first information, it queries the rabbit number in the rabbit database according to the rabbit coloration, and generates an identification request with the rabbit coloration, rabbit number, and the first information and sends it to the first terminal; after receiving and displaying the identification request, the first terminal receives the entered difference confirmation or cage separation request, and sends the difference confirmation or cage separation request to the processing terminal.

[0046] If the processing terminal receives the difference confirmation, it learns the rabbit outline and the rabbit coloration corresponding to the rabbit outline in the difference confirmation, then queries the rabbit number in the rabbit database according to the rabbit coloration, and sends the query result to the first terminal; if the processing terminal receives the cage separation request, it generates a cage separation query according to the rabbit coloration and sends it to the communicatively connected processing terminal; after receiving the cage separation query, the processing terminal uses the rabbit coloration in the cage separation query to query the rabbit number in the rabbit database. If the rabbit number is not found, it sends the cage separation reception as a response feedback to the processing terminal that sent the cage separation query; otherwise, it sends the rejection reception as a response feedback; after receiving the response feedback, the processing terminal selects a processing terminal with a response feedback of cage separation reception as the cage separation recommendation and sends it to the first terminal.

[0047] Among them, when the processing terminal controls the time for the water outlet device to discharge drinking water into the water outlet pipe according to the first positional relationship, it takes the time for the water outlet device to discharge drinking water into the water outlet pipe as the drinking time, generates a drinking record with the drinking time and the corresponding rabbit number, uses the drinking record to train the model established by the deep learning algorithm, and then puts the rabbit number into the model established by the deep learning algorithm to predict the rabbit drinking time as the drinking prediction time. If the gap between the drinking time obtained from the first information and the drinking prediction time is greater than the preset drinking threshold (which needs to be set by the breeder according to the weight of the young rabbits), it sends the drinking record corresponding to the rabbit number as a drinking anomaly to the first terminal. When the processing terminal is used to record the drinking time, it is also used to record the dose of the medicine discharged by the medicine outlet device into the medicine outlet pipe.

[0048] Among them, the processing terminal is also used to compare the shape differences of the rabbit outlines in the first information collected twice adjacent to each other according to the rabbit color pattern. When the compared shape difference is greater than the preset rest difference (the default is 60%, specifically set by the administrator), the collection time is recorded as the exercise time. According to the preset exercise interval time (generally 2 hours, specifically can be set by the breeder according to the individual situation of the young rabbit), the number of exercise time corresponding to the exercise interval time is regularly obtained as the exercise frequency, and the exercise frequency and the corresponding rabbit number are used to generate an exercise record;

[0049] The processing terminal is used to obtain the rabbit number according to the illness mark, obtain the drinking time containing the drug dose in the drinking record as the medicine-taking time and the diagnosis and treatment opinion according to the rabbit number, obtain the medicine-taking duration according to the medicine-taking time and the diagnosis and treatment opinion, obtain the exercise frequency in the exercise record according to the medicine-taking duration. If the exercise frequency gradually increases during the medicine-taking duration, a medicine dose reduction prompt is generated and sent to the first terminal; if the exercise frequency gradually decreases during the medicine-taking duration, a medicine dose increase prompt is generated and sent to the first terminal; if the exercise frequency gradually decreases three times in a row, an abnormal alarm is generated and sent to the first terminal.

[0050] Among them, the processing terminal compares the rabbit outline and the rabbit color pattern collected twice adjacent to each other according to the first information collected by the collection terminal at a preset specific time (generally a fixed time every day, specifically set by the administrator. When the young rabbit is relatively young, the administrator can set it to two or even more fixed times every day. When the young rabbit is older, the administrator can set the interval between the growth specific times), and identifies the change characteristics of the rabbit outline and the rabbit color pattern, records the change characteristics and the collection time of the first information as the growth record, and uses the rabbit outline and the rabbit color pattern collected most recently in the growth record to replace the corresponding data in the rabbit database; the processing terminal also uses the growth record as the learning and training data for identifying and extracting the rabbit outline and the rabbit color pattern.

[0051] Among them, after the processing terminal sends the cage separation recommendation to the first terminal, it takes the processing terminal corresponding to the cage separation recommendation as the response terminal and sends the reservation information to the response terminal; after receiving the reservation information, the response terminal assigns a rabbit number and sends the assigned rabbit number as a virtual number to the first terminal; the first terminal plans the running path for the breeder according to the received virtual number and the collection location corresponding to the virtual number.

[0052] Among them, the processing terminal is also used to encrypt the data stored locally and send it to other processing terminals; the processing terminal is also used to receive the encrypted data sent by other processing terminals and store the encrypted data together with the address of the processing terminal that sent the encrypted data.

[0053] This embodiment also includes an Internet of Things-based rabbit farm intelligent management method using an Internet of Things-based rabbit farm intelligent management system.

[0054] During specific implementation:

[0055] A rabbit farm uses an Internet of Things-based rabbit farm intelligent management system and method to assist in managing the rabbit farm. An automatic water supply device is installed on each rabbit cage (second terminal), and a personal work mobile phone (first terminal) is configured for each breeder. The rabbit cages are divided into multiple breeding areas according to geographical locations, and a fixed high-definition camera device (acquisition terminal) is configured for each breeding area. The processing terminal is deployed in the central control room of the rabbit farm.

[0056] Every morning, when the preset specific time arrives, each acquisition terminal starts to collect image information of the young rabbits according to the established acquisition directions. The first information captured by the high-definition camera is quickly transmitted to the processing terminal through the network. The processing terminal operates efficiently, immediately identifies and segments the rabbit contours in the first information, and accurately extracts the rabbit coat colors.

[0057] During one acquisition process, the processing terminal discovers that two rabbits have the same coat color (it is possible that rabbits from other areas bit through the cage and ran into the rabbit cage in this area). Immediately, it queries the rabbit database for the rabbit numbers, and generates an identification request including the rabbit coat color, rabbit number, and the first information, and sends it to the first terminal. After receiving the identification request, breeder Li carefully observes the image displayed on the mobile phone and, relying on rich experience, accurately makes a differentiation confirmation. After receiving the differentiation confirmation, the processing terminal learns the relevant information and feeds back the query result to Li.

[0058] Meanwhile, the processing terminal is also monitoring various data of the rabbit cage terminal in real time. When obtaining the positional relationship between the rabbit contour and the water outlet pipe contour, it controls the water supply device to discharge drinking water into the water outlet pipe according to the first positional relationship. For example, when it is found that a young rabbit numbered 005 approaches the water outlet pipe, the processing terminal instructs the water supply device to discharge an appropriate amount of drinking water within the next 30 seconds, and generates a drinking record with this drinking time and the rabbit number 005.

[0059] In daily management, the breeder sets a suitable drinking threshold on the first terminal according to the weight of the young rabbits. One day, the processing terminal predicts through the model established by the deep learning algorithm that the drinking time of the young rabbit numbered 008 should be around 10 am, but the actual collected drinking time is 11:30 am, and the difference between the two is greater than the preset drinking threshold. The processing terminal immediately sends the drinking record corresponding to the rabbit number 008 as a drinking anomaly to the mobile phone of breeder Zhang. After receiving the information, Zhang views the monitoring images related to 008 through the mobile phone, decides to go to the rabbit cage area to check the health status of the young rabbit numbered 008, and timely adjusts the water supply strategy.

[0060] As the young rabbits grow, the administrator will reasonably adjust the preset specific time according to the age of the young rabbits. When the young rabbits are relatively young, the administrator sets the preset specific time to 8:00 am and 3:00 pm every day to pay closer attention to the growth changes of the young rabbits. After the first piece of information is collected each time, the processing terminal will compare the rabbit outlines and rabbit coat colors collected in two adjacent times, identify the change characteristics, and record them as growth records. These growth records are not only used to update the data in the rabbit database but also serve as learning and training data to continuously optimize the ability of the processing terminal to identify and extract rabbit outlines and rabbit coat colors.

[0061] In terms of disease prevention and control, when the processing terminal extracts the sick characteristics in the rabbit's facial coat color, it will match the rabbit's sick type from the stored rabbit sick knowledge graph and generate a sick mark. After receiving the first piece of information containing the sick mark, the breeder Wang carefully checks the rabbit number and the sick type, and combines the actual situation to enter detailed diagnosis and treatment opinions on the first terminal. After receiving the diagnosis and treatment opinions, the processing terminal establishes an association relationship between the diagnosis and treatment opinions and the sick mark, and controls the medicine dispensing device to discharge the corresponding dose of medicine into the medicine pipe according to the sick mark.

[0062] In terms of exercise management, the processing terminal will compare the shape differences of the rabbit outlines in the first piece of information collected in two adjacent times according to the rabbit coat color. When the shape difference of a young rabbit numbered 012 is greater than the preset rest difference, the processing terminal records the collection time as the exercise time. After a period of monitoring, the processing terminal counts that the young rabbit has 5 exercise times within a 2-hour exercise interval, and thus calculates the exercise frequency. During the treatment of young rabbit No. 012, the processing terminal finds that its exercise frequency gradually decreases during the medicine-taking duration, and then generates a medicine dosage increase prompt and sends it to the breeder Zhao to adjust the treatment plan in time.

[0063] When the processing terminal sends a cage separation inquiry to other processing terminals according to the breeder's cage separation request, if a certain processing terminal does not query the numbers of rabbits with the same coat color, it will send back the cage separation reception as a reply feedback. After receiving the reply feedback, the processing terminal selects one of them as the cage separation recommendation and sends it to the first terminal. Subsequently, the processing terminal selected as the reply terminal assigns a rabbit number and sends the virtual number to the first terminal. The breeder plans the most reasonable running path according to the received virtual number and the collection location, and efficiently completes the cage separation work.

[0064] In addition, to ensure data security, the processing terminal regularly encrypts the locally stored data and sends it to other processing terminals for backup. At the same time, it also receives and stores the encrypted data and the sending addresses sent by other processing terminals. During a system upgrade, the technicians successfully completed data migration and system update through these backup data, ensuring the continuous and stable operation of the intelligent management system for the rabbit farm.

[0065] The above are only embodiments of the present invention, and common knowledge such as the specific structures and characteristics known in the solution is not described in detail herein. It should be noted that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the implementation effect of the present invention and the practicality of the patent. The protection scope required by this application should be based on the content of its claims, and the specific implementation manners described in the specification can be used to interpret the content of the claims.

Claims

1. The intelligent management system of rabbit farm based on Internet of Things is characterized by: include: A collection terminal collects image information of the baby rabbit as first information, wherein the first information includes an overall appearance and facial details of the baby rabbit; The processing terminal identifies and segments the rabbit outline in the first information, performs feature recognition on the segmented rabbit outline, extracts the rabbit color corresponding to the rabbit outline, uses the rabbit color in combination with a stored rabbit database to query the rabbit number corresponding to the rabbit color, obtains the rabbit facial color according to the rabbit outline, extracts the disease features in the rabbit facial color, matches the rabbit's disease type from the stored rabbit disease knowledge graph according to the extracted disease features, generates a disease mark from the disease type and the rabbit number, and adds the corresponding disease mark to the rabbit outline in the first information; The first terminal obtains and displays the first information containing the sick mark, receives the entered diagnosis and treatment opinions according to the rabbit number, and transmits the diagnosis and treatment opinions and the rabbit number to the processing terminal; The processing terminal is also used to obtain the diagnosis and treatment opinion and the rabbit number. When the processing terminal is used to obtain the first information, the rabbit outline in the first information is obtained, the water outlet pipe outline in the first information is identified and marked, and the positional relationship between the rabbit outline and the water outlet pipe outline is obtained as the first positional relationship. The processing terminal is used to establish an association relationship between the diagnosis and treatment opinion and the disease mark; The second terminal includes a water outlet pipe and a medicine outlet pipe, one end of the water outlet pipe is connected to a water outlet device, the medicine outlet pipe is connected to a medicine outlet device, and the water outlet pipe is wrapped around the outside of the medicine outlet pipe; the second terminal is used to obtain the first position relationship, and control the time for the water outlet device to discharge drinking water into the water outlet pipe according to the first position relationship; it is also used to obtain the sick mark in the rabbit outline when obtaining the first position relationship, and control the dosage of the medicine discharged into the medicine outlet pipe by the medicine outlet device according to the diagnosis and treatment opinions associated with the sick mark.

2. The rabbit farm intelligent management system based on the Internet of Things according to claim 1 is characterized in that: The processing terminal is communicatively connected with a plurality of collecting terminals, the collecting positions of the collecting terminals in the rabbit farm are fixed, the processing terminal stores the unique device number of each collecting terminal and the rabbit number corresponding to the rabbit outline in the first information, and the processing terminal generates a number that does not overlap with the stored rabbit number as the rabbit number according to the device number of the collecting terminal and the rabbit color.

3. The rabbit farm intelligent management system based on the Internet of Things according to claim 2 is characterized in that: The method comprises a plurality of processing terminals, each of which is in communication connection with each other; when the processing terminal extracts the rabbit color corresponding to the rabbit outline, if there are different rabbit outlines with the same rabbit color in the first information, the rabbit number is queried in the rabbit database according to the rabbit color, and an identification request is generated by the rabbit color, the rabbit number and the first information and sent to the first terminal; after the first terminal receives and displays the identification request, it receives the inputted distinction confirmation or cage separation request, and sends the distinction confirmation or cage separation request to the processing terminal; If the processing terminal receives the distinction confirmation, it learns the rabbit outline and the rabbit color corresponding to the rabbit outline in the distinction confirmation, and then searches the rabbit database for the rabbit number according to the rabbit color, and sends the query result to the first terminal; If the processing terminal receives a cage separation request, it generates a cage separation query according to the rabbit color and sends it to the processing terminal connected to the communication; after receiving the cage separation query, the processing terminal uses the rabbit color in the cage separation query to query the rabbit number in the rabbit database. If the rabbit number is not found, the cage separation reception is sent as a response feedback to the processing terminal that sent the cage separation query; otherwise, the rejection of reception is sent as a response feedback; after receiving the response feedback, the processing terminal selects a processing terminal whose response feedback is cage separation reception as a cage separation recommendation and sends it to the first terminal.

4. The rabbit farm intelligent management system based on the Internet of Things according to claim 3 is characterized in that: When the processing terminal controls the time when the water outlet device discharges drinking water into the water outlet pipe according to the first position relationship, the time when the water outlet device discharges drinking water into the water outlet pipe is used as the drinking time, and a drinking record is generated based on the drinking time and the corresponding rabbit number. The drinking record is used to train the model established by the deep learning algorithm, and then the rabbit number is put into the model established by the deep learning algorithm to predict the rabbit's drinking time as the predicted drinking time. If the difference between the drinking time and the predicted drinking time obtained in the first information is greater than a preset drinking threshold, the drinking record corresponding to the rabbit number is sent to the first terminal as a drinking anomaly. The processing terminal is used to record the drinking time and also to record the dosage of the medicine discharged by the medicine dispensing device into the medicine dispensing pipe.

5. The rabbit farm intelligent management system based on the Internet of Things according to claim 4 is characterized in that: The processing terminal is also used to compare the shape difference of the rabbit outline in the first information collected twice adjacently according to the rabbit color, and when the compared shape difference is greater than the preset rest difference, record the collection time as the exercise time, regularly obtain the number of exercise time items corresponding to the exercise interval time as the number of exercises according to the preset exercise interval time, obtain the exercise frequency according to the exercise interval time and the number of exercises, and generate an exercise record with the exercise frequency and the corresponding rabbit number; The processing terminal is used to obtain the rabbit number according to the sick mark, obtain the drinking time containing the medicine dosage in the drinking record according to the rabbit number as the medicine taking time and treatment advice, obtain the medicine taking duration by combining the medicine taking time and the treatment advice, obtain the exercise frequency in the exercise record according to the medicine taking duration, if the exercise frequency gradually increases during the medicine taking duration, generate a medicine reduction prompt and send it to the first terminal; if the exercise frequency gradually decreases during the medicine taking duration, generate a medicine increase prompt and send it to the first terminal; if the exercise frequency gradually decreases for three consecutive times, generate an abnormal alarm and send it to the first terminal.

6. The rabbit farm intelligent management system based on the Internet of Things according to claim 5 is characterized in that: The processing terminal compares the rabbit outlines and rabbit colors collected twice adjacently based on the first information collected by the collection terminal at a preset specific time, identifies the change characteristics of the rabbit outlines and rabbit colors, records the change characteristics and the collection time of the first information as a growth record, and uses the rabbit outlines and rabbit colors collected most recently in the growth record to replace the corresponding data in the rabbit database; the processing terminal also uses the growth record as learning training data for identifying and extracting rabbit outlines and rabbit colors.

7. The rabbit farm intelligent management system based on the Internet of Things according to claim 6 is characterized in that: After sending the cage recommendation to the first terminal, the processing terminal corresponding to the cage recommendation is used as a response terminal, and the reservation information is sent to the response terminal; after receiving the reservation information, the response terminal assigns a rabbit number, and sends the assigned rabbit number as a virtual number to the first terminal; the first terminal plans a running path for the breeder according to the received virtual number and the collection position corresponding to the virtual number.

8. The intelligent management system for rabbit farms based on Internet of Things according to claim 7 is characterized in that: The processing terminal is also used to encrypt the locally stored data and send it to other processing terminals; the processing terminal is also used to receive encrypted data sent by other processing terminals, and store the encrypted data together with the address of the processing terminal that sent the encrypted data.

9. The intelligent management method of rabbit farm based on Internet of Things is characterized by: The rabbit farm intelligent management system based on the Internet of Things described in any one of claims 1 to 8 is used.

Images