Farm environment control system and method based on computer vision

Abstract

The invention discloses a farm environment control system and method based on computer vision, and relates to the technical field of environment control, the farm environment control system comprises an information acquisition module, an activity area health analysis module, a shower area health analysis module and a display terminal, the information acquisition module acquires video and environment data of each activity area of dairy cattle, and displays the video and environment data on the display terminal; the system achieves the monitoring of the environment state, judges whether the environment state meets the breeding requirements or not, carries out the automatic adjustment if not, judges whether each activity area of the dairy cow needs to be cleaned or not through an activity area health analysis module, and further determines the water spraying rate and water spraying amount of each activity area of the dairy cow. Whether the shower system of each shower area of the dairy cow is started or not is judged, and the stay duration of each dairy cow in each shower area of the dairy cow is analyzed, so that the control of the system is more intelligent and accurate, and the sanitary level and the production efficiency of a farm are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental control, and in particular to a farm environmental control system and method based on computer vision. Background Art

[0002] As people's living standards improve, the demand for milk and dairy products continues to increase, which requires dairy farms to improve production efficiency and milk quality. Good environmental control helps improve the production performance and milk quality of dairy cows, thereby better meeting market demand. Traditional farm environmental control usually uses timers or simple manual operations to control the opening and closing of equipment. This method cannot be intelligently adjusted according to the actual behavior of dairy cows and changes in the breeding environment, resulting in energy waste of equipment and inefficient cow health management. Therefore, it is particularly important to invent a farm environmental control system and method based on computer vision.

[0003] Prior art, such as the invention application patent with announcement number: CN111417304A, discloses a feeding place management device and method, which includes: a shooting unit shoots the feeding place; a controller detects abnormal signs in the feeding place by comparing the behavior pattern in the image data obtained by the shooting unit with the behavior pattern in the normal image data; a communicator transmits information about the abnormal signs to an administrator terminal; the controller extracts a region of interest including the abnormal signs from the obtained image data; the controller transmits information about the region of interest to the shooting unit; the controller receives image data obtained by shooting the region of interest from the shooting unit; and the controller extracts an abnormal region narrower than the region of interest from the behavior pattern in the image data obtained by shooting the region of interest.

[0004] The existing technology also has the following defects, which are specifically reflected in: 1. In the existing technology, not much attention is paid to the water spraying rate and amount in the activity area. Inappropriate water spraying rate and amount cannot completely wash away stains such as feces, sewage and feed residues in the activity area, resulting in a large amount of pollutants remaining on the ground. Long-term accumulation will breed bacteria, viruses and parasites, increase the risk of dairy cows contracting diseases, and affect the health of dairy cows.

[0005] 2. In the prior art, not much attention is paid to the cleanliness of dairy cows. The udders and surrounding skin of dairy cows are not clean. Bacteria can easily enter the milk during milking, which reduces the quality of the milk, shortens the shelf life of the milk, and affects the quality and safety of the milk. Summary of the invention

[0006] The purpose of the present invention is to provide a farm environment control system and method based on computer vision, which solves the problems existing in the background technology.

[0007] In order to solve the above technical problems, the present invention adopts the following technical solutions: The present invention provides a farm environment control system based on computer vision, including: an information acquisition module, which obtains videos of various activity areas of dairy cows through high-definition cameras, collects data of various activity areas of dairy cows, obtains environmental data of various activity areas of dairy cows, and determines whether various activity areas of dairy cows meet the needs of dairy cow breeding. If the environmental data of a certain activity area of ​​dairy cows does not meet the needs of dairy cow breeding, the relevant equipment automatically adjusts.

[0008] The activity area health analysis module evaluates the environmental cleanliness index of each activity area of ​​the cow based on the environmental data of each activity area of ​​the cow. Based on the environmental cleanliness index of each activity area of ​​the cow, it determines whether each activity area of ​​the cow needs to be sprayed with water for cleaning. If a certain activity area of ​​the cow needs to be sprayed with water for cleaning, the cleaning difficulty coefficient of each activity area of ​​the cow is evaluated, and the spraying rate and amount of water in each activity area of ​​the cow are determined.

[0009] The shower area health analysis module obtains videos of each cow's shower area through high-definition cameras, collects data from each cow's shower area, obtains the location of each cow through GPS, evaluates the behavioral health coefficient of each cow in each shower area, evaluates the skin cleanliness index of each cow in each shower area, and then determines whether the shower system in each cow's shower area is started. If the shower system in each cow's shower area is not started, the length of time each cow stays in each shower area is analyzed.

[0010] The display terminal displays the video of each activity area of ​​the cows, displays the environmental cleanliness index of each activity area of ​​the cows, displays the water spraying rate and water spraying amount of each activity area of ​​the cows, and displays the video of each shower area of ​​the cows.

[0011] Preferably, the specific analysis method for evaluating the environmental cleanliness index of each activity area of ​​the dairy cow is as follows: From the videos of the cows’ various activity areas, we extracted the feces area, sewage area, and feed residue area, and then evaluated the environmental cleanliness index of the cows’ various activity areas. The calculation formula is: ,in Indicates the cow Environmental cleanliness index of each activity area, , , Respectively represent the cow The area of ​​feces, sewage, and feed residue in each activity area. , , They represent the suitable feces area, sewage area and feed residue area stored in the database respectively. , , They represent the appropriate feces area weight factor, the appropriate sewage area weight factor, and the appropriate feed residue area weight factor stored in the database, respectively; e represents a natural constant, Indicates the number of the active area, , is a positive integer greater than 2.

[0012] Preferably, the specific analysis method for judging whether each activity area of ​​the cow needs to be sprayed with water for cleaning is as follows: The environmental cleanliness index of each activity area of ​​the cows is extracted and compared with the appropriate environmental cleanliness index stored in the database. If the environmental cleanliness index of a certain activity area of ​​the cows is higher than the appropriate environmental cleanliness index stored in the database, it is determined that the activity area needs to be sprayed with water for cleaning; if the environmental cleanliness index of a certain activity area of ​​the cows is lower than the appropriate environmental cleanliness index stored in the database, it is determined that the activity area does not need to be sprayed with water for cleaning.

[0013] Preferably, the cleaning difficulty coefficient of each activity area of ​​the cows is evaluated by a specific analysis method: the area and ground material of each activity area of ​​the cows are extracted from the database, the cleaning difficulty coefficient of the ground material is obtained by analysis, and the cleaning difficulty coefficient of each activity area of ​​the cows is evaluated in combination with the environmental cleaning index of each activity area of ​​the cows. The calculation formula is: ,in Indicates the cow The cleaning difficulty coefficient of each activity area, , Respectively represent the cow The area of ​​each activity area and the cleaning difficulty of the floor material, , They represent the area of ​​the suitable activity area and the cleaning difficulty coefficient of the suitable floor material stored in the database respectively. , They respectively represent the area weight factor of the suitable activity area and the cleaning difficulty coefficient weight factor of the suitable floor material stored in the database.

[0014] Preferably, the specific analysis method for determining the water spraying rate and water spraying amount in each activity area of ​​the cow is as follows: The cleaning difficulty coefficient of each activity area of ​​the cow is extracted, and compared with the cleaning difficulty coefficient range corresponding to each water spray rate and the cleaning difficulty coefficient range corresponding to each water spray amount stored in the database. If the cleaning difficulty coefficient of a certain activity area of ​​the cow is within the cleaning difficulty coefficient range corresponding to a certain water spray rate, the water spray rate is used as the water spray rate of the activity area of ​​the cow, and the water spray rate of each activity area of ​​the cow is obtained. If the cleaning difficulty coefficient of a certain activity area of ​​the cow is within the cleaning difficulty coefficient range corresponding to a certain water spray amount, the water spray amount is used as the water spray amount of the activity area of ​​the cow, and the water spray amount of each activity area of ​​the cow is obtained.

[0015] Preferably, the behavioral health coefficient of each cow in each shower area of ​​the cows is evaluated, and the specific analysis method is: based on the video of each shower area of ​​the cows, extraction is performed according to a preset frequency to obtain each picture, and each body contour of each cow in each picture is extracted, and it is compared with each body contour of each cow's scratching behavior preset in the database; if a body contour of a cow in a certain picture is consistent with a body contour of a cow's scratching behavior preset in the database, then the cow is determined to be in a scratching state, and the number of pictures in which the cow has the scratching behavior is obtained by statistics, and the scratching frequency and scratching duration of the cow are obtained by analysis, and then the scratching frequency of each cow in each shower area of ​​the cow is obtained, and the scratching duration of each cow in each shower area of ​​the cow is obtained.

[0016] Based on the videos of each shower area for cows, the tail-flicking frequency of each cow in each shower area was analyzed, and the behavioral cleaning coefficient of each cow in each shower area was evaluated. The calculation formula is: ,in Indicates Shower area The behavioral cleanliness coefficient of each cow, , , Respectively represent Shower area The frequency, duration and tail-flicking frequency of each cow. , , They represent the appropriate cow scratching frequency, the appropriate cow scratching duration, and the appropriate cow tail flicking frequency stored in the database, respectively. Indicates the number of each shower area, , is a positive integer greater than 2, Indicates the number of each shower area, , is a positive integer greater than 2, , , They respectively represent the appropriate cow scratching frequency weight factor, the appropriate cow scratching duration weight factor, and the appropriate cow tail flicking frequency weight factor stored in the database.

[0017] Preferably, the specific analysis method for evaluating the skin cleanliness index of each cow in each shower area is as follows: The stain area and number of stains on the udder of each cow in each shower area were extracted from the video, and the skin cleanliness index of each cow in each shower area was evaluated. The calculation formula is: ,in Indicates Shower area Skin cleanliness index of a cow, , Respectively represent Shower area The stain area and number of stains on the udder of each cow, , They respectively represent the stain area of ​​a suitable cow's udder and the number of stains of a suitable cow's udder stored in the database.

[0018] Preferably, the specific analysis method for judging whether the shower system of each shower area is started is as follows: extracting the behavior cleaning coefficient of each cow in each shower area and the skin cleaning index of each cow in each shower area, and comparing them with the behavior cleaning coefficient of a suitable cow and the skin cleaning index of a suitable cow stored in a database, respectively; if the behavior cleaning coefficient of a cow in a certain shower area is higher than the behavior cleaning coefficient of a suitable cow stored in the database, then it is determined that the cow needs to be cleaned; if the skin cleaning index of a cow in a certain shower area is higher than the skin cleaning index of a suitable cow stored in the database, then it is determined that the cow needs to be cleaned; the number of cows that need to be cleaned in each shower area is obtained by counting; the total number of cows in each shower area is counted and compared; the percentage of cows that need to be cleaned in each shower area is calculated; and it is compared with the percentage of cows that need to be cleaned stored in the database; if the percentage of cows that need to be cleaned in a certain shower area is higher than the percentage of cows that need to be cleaned stored in the database, then it is determined that the shower system of the shower area is started.

[0019] Preferably, the staying time of each cow in each shower area is analyzed, and the specific analysis method is: based on the video of each shower area and the position of each cow, the staying time of each cow in each shower area is extracted, and input into the shower system control formula, and the formula expression is: ,like , then it means If the total length of time that a cow stays in each shower area reaches the shower system opening standard, , then it means The total time that the cows stay in each shower area reaches the shower system opening standard. Indicates Shower area The length of stay of a cow, Indicates the total suitable duration of stay stored in the database. Indicates the number of shower areas.

[0020] A second aspect of the present invention provides a method for a farm environment control system based on computer vision, comprising: Step 1: Information acquisition: Use high-definition cameras to obtain videos of various activity areas of dairy cows, collect data from various activity areas of dairy cows, obtain environmental data from various activity areas of dairy cows, and determine whether various activity areas of dairy cows meet the needs of dairy cow breeding. If the environmental data of a certain activity area of ​​dairy cows does not meet the needs of dairy cow breeding, the relevant equipment will automatically adjust.

[0021] Step 2: Activity area health analysis: Based on the environmental data of each activity area of ​​the cow, the environmental cleanliness index of each activity area of ​​the cow is evaluated. Based on the environmental cleanliness index of each activity area of ​​the cow, it is determined whether each activity area of ​​the cow needs to be sprayed with water for cleaning. If a certain activity area of ​​the cow needs to be sprayed with water for cleaning, the cleaning difficulty coefficient of each activity area of ​​the cow is evaluated, and the spraying rate and amount of water in each activity area of ​​the cow are determined.

[0022] Step 3: Shower area health analysis: obtain videos of each cow's shower area through high-definition cameras, collect data from each cow's shower area, obtain the location of each cow through GPS, evaluate the behavioral health coefficient of each cow in each shower area, evaluate the skin cleanliness index of each cow in each shower area, and then determine whether the shower system in each cow's shower area is started. If the shower system in each cow's shower area is not started, analyze the length of time each cow stays in each shower area.

[0023] The beneficial effects of the present invention are: 1. In the present invention, more attention is paid to the water spraying rate and water spraying amount in the activity area. The appropriate water spraying rate and water spraying amount can ensure that the water flow has sufficient impact force on the stains such as feces, sewage and feed residues in the activity area, making it easier to wash them away, thereby improving the cleaning efficiency, reducing the residual stains, ensuring that the cows are in a clean living environment, and reducing the risk of cows contracting diseases.

[0024] 2. In the present invention, more attention is paid to the cleaning of dairy cows. The udders and surrounding skin of dairy cows are cleaned in time, and it is difficult for bacteria to enter the milk during milking, which improves the quality of milk, extends the shelf life of milk, improves the quality and safety of milk, enhances the brand reputation of the farm, and enhances consumers' trust in their products. At the same time, a voting mechanism is adopted, and the system can more accurately determine when to start the spraying equipment. If the cow disappears temporarily in the shadow, the system will not shut down the equipment immediately, but continue to judge the behavior through the cumulative scoring mechanism, which increases the fault tolerance rate and ensures that the cow can be cleaned in time when it really needs a shower. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0026] Figure 1 It is a schematic diagram of the system structure connection of the present invention.

[0027] Figure 2 The present invention is a schematic flow chart of the steps for implementing the method. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] Reference Figure 1 As shown, the present invention provides a farm environment control system based on computer vision, including: an information acquisition module, which acquires videos of various activity areas of dairy cows through high-definition cameras, collects data of various activity areas of dairy cows, acquires environmental data of various activity areas of dairy cows, and determines whether various activity areas of dairy cows meet the needs of dairy cow breeding. If the environmental data of a certain activity area of ​​dairy cows does not meet the needs of dairy cow breeding, the relevant equipment automatically adjusts.

[0030] It should be noted that the determination of whether the various activity areas of the cows meet the requirements of dairy farming is performed by respectively obtaining the temperature, humidity and harmful gas concentration of each activity area of ​​the cows through temperature sensors, humidity sensors and gas sensors. The harmful gases include ammonia, hydrogen sulfide, carbon dioxide and the like. The temperature, humidity and harmful gas concentration of each activity area of ​​the cows are extracted and compared with the appropriate temperature range, the appropriate humidity range and the appropriate harmful gas concentration range stored in the database. If the temperature of a certain activity area of ​​the cows is not within the appropriate temperature range stored in the database, the humidity is not within the appropriate humidity range stored in the database or the harmful gas concentration is not within the appropriate harmful gas concentration range stored in the database, then it is determined that the activity area of ​​the cows does not meet the requirements of dairy farming, and the relevant equipment is automatically adjusted. If the temperature of a certain activity area of ​​the cows does not meet the requirements of dairy farming, the air conditioner of the activity area automatically starts to run until the temperature of the activity area meets the requirements of dairy farming. If the humidity and harmful gas concentration of a certain activity area of ​​the cows do not meet the requirements of dairy farming, the exhaust fan of the activity area automatically starts to run until the humidity and harmful gas concentration of the activity area meet the requirements of dairy farming.

[0031] The activity area health analysis module evaluates the environmental cleanliness index of each activity area of ​​the cow based on the environmental data of each activity area of ​​the cow. Based on the environmental cleanliness index of each activity area of ​​the cow, it determines whether each activity area of ​​the cow needs to be sprayed with water for cleaning. If a certain activity area of ​​the cow needs to be sprayed with water for cleaning, the cleaning difficulty coefficient of each activity area of ​​the cow is evaluated, and the spraying rate and amount of water in each activity area of ​​the cow are determined.

[0032] In a specific embodiment, the environmental cleanliness index of each activity area of ​​the dairy cow is evaluated by a specific analysis method: extracting the feces area, sewage area, and feed residue area from the video of each activity area of ​​the dairy cow, and then evaluating the environmental cleanliness index of each activity area of ​​the dairy cow. The calculation formula is: ,in Indicates the cow Environmental cleanliness index of each activity area, , , Respectively represent the cow The area of ​​feces, sewage, and feed residue in each activity area. , , They represent the suitable feces area, sewage area and feed residue area stored in the database respectively. , , They represent the appropriate feces area weight factor, the appropriate sewage area weight factor, and the appropriate feed residue area weight factor stored in the database, respectively; e represents a natural constant, Indicates the number of the active area, , is a positive integer greater than 2.

[0033] It should be noted that the suitable feces area, suitable sewage area, and suitable feed residue area stored in the database, and the suitable feces area weight factor, suitable sewage area weight factor, and suitable feed residue area weight factor stored in the database are set by professionals.

[0034] In a specific embodiment, the determination of whether each activity area of ​​the cow needs to be sprayed with water for cleaning is performed by a specific analysis method as follows: extracting the environmental cleanliness index of each activity area of ​​the cow, and comparing it with the appropriate environmental cleanliness index stored in the database; if the environmental cleanliness index of a certain activity area of ​​the cow is higher than the appropriate environmental cleanliness index stored in the database, then it is determined that the activity area needs to be sprayed with water for cleaning; if the environmental cleanliness index of a certain activity area of ​​the cow is lower than the appropriate environmental cleanliness index stored in the database, then it is determined that the activity area does not need to be sprayed with water for cleaning.

[0035] It should be noted that the appropriate environmental cleanliness index stored in the database is set by professionals.

[0036] In a specific embodiment, the cleaning difficulty coefficient of each activity area of ​​the cow is evaluated by a specific analysis method: the area and ground material of each activity area of ​​the cow are extracted from the database, the cleaning difficulty coefficient of the ground material is obtained by analysis, and the cleaning difficulty coefficient of each activity area of ​​the cow is evaluated in combination with the environmental cleaning index of each activity area of ​​the cow. The calculation formula is: ,in Indicates the cow The cleaning difficulty coefficient of each activity area, , Respectively represent the cow The area of ​​each activity area and the cleaning difficulty of the floor material, , They represent the area of ​​the suitable activity area and the cleaning difficulty coefficient of the suitable floor material stored in the database. , They respectively represent the area weight factor of the suitable activity area and the cleaning difficulty coefficient weight factor of the suitable floor material stored in the database.

[0037] It should be noted that the area of ​​suitable activity areas and the cleaning difficulty coefficient of suitable ground materials stored in the database, the area weight factors of suitable activity areas and the cleaning difficulty coefficient weight factors of suitable ground materials stored in the database are set by professionals, and the cleaning difficulty coefficient of ground materials is obtained by analysis. For example, a mud floor will absorb moisture quickly, but it will also easily become muddy, so the cleaning difficulty coefficient is high; while some new anti-slip floor tiles have extremely low water absorption, and moisture will form water droplets on the surface, so the cleaning difficulty coefficient is low.

[0038] In a specific embodiment, the water spraying rate and water spraying amount of each activity area of ​​the cow are determined, and the specific analysis method is: extract the cleaning difficulty coefficient of each activity area of ​​the cow, and compare it with the cleaning difficulty coefficient range corresponding to each water spraying rate and the cleaning difficulty coefficient range corresponding to each water spraying amount stored in the database; if the cleaning difficulty coefficient of a certain activity area of ​​the cow is within the cleaning difficulty coefficient range corresponding to a certain water spraying rate, then the water spraying rate is used as the water spraying rate of the activity area of ​​the cow, and the water spraying rate of each activity area of ​​the cow is obtained; if the cleaning difficulty coefficient of a certain activity area of ​​the cow is within the cleaning difficulty coefficient range corresponding to a certain water spraying amount, then the water spraying amount is used as the water spraying amount of the activity area of ​​the cow, and the water spraying amount of each activity area of ​​the cow is obtained.

[0039] It should be noted that the range of cleaning difficulty coefficients corresponding to each water spray rate and the range of cleaning difficulty coefficients corresponding to each water spray volume stored in the database are set by professionals.

[0040] In the present invention, more attention is paid to the water spraying rate and amount in the active area. Appropriate water spraying rate and amount can ensure that the water flow has sufficient impact force on stains such as feces, sewage and feed residues in the active area, making them easier to be washed away, thereby improving cleaning efficiency, reducing stain residues, ensuring that cows are in a clean living environment, and reducing the risk of cows contracting diseases.

[0041] The shower area health analysis module obtains videos of each cow's shower area through high-definition cameras, collects data from each cow's shower area, obtains the location of each cow through GPS, evaluates the behavioral health coefficient of each cow in each shower area, evaluates the skin cleanliness index of each cow in each shower area, and then determines whether the shower system in each cow's shower area is started. If the shower system in each cow's shower area is not started, the length of time each cow stays in each shower area is analyzed.

[0042] It should be noted that all cows wear ear tags, and the location of each ear tag is obtained through GPS to obtain the location of each cow. The ear tag also contains the cow's number.

[0043] In a specific embodiment, the behavioral health coefficient of each cow in each shower area of ​​the cows is evaluated, and the specific analysis method is as follows: based on the video of each shower area of ​​the cows, extraction is performed according to a preset frequency to obtain each picture, and each body contour of each cow in each picture is extracted, and it is compared with each body contour of each cow's scratching behavior preset in the database; if a body contour of a cow in a certain picture is consistent with a body contour of a cow's scratching behavior preset in the database, then the cow is determined to be in a scratching state, and the number of pictures in which the cow has scratching behavior is obtained by statistics, and the scratching frequency and scratching duration of the cow are obtained by analysis, and then the scratching frequency of each cow in each shower area of ​​the cow is obtained, and the scratching duration of each cow in each shower area of ​​the cow is obtained.

[0044] It should be noted that the analysis obtains the frequency and duration of the cow's scratching, carefully checks the numbers of the pictures with scratching behavior, and determines the starting and ending pictures of each scratching behavior by comparing the body contours of the cow in adjacent pictures, so as to obtain the number of scratching behaviors in the analysis time period, i.e., the scratching frequency. The preset time interval for extracting pictures from the video is known, and the duration of the scratching behavior is calculated by subtracting the number of the starting picture from the number of the ending picture of each scratching behavior. The duration of all single scratching behaviors is added together to obtain the total scratching duration of the cow in the analysis time period, i.e., the scratching duration. Similarly, the tail-flashing frequency of each cow in each showering area of ​​the cow is obtained, which will not be elaborated here.

[0045] Based on the videos of each shower area for cows, the tail-flicking frequency of each cow in each shower area was analyzed, and the behavioral cleaning coefficient of each cow in each shower area was evaluated. The calculation formula is: ,in Indicates Shower area The behavioral cleanliness coefficient of each cow, , , Respectively represent Shower area The frequency, duration and tail-flicking frequency of each cow. , , They represent the appropriate cow scratching frequency, the appropriate cow scratching duration, and the appropriate cow tail flicking frequency stored in the database, respectively. Indicates the number of each shower area, , is a positive integer greater than 2, Indicates the number of each shower area, , is a positive integer greater than 2, , , They respectively represent the appropriate cow scratching frequency weight factor, the appropriate cow scratching duration weight factor, and the appropriate cow tail flicking frequency weight factor stored in the database.

[0046] It should be noted that the body contours of each cow scratching behavior preset in the database, the appropriate cow scratching frequency, the appropriate cow scratching duration, the appropriate cow tail-flicking frequency stored in the database, the appropriate cow scratching frequency weight factor, the appropriate cow scratching duration weight factor, and the appropriate cow tail-flicking frequency weight factor stored in the database are set by professionals.

[0047] In a specific embodiment, the skin cleanliness index of each cow in each shower area is evaluated by a specific analysis method: extracting the stain area and the number of stains on the udder of each cow in each shower area from the video, and evaluating the skin cleanliness index of each cow in each shower area. The calculation formula is: ,in Indicates Shower area Skin cleanliness index of a cow, , Respectively represent Shower area The stain area and number of stains on the udder of each cow, , They respectively represent the stain area of ​​a suitable cow's udder and the number of stains of a suitable cow's udder stored in the database.

[0048] It should be noted that the suitable stain area of ​​the cow's udder and the suitable number of stains on the cow's udder stored in the database are set by professionals.

[0049] In the present invention, more attention is paid to the cleaning of dairy cows, the udders and surrounding skin of dairy cows are cleaned in time, and it is difficult for bacteria to enter the milk during milking, thereby improving the quality of milk, extending the shelf life of milk, improving the quality and safety of milk, enhancing the brand reputation of the farm, and enhancing consumers' trust in their products. At the same time, a voting mechanism is adopted, and the system can more accurately determine when to start the spraying equipment. If the cow disappears temporarily in the shadow, the system will not shut down the equipment immediately, but continue to judge the behavior through the cumulative scoring mechanism, which increases the fault tolerance rate and ensures that the cow can be cleaned in time when it really needs a shower.

[0050] In a specific embodiment, the specific analysis method for determining whether the shower system of each shower area is started is as follows: extracting the behavior cleaning coefficient of each cow in each shower area and the skin cleaning index of each cow in each shower area, and comparing them with the behavior cleaning coefficient of a suitable cow and the skin cleaning index of a suitable cow stored in a database, respectively; if the behavior cleaning coefficient of a cow in a certain shower area is higher than the behavior cleaning coefficient of a suitable cow stored in the database, then it is determined that the cow needs to be cleaned; if the skin cleaning index of a cow in a certain shower area is higher than the skin cleaning index of a suitable cow stored in the database, then it is determined that the cow needs to be cleaned; the number of cows that need to be cleaned in each shower area is obtained by counting; the total number of cows in each shower area is counted and compared; the percentage of cows that need to be cleaned in each shower area is calculated; and it is compared with the percentage of cows that need to be cleaned stored in the database; if the percentage of cows that need to be cleaned in a certain shower area is higher than the percentage of cows that need to be cleaned stored in the database, then it is determined that the shower system of the shower area is started.

[0051] It should be noted that the appropriate cow behavior cleaning coefficient, the appropriate cow skin cleaning index stored in the database, and the percentage of cows that need cleaning stored in the database are set by professionals.

[0052] In a specific embodiment, the length of stay of each cow in each shower area is analyzed. The specific analysis method is: based on the video of each shower area and the position of each cow, the length of stay of each cow in each shower area is extracted and input into the shower system control formula. The formula expression is: ,like , then it means If the total length of time that a cow stays in each shower area reaches the shower system opening standard, , then it means The total time that the cows stay in each shower area reaches the shower system opening standard. Indicates Shower area The length of stay of a cow, Indicates the total suitable duration of stay stored in the database. Indicates the number of shower areas.

[0053] It should be noted that the appropriate stay time stored in the database is set by professionals, and if the shower system start-up standard is reached, the shower system is turned on.

[0054] The display terminal displays the video of each activity area of ​​the cows, displays the environmental cleanliness index of each activity area of ​​the cows, displays the water spraying rate and water spraying amount of each activity area of ​​the cows, and displays the video of each shower area of ​​the cows.

[0055] The information acquisition module is connected to the activity area health analysis module, the activity area health analysis module is connected to the shower area health analysis module, and the activity area health analysis module and the shower area health analysis module are simultaneously connected to a database and a display terminal.

[0056] It should be noted that the present invention also includes a database, which specifically includes a suitable temperature range, a suitable humidity range, a suitable harmful gas concentration range, a suitable feces area, a suitable sewage area, a suitable feed residue area, a suitable feces area weight factor, a suitable sewage area weight factor, a suitable feed residue area weight factor, a suitable environmental cleanliness index, a suitable area of ​​an activity area, a suitable cleaning difficulty coefficient of a ground material, a suitable area weight factor of an activity area, a suitable cleaning difficulty coefficient weight factor of a ground material, a cleaning difficulty coefficient range corresponding to each water spray rate, a cleaning difficulty coefficient range corresponding to each water spray amount, preset body contours of each cow's scratching behavior, a suitable cow's scratching frequency, a suitable cow's scratching duration, a suitable cow's tail-swiping frequency, a suitable cow's scratching frequency weight factor, a suitable cow's scratching duration weight factor, a suitable cow's tail-swiping frequency weight factor, a suitable cow's udder stain area, a suitable cow's udder stain number, a suitable cow's behavioral cleaning coefficient, a suitable cow's skin cleaning index, the percentage of cows that need to be cleaned, and a suitable stay duration.

[0057] refer to Figure 2 , a method for a farm environment control system based on computer vision, comprising: Step 1: Information acquisition: Use high-definition cameras to obtain videos of each cow's activity area, collect data from each cow's activity area, obtain environmental data from each cow's activity area, and determine whether each cow's activity area meets the needs of dairy farming. If the environmental data of a cow's activity area does not meet the needs of dairy farming, the relevant equipment will automatically adjust.

[0058] Step 2: Activity area health analysis: Based on the environmental data of each activity area of ​​the cow, the environmental cleanliness index of each activity area of ​​the cow is evaluated. Based on the environmental cleanliness index of each activity area of ​​the cow, it is determined whether each activity area of ​​the cow needs to be sprayed with water for cleaning. If a certain activity area of ​​the cow needs to be sprayed with water for cleaning, the cleaning difficulty coefficient of each activity area of ​​the cow is evaluated, and the spraying rate and amount of water in each activity area of ​​the cow are determined.

[0059] Step 3: Shower area health analysis: obtain videos of each cow's shower area through high-definition cameras, collect data from each cow's shower area, obtain the location of each cow through GPS, evaluate the behavioral health coefficient of each cow in each shower area, evaluate the skin cleanliness index of each cow in each shower area, and then determine whether the shower system in each cow's shower area is started. If the shower system in each cow's shower area is not started, analyze the length of time each cow stays in each shower area.

[0060] The above contents are merely examples and explanations of the concept of the present invention. Those skilled in the art may make various modifications or additions to the specific embodiments described or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they shall all fall within the protection scope of the present invention.

Claims

1. A farm environment control system based on computer vision, characterized in that: include: The information acquisition module uses high-definition cameras to obtain videos of various cow activity areas, collect data from various cow activity areas, obtain environmental data from various cow activity areas, and determine whether various cow activity areas meet the requirements of dairy farming. If the environmental data of a certain cow activity area does not meet the requirements of dairy farming, the relevant equipment will automatically adjust; The activity area health analysis module evaluates the environmental cleanliness index of each activity area of ​​the cow based on the environmental data of each activity area of ​​the cow. Based on the environmental cleanliness index of each activity area of ​​the cow, it is determined whether each activity area of ​​the cow needs to be sprayed with water for cleaning. If a certain activity area of ​​the cow needs to be sprayed with water for cleaning, the cleaning difficulty coefficient of each activity area of ​​the cow is evaluated, and the spraying rate and amount of water in each activity area of ​​the cow are determined; The shower area health analysis module uses high-definition cameras to obtain videos of each cow's shower area, collects data from each cow's shower area, obtains the location of each cow through GPS, evaluates the behavioral health coefficient of each cow in each shower area, evaluates the skin cleanliness index of each cow in each shower area, and then determines whether the shower system in each cow's shower area is activated. If the shower system in each cow's shower area is not activated, the length of stay of each cow in each shower area is analyzed; The display terminal displays the video of each activity area of ​​the cows, displays the environmental cleanliness index of each activity area of ​​the cows, displays the water spraying rate and water spraying amount of each activity area of ​​the cows, and displays the video of each shower area of ​​the cows.

2. A farm environment control system based on computer vision according to claim 1, characterized in that: The specific analysis method for evaluating the environmental cleanliness index of each dairy cow activity area is as follows: From the videos of the cows’ various activity areas, we extracted the feces area, sewage area, and feed residue area, and then evaluated the environmental cleanliness index of the cows’ various activity areas. The calculation formula is: ,in Indicates the cow Environmental cleanliness index of each activity area, , , Respectively represent the cow The area of ​​feces, sewage, and feed residue in each activity area. , , They represent the suitable feces area, sewage area and feed residue area stored in the database respectively. , , They represent the appropriate feces area weight factor, the appropriate sewage area weight factor, and the appropriate feed residue area weight factor stored in the database, respectively; e represents a natural constant, Indicates the number of the active area, , is a positive integer greater than 2.

3. A farm environment control system based on computer vision according to claim 2, characterized in that: The specific analysis method for judging whether each activity area of ​​the dairy cow needs to be sprayed with water for cleaning is as follows: The environmental cleanliness index of each activity area of ​​the cows is extracted and compared with the appropriate environmental cleanliness index stored in the database. If the environmental cleanliness index of a certain activity area of ​​the cows is higher than the appropriate environmental cleanliness index stored in the database, it is determined that the activity area needs to be sprayed with water for cleaning; if the environmental cleanliness index of a certain activity area of ​​the cows is lower than the appropriate environmental cleanliness index stored in the database, it is determined that the activity area does not need to be sprayed with water for cleaning.

4. A farm environment control system based on computer vision according to claim 3, characterized in that: The specific analysis method for evaluating the cleaning difficulty coefficient of each activity area of ​​the dairy cow is as follows: The area and ground material of each cow activity area are extracted from the database, and the cleaning difficulty coefficient of the ground material is analyzed. Combined with the environmental cleanliness index of each cow activity area, the cleaning difficulty coefficient of each cow activity area is evaluated. The calculation formula is: ,in Indicates the cow The cleaning difficulty coefficient of each activity area, , Respectively represent the cow The area of ​​each activity area and the cleaning difficulty of the floor material, , They represent the area of ​​the suitable activity area and the cleaning difficulty coefficient of the suitable floor material stored in the database. , They respectively represent the area weight factor of the suitable activity area and the cleaning difficulty coefficient weight factor of the suitable floor material stored in the database.

5. The computer vision-based farm environment control system according to claim 4, characterized in that: The specific analysis method for determining the water spraying rate and water spraying amount in each activity area of ​​the dairy cow is as follows: The cleaning difficulty coefficient of each activity area of ​​the cow is extracted, and compared with the cleaning difficulty coefficient ranges corresponding to each water spray rate and each water spray amount stored in the database. If the cleaning difficulty coefficient of a certain activity area of ​​the cow is within the cleaning difficulty coefficient range corresponding to a certain water spray rate, the water spray rate is used as the water spray rate of the activity area of ​​the cow to obtain the water spray rate of each activity area of ​​the cow. If the cleaning difficulty coefficient of a certain activity area of ​​the cow is within the cleaning difficulty coefficient range corresponding to a certain water spray amount, the water spray amount is used as the water spray amount of the activity area of ​​the cow to obtain the water spray amount of each activity area of ​​the cow.

6. The computer vision-based farm environment control system according to claim 5, characterized in that: The specific analysis method for evaluating the behavioral health coefficient of each cow in each shower area is as follows: Based on the videos of each shower area of ​​the cow, extract according to the preset frequency to obtain each picture, extract each body contour of each cow in each picture, and compare it with each body contour of each cow's scratching behavior preset in the database. If a body contour of a cow in a certain picture is consistent with a body contour of a cow's scratching behavior preset in the database, it is determined that the cow is in a scratching state, and the number of pictures in which the cow has scratching behavior is obtained by counting, and the scratching frequency and scratching duration of the cow are obtained by analysis, and then the scratching frequency of each cow in each shower area of ​​the cow is obtained, and the scratching duration of each cow in each shower area of ​​the cow is obtained; Based on the videos of each shower area for cows, the tail-flicking frequency of each cow in each shower area was analyzed, and the behavioral cleaning coefficient of each cow in each shower area was evaluated. The calculation formula is: ,in Indicates Shower area The behavioral cleanliness coefficient of each cow, , , Respectively represent Shower area The frequency, duration and tail-flicking frequency of each cow. , , They represent the appropriate cow scratching frequency, the appropriate cow scratching duration, and the appropriate cow tail flicking frequency stored in the database, respectively. Indicates the number of each shower area, , is a positive integer greater than 2, Indicates the number of each shower area, , is a positive integer greater than 2, , , They respectively represent the appropriate cow scratching frequency weight factor, the appropriate cow scratching duration weight factor, and the appropriate cow tail flicking frequency weight factor stored in the database.

7. The computer vision-based farm environment control system according to claim 6, characterized in that: The specific analysis method for evaluating the skin cleanliness index of each cow in each shower area is as follows: The stain area and number of stains on the udder of each cow in each shower area were extracted from the video, and the skin cleanliness index of each cow in each shower area was evaluated. The calculation formula is: ,in Indicates Shower area Skin cleanliness index of a cow, , Respectively represent Shower area The stain area and number of stains on the udder of each cow, , They respectively represent the stain area of ​​a suitable cow's udder and the number of stains of a suitable cow's udder stored in the database.

8. The computer vision-based farm environment control system according to claim 7, characterized in that: The specific analysis method for determining whether the shower system of each shower area is started is as follows: The behavior cleaning coefficient of each cow in each shower area and the skin cleaning index of each cow in each shower area are extracted, and they are compared with the behavior cleaning coefficient of a suitable cow and the skin cleaning index of a suitable cow stored in the database. If the behavior cleaning coefficient of a cow in a shower area is higher than the behavior cleaning coefficient of a suitable cow stored in the database, it is determined that the cow needs to be cleaned. If the skin cleaning index of a cow in a shower area is higher than the skin cleaning index of a suitable cow stored in the database, it is determined that the cow needs to be cleaned. The number of cows that need to be cleaned in each shower area is obtained by counting, the total number of cows in each shower area is counted and compared, and the percentage of cows that need to be cleaned in each shower area is calculated and compared with the percentage of cows that need to be cleaned stored in the database. If the percentage of cows that need to be cleaned in a shower area is higher than the percentage of cows that need to be cleaned stored in the database, it is determined that the shower system in the shower area is started.

9. The computer vision-based farm environment control system according to claim 8, characterized in that: The specific analysis method for analyzing the duration of stay of each cow in each shower area is as follows: Based on the videos of each cow shower area and the position of each cow, the stay time of each cow in each shower area is extracted and input into the shower system control formula. The formula expression is: ,like , then it means If the total time that a cow stays in each shower area reaches the shower system opening standard, , then it means The total time that the cows stay in each shower area reaches the shower system opening standard. Indicates Shower area The length of stay of a cow, Indicates the total suitable duration of stay stored in the database. Indicates the number of shower areas.

10. A method for executing a farm environment control system based on computer vision according to any one of claims 1 to 9, characterized in that: include: Step 1: Information acquisition: obtain videos of cows’ various activity areas through high-definition cameras, collect data of cows’ various activity areas, obtain environmental data of cows’ various activity areas, and judge whether cows’ various activity areas meet the requirements of dairy farming. If the environmental data of a cow’s activity area does not meet the requirements of dairy farming, the relevant equipment will automatically adjust; Step 2: Activity area health analysis: Based on the environmental data of each activity area of ​​the cow, the environmental cleanliness index of each activity area of ​​the cow is evaluated. Based on the environmental cleanliness index of each activity area of ​​the cow, it is determined whether each activity area of ​​the cow needs to be sprayed with water for cleaning. If a certain activity area of ​​the cow needs to be sprayed with water for cleaning, the cleaning difficulty coefficient of each activity area of ​​the cow is evaluated, and the spraying rate and amount of each activity area of ​​the cow are determined; Step 3: Shower area health analysis: obtain videos of each cow's shower area through high-definition cameras, collect data from each cow's shower area, obtain the location of each cow through GPS, evaluate the behavioral health coefficient of each cow in each shower area, evaluate the skin cleanliness index of each cow in each shower area, and then determine whether the shower system in each cow's shower area is started. If the shower system in each cow's shower area is not started, analyze the length of time each cow stays in each shower area.

Images