Cup patrolling operation auxiliary lighting system of automatic milking rotary table and application method of cup patrolling operation auxiliary lighting system

Abstract

A cup patrolling operation auxiliary lighting system of an automatic milking rotary table relates to the technical field of livestock breeding and comprises a PC (personal computer), a camera, an AI (artificial intelligence) visual recognition module, a data acquisition module, a data processing module, an indicator light module and a cloud data management module. The PC is installed in the control box, the AI visual recognition module, the data acquisition module, the data processing module and the indicator light module are application software installed in the PC, and the multiple cameras are installed at the side ends of medicated bath robots of multiple milking stations and medicated bath stations of the milking rotary table respectively; the application method of the cup patrolling operation auxiliary lighting system of the automatic milking rotary table comprises three steps. The system is used in cooperation with an automatic milking rotary table, under the combined action of related modules, by combining AI visual identification, data acquisition and analysis and other schemes, intelligent monitoring and abnormity processing of the milk cup separation state can be achieved, the operation efficiency and reliability of the automatic milking rotary table are improved, and cow health and milk production quality are guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of livestock breeding automation equipment and application methods, and particularly to a cup inspection operation auxiliary lighting system for an automatic milking turntable and an application method thereof. Background Art

[0002] In modern dairy farming, automatic milking turntables have been widely used. Automatic milking turntables can achieve functions such as automatic positioning of dairy cows, milking, and automatic detachment of milking cups, greatly improving milking efficiency and production management levels. However, in the actual production process, affected by various factors, abnormal detachment of milking cups during milking and failure to detach after milking occur frequently. Premature detachment of milking cups will lead to insufficient milking and affect milk production; if the milking cups do not detach normally after milking, it may cause damage to the udders of dairy cows and trigger diseases.

[0003] Currently, for the monitoring of the detachment status of milking cups on automatic milking turntables, most rely on manual inspections. Manual inspections have problems such as low efficiency, being easily affected by subjective factors, being unable to monitor in real time, and increasing production costs. Although there are also simple sensors used in the prior art to monitor the status of milking cups, due to technical limitations, the sensor detection method cannot accurately determine whether the detachment of milking cups is normal and is difficult to meet the requirements of modern and intelligent breeding. Therefore, it is very necessary to provide a system and application method that can accurately judge the detachment status of milking cups and can promptly remind the staff to handle abnormalities. Summary of the Invention

[0004] In order to overcome the drawbacks described in the background art due to structural limitations of existing sensor devices for monitoring milking cups, the present invention provides an automatic milking turntable cup inspection operation auxiliary lighting system and an application method thereof that are used in conjunction with an automatic milking turntable. Under the combined action of relevant modules, through the combination of AI visual recognition, data collection and analysis, etc., intelligent monitoring and abnormal handling of the detachment status of milking cups can be achieved, improving the operating efficiency and reliability of the automatic milking turntable, and ensuring the health of dairy cows and the quality of milk production.

[0005] The technical solution adopted by the present invention to solve its technical problems is: Automatic milking turntable cup inspection operation auxiliary lighting system, including a PC, a camera, and also has an AI vision recognition module, a data acquisition module, a data processing module, an indicator light module, and a cloud data management module; the PC is installed in the control box, and the AI vision recognition module, the data acquisition module, the data processing module, and the indicator light module are application software installed in the PC. Multiple cameras are respectively installed on the side of the milking robots at multiple milking stations and the medicating station of the automatic milking turntable; the AI vision recognition module is based on the camera to perform real-time image acquisition of the milk scale, and through the AI vision recognition algorithm, analyze the acquired image to identify whether the milk scale has a cup-off situation; the data acquisition module can collect the cow breast structure information and related data obtained by the medicating robot based on the camera. At the same time, the data acquisition module can read the operating status information of the turntable from the milking turntable data panel, and the collected data will provide multi-dimensional data support for subsequent analysis and judgment; the data processing module: can receive the cup-off recognition result transmitted by the AI vision recognition module, the cow breast structure information collected by the data acquisition module, and the turntable operating status information. Then, the data processing module automatically matches the cow breast structure and combines historical data to compare the fullness degree of the cow breast, automatically discriminates whether the cup-off of the milk cup is normal. At the same time, the data processing module analyzes the abnormal error information on the turntable data panel to further confirm whether the cup-off status of the milk cup is abnormal; there are multiple sets of the indicator light module, and multiple sets of indicator light modules are respectively installed at each station of the automatic milking turntable. The indicator light module is connected to the data processing module through a data cable. When the data processing module determines that there is an abnormal cup-off situation at a certain station, it sends a control signal to the indicator light module, and the indicator light module lights up the indicator light of the corresponding station to remind the cup inspector to process this station; the cloud data management module: is connected to the data processing module wirelessly and receives the cup-off data summarized and sorted out by the data processing module.

[0006] Further, the AI vision recognition module can accurately capture the connection state details between the milk cup and the milk scale, that is, whether the milk cup is on the cow's nipple, with high recognition accuracy and flexibility.

[0007] Further, in the data acquisition module, the specific operating status information is whether there is an abnormal error situation on the data panel.

[0008] Further, in the data processing module, an association model of the cow breast fullness degree, normal milking time, and cup-off status is established based on historical data. When the real-time collected data deviates from the model, it is judged as abnormal cup-off.

[0009] Further, in the indicator light module, the indicator lights can use different colors or blinking frequencies to distinguish different types of abnormal situations, which is convenient for the cup inspector to quickly identify and process.

[0010] Furthermore, in the cloud data management module, it is also possible to store, analyze, and manage the cup-off data. The breeding personnel can access the cloud data management module through the network to view the historical cup-off data and analyze the patterns of abnormal cup-off, providing a data basis for optimizing the milking process and improving the operation efficiency of the equipment.

[0011] The application method of the auxiliary lighting system for the cup inspection operation of the automatic milking turntable includes the following steps: S1: Data collection. The AI vision recognition module captures images of the milk scale, and the data collection module collects the cow's breast structure information of the medicating robot and the operation status information of the turntable data panel; S2: Cup-off status recognition and judgment. The AI vision recognition module analyzes the captured images to identify whether the milk scale is cup-off; the data processing module receives the recognition results of the AI vision recognition module and the data collected by the data collection module, automatically matches the breast structure, combines historical data to compare the fullness of the cow's breast, and at the same time analyzes the abnormal error information of the turntable data panel to comprehensively judge whether the cup-off of the milk cup is normal; S3: Data management. The data processing module summarizes and organizes the cup-off data and uploads it to the cloud data management module, and the cloud data management module stores, analyzes, and manages the data.

[0012] Furthermore, in step S2, if the data processing module determines that it is an abnormal cup-off, it sends a control signal to the indicator light module, and the indicator light module lights up the indicator light of the corresponding work station to remind the cup inspection personnel to handle it.

[0013] Furthermore, in step S3, the cloud data management module can also optimize the milking process and equipment operation parameters, promoting the intelligent and scientific development of breeding production.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: Under the combined action of the camera and related software, the present invention can more accurately judge the cup-off state of the milk cup, avoid milking problems caused by misjudgment or missed judgment, and ensure the health of cows and the quality of milk production; it can timely and prominently remind the cup inspection personnel to handle abnormal cup-off situations, reduce the manual inspection time, improve the efficiency of abnormal handling, and reduce the labor cost; it can also realize the centralized storage and analysis of cup-off data, provide data support for breeding personnel, facilitate the optimization of the milking process and equipment operation parameters, and promote the intelligent and scientific development of breeding production. The present invention can be seamlessly docked with existing equipment such as automatic milking turntables and medicating robots, without large-scale transformation of existing equipment, and has good compatibility and scalability, so it has good application prospects. Brief Description of the Drawings

[0015] Figure 1 It is a block diagram of the architecture of the auxiliary lighting system for the cup inspection operation of the automatic milking turntable of the present invention. Detailed Embodiments

[0016] Figure 1 As shown, the auxiliary lighting system for cup inspection operation of the automatic milking turntable includes a PC and a camera, and also has an AI vision recognition module, a data acquisition module, a data processing module, an indicator light module, and a cloud data management module; the PC is installed in the control box of the milking turntable, and the AI vision recognition module, the data acquisition module, the data processing module, and the indicator light module are application software installed in the PC. Multiple cameras are respectively installed on the side of the milking robots at multiple milking stations and the medicating station of the milking turntable. The AI vision recognition module performs real-time image acquisition on the milk weighing cup (also known as the milking cup liner, which is a key component in the milking equipment that directly contacts the cow's nipple) based on the camera, and analyzes the collected images through the AI vision recognition algorithm to identify whether the milk weighing cup has a cup-off situation. Since the AI vision recognition module can accurately capture the connection state details between the milk cup and the milk weighing cup, it has higher recognition accuracy and flexibility compared with the traditional sensor monitoring method.

[0017] Figure 1As shown, the AI visual recognition algorithm uses AI vision to determine the cup removal situation on the automatic milking carousel, mainly involving several key steps such as image acquisition, feature extraction, model training, and recognition. The following is the specific theoretical elaboration. A: Image acquisition. Install a high-definition camera at a suitable position above or around the milking carousel to ensure that the contact area between the milking cup and the cow's teat can be clearly photographed. To ensure image quality, it is necessary to reasonably set the lighting conditions to avoid interference caused by shadows or reflections on the image. A uniform scattering light source can be used to make the light in the photographed area sufficient and evenly distributed. B: Image preprocessing. Perform grayscale processing on the acquired image to convert the color image into a grayscale image, which can reduce the amount of data and at the same time highlight the brightness information of the image, facilitating subsequent feature extraction. Perform filtering operations, such as using Gaussian filtering to remove noise in the image, making the image smoother, improving the signal-to-noise ratio of the image, and thus better identifying the target object in the image. Image enhancement processing can also be carried out, such as improving the contrast of the image through methods such as histogram equalization, making the details in the image clearer, and facilitating subsequent feature extraction of the milking cup and the teat.C: Feature extraction. (1) Geometric features: Extract geometric information such as the shape, size, and position of the milking cup and the cow's teat. For example, obtain the contours of the milking cup and the teat through edge detection algorithms, and calculate parameters such as area, perimeter, major axis, and minor axis. In the normal milking state, the milking cup fits closely with the teat, presenting a specific geometric shape and positional relationship. When the cup is removed, this geometric relationship changes significantly, such as a gap appearing between the milking cup and the teat, or the position of the milking cup shifting. (2) Texture features: Analyze the texture information on the surfaces of the milking cup and the teat. During normal milking, the texture of the part where the milk liner contacts the teat is relatively uniform and stable. When cup removal occurs, due to the deformation of the milk liner or the movement of the teat, the texture will show changes such as distortion, breakage, or discontinuity. Methods such as gray-level co-occurrence matrix can be used to extract texture features to describe the spatial relationship and gray-level change law between pixels in the image. (3) Motion features: If it is real-time monitoring, analyze the motion states of the milking cup and the teat over a period of time. For example, calculate the motion speed and direction of objects in the image through optical flow method. During normal milking, the milking cup and the teat are relatively stationary or have only minor periodic motions. When cup removal occurs, there will be obvious relative motions, such as the milking cup suddenly moving upward or outward, and the teat coming out of the milking cup. (4) Model training: Collect a large amount of image data containing normal milking states and cup removal states, and perform manual annotation to mark whether there is a cup removal situation in the image and the position of cup removal, etc. Select a suitable deep learning model, such as a convolutional neural network (CNN). CNN has strong feature extraction capabilities and can automatically learn the feature patterns in the image. Divide the annotated image data into a training set, a validation set, and a test set. Use the training set to train the model. By adjusting the parameters of the model, make the model accurately identify the cup removal features in the image. During the training process, use the validation set to monitor the training effect of the model to prevent overfitting. After training, use the test set to evaluate the model and calculate metrics such as accuracy and recall rate of the model to measure the performance of the model. D: Cup removal determination. After preprocessing and feature extraction of the real-time collected image, input it into the trained AI model for recognition. The model judges whether there is a cup removal situation in the image according to the learned feature patterns. If the result output by the model indicates cup removal, the position of cup removal can be further located. For example, mark the area where the milking cup and the teat with cup removal are located through a bounding box. To improve the accuracy of determination, certain thresholds and confidence levels can be set. When the cup removal probability output by the model exceeds the set threshold, it is determined as cup removal. At the same time, combine the image information of multiple consecutive frames for comprehensive judgment to avoid false cup removal alarms caused by noise or misjudgment in a single image. Through the above AI vision-based method, accurate determination of cup removal situations on the automatic milking carousel can be achieved, timely detection of abnormal situations during milking, improvement of milking efficiency and milk quality, and reduction of the workload and error of manual monitoring.

[0018] Figure 1 As shown, the data acquisition module can collect the udder structure information and related data of the medicated bath robot based on the camera. At the same time, the data acquisition module can also read the operation status information of the turntable from the turntable data panel, including whether there are abnormal errors, etc. The data collected by this module will provide multi-dimensional data support for subsequent analysis and judgment. The data processing module can receive the milk scale cup removal recognition results transmitted by the AI ​​visual recognition module, the udder structure information of the cow collected by the data acquisition module, the turntable operation status information and other data. The data processing module automatically matches the udder structure of the cow and can automatically judge whether the milk cup is normal by comparing the fullness of the cow's udder with historical data. Specifically, a correlation model between the fullness of the cow's udder and the normal milking time and the cup removal state is established based on historical data. When the real-time collected data deviates from the model, it is judged as abnormal cup removal. At the same time, the data processing module analyzes the abnormal error information of the turntable data panel to further confirm whether the milk cup removal state is abnormal, and the data obtained is more real and effective. The indicator light module is set near each station of the automatic milking turntable, and realizes information exchange with the data processing module. When the data processing module determines that an abnormal cup removal occurs at a certain station, it sends a control signal to the indicator light module, and the indicator light module lights up the indicator light of the corresponding station, reminding the cup patrol personnel to handle the station in a striking way. Specifically, the indicator light can use different colors or flashing frequencies to distinguish different types of abnormal situations, which is convenient for the cup patrol personnel to quickly identify and handle. The cloud data management module realizes information exchange with the data processing module, receives the cup removal data summarized and sorted by the data processing module, and stores, analyzes and manages the cup removal data. Farming personnel can access the cloud data management module through the network, view historical cup removal data, analyze the laws of abnormal cup removal, and provide data basis for optimizing milking technology and improving equipment operation efficiency.

[0019] Figure 1As shown, the correlation model is to establish a correlation model between the degree of udder filling of dairy cows and the normal milking time and milking cup-off state. It can be carried out in the steps of data collection and preprocessing, feature engineering, model selection and training, and model evaluation and optimization. The following is a specific description: A: Data collection and preprocessing, (1) Data collection: Collect dairy cow milking-related data within a certain time range from the recording system of milking equipment, sensor data, and manual observation records, including the start time, end time, milk yield, quantitative indicators of udder filling degree (such as pressure values, volume measurement values obtained through udder sensors, etc.) and milking cup-off state (normal cup-off or abnormal cup-off, which can be recorded by the aforementioned AI vision system) and other information; (2) Data cleaning: Check whether there are missing values, outliers, and incorrect data in the data. For missing values, appropriate filling methods can be selected according to specific situations, such as mean filling, median filling, or filling based on similar dairy cow individuals. For outliers, the reasons for their generation need to be analyzed to determine whether it is a real abnormal situation or a data recording error. If it is incorrect data, it is corrected or deleted. B: Data standardization: Standardize the collected different types of data and convert them into data with the same scale and distribution for the convenience of model training and comparison. For example, normalize numerical data such as milking time and milk yield, and also convert the indicator data of udder filling degree to the same numerical range. C: Feature engineering, (1) Feature selection: Select features that are closely related to the degree of udder filling of dairy cows, normal milking time, and milking cup-off state from the collected original features. For example, milk yield, multiple measurement indicators of udder filling degree, and the udder state at the start of milking may be relatively important features, while some features unrelated to the milking process, such as the dairy cow number and milking worker information, can be excluded; (2) Feature derivation: Create some new features based on the existing features to better describe the relationship between data. For example, milking speed (milk yield divided by milking time), change rate of udder filling degree (the difference in udder filling degree between this milking and the previous milking divided by the time interval) and other features can be calculated. These derived features may be very helpful for the prediction ability of the model. D: Model selection and training, (1) Select a model: Select an appropriate machine learning or statistical model according to the characteristics of the problem and the nature of the data. Commonly used models include linear regression models, decision tree models, random forest models, support vector machine models, etc.If the data has time series characteristics, models such as recurrent neural network (RNN) or long short-term memory network (LSTM) can also be considered. For example, if it is considered that there is an approximate linear relationship between the degree of udder fullness of cows, normal milking time, and cup-off state, a linear regression model can be tried first; if there are complex non-linear relationships in the data, models such as random forest or neural network may be more appropriate; (2) Divide the data set: Divide the preprocessed data into a training set, a validation set, and a test set. Generally, the training set is used for model training, the validation set is used to adjust the hyperparameters of the model and prevent overfitting, and the test set is used to evaluate the final performance of the model. Usually, it is divided according to the ratio of 7:2:1 or 8:1:1; (3) Model training: Use the training set to train the selected model. By adjusting the parameters of the model, the model can best fit the relationship between the features and the target variable in the training data. During the training process, according to the type and characteristics of the model, select a suitable optimization algorithm, such as stochastic gradient descent, Adagrad, Adadelta, etc., to minimize the loss function. For example, for a linear regression model, the mean squared error (MSE) can be selected as the loss function; for classification problems (such as determining whether the cup-off state is normal or abnormal), the cross-entropy loss function can be used; (4) Model evaluation and optimization; Evaluation metrics: Select appropriate evaluation metrics to measure the performance of the model. For regression problems (such as predicting normal milking time or udder fullness), common metrics include mean squared error (MSE), mean absolute error (MAE), coefficient of determination (R²), etc.; for classification problems (such as determining the cup-off state), common metrics include accuracy, recall, and F1 score; (5) Model evaluation: Use the test set to evaluate the trained model and calculate the values of various evaluation metrics. If the performance of the model does not meet the requirements, analyze in which aspects the model is lacking. For example, is the prediction inaccurate for certain specific cow individuals or specific milking conditions, or is the overall prediction effect poor; (6) Model optimization: Optimize the model according to the evaluation results. If it is an overfitting problem, you can try to increase the amount of data, reduce the complexity of the model (such as reducing the number of layers or neurons in the neural network), use regularization methods (such as L1 and L2 regularization), etc.; if it is an underfitting problem, increase the complexity of the model, select more appropriate features or adjust the hyperparameters of the model. After optimization, train and evaluate again until the performance of the model reaches a satisfactory effect. Through the above steps, a model can be established that can better describe the relationship between the degree of udder fullness of cows, normal milking time, and cup-off state, providing valuable reference for the monitoring and management of the milking process.

[0020] Figure 1As shown below, the calculation and analysis for judging the abnormal cup-off process are as follows. A: Set the deviation threshold: After establishing the correlation model among the degree of udder fullness of cows, the normal milking time, and the cup-off state, it is necessary to determine a deviation threshold, which represents the acceptable difference range between the real-time collected data and the model prediction value. For example, based on the analysis of historical data and practical experience, it is determined that when the deviation of the real-time collected milking time from the milking time predicted by the model exceeds 10%, or the deviation of the change rate of the udder fullness from the model prediction value exceeds 20%, a large deviation is considered to have occurred. B: Real-time data collection and calculation: During the milking process, relevant data of cows are collected in real time, including information such as the current degree of udder fullness and milking time. The collected data are input into the established correlation model to obtain the normal milking time predicted by the model and the change of the udder fullness degree, and then the deviation between the real-time data and the model prediction value is calculated. For example, assume that the model predicts the normal milking time of a certain cow to be 30 minutes, but the cup-off occurs when the actual milking reaches 20 minutes. Then the deviation of the milking time is (30 - 20) / 30×100%≈33.3%, exceeding the set threshold of 10%. At this time, it can be preliminarily judged as abnormal cup-off. C: Comprehensive judgment considering multiple factors: In addition to considering the deviation of milking time and udder fullness degree, other relevant factors can be combined for comprehensive judgment. For example, observe whether the milk yield is consistent with the milk yield predicted by the model. If the milk yield is significantly lower than the normal level and at the same time there are deviations in milking time and udder fullness degree, then it is more likely to be abnormal cup-off. In addition, factors such as individual differences of cows and the diet situation on that day can be considered to further correct the judgment result. The abnormal error messages analyzed from the turntable data panel are as follows. A: Parse the error messages. The abnormal error messages on the turntable data panel usually contain some key codes or prompt messages, which are used to indicate the type or location of the problem. First, these error messages need to be parsed to convert them into understandable text descriptions. For example, the error message may be "ErrorCode101:Milkflowsensormalfunction". By referring to relevant technical documents or communicating with the equipment supplier, it is learned that "ErrorCode101" indicates a failure of the milk flow sensor. B: Correlate relevant data. Correlate the abnormal error messages with the real-time collected cow data and the calculation results of the model. For example, if an error message indicating a failure of the milk flow sensor appears on the data panel at the same time as abnormal cup-off, then it is necessary to further analyze the change of the milk flow data during the milking process to check whether the milk flow shows abnormal fluctuations or a sudden drop before cup-off, which may be related to the sensor failure or one of the reasons for abnormal cup-off.C: Fault location and cause analysis. Based on the analysis of error messages and relevant data, attempt to locate the specific location and cause of the fault. If abnormal cup removal is caused by a sensor fault, it is necessary to check whether the connection of the sensor is normal, whether it is interfered with or damaged, etc. If it is due to other reasons, such as mechanical faults of the milking equipment, health problems of the cows, etc., more information needs to be combined for judgment. For example, check the sealing of the milking cup, the working status of the vacuum pump, etc., and at the same time check whether the cows have symptoms of diseases such as mastitis. D: Recording and feedback: Record the results obtained from the analysis, including the detailed content of the abnormal error message, associated data, fault location and cause analysis, etc., and feedback this information to relevant technical personnel or management personnel for timely measures to be taken for repair and handling. At the same time, track and record the processing results, observe whether similar problems will occur in subsequent milking processes, and continuously optimize the data processing and fault analysis methods.

[0021] Figure 1 As shown, the application method of the cup inspection operation auxiliary lighting system for the automatic milking turntable includes the following steps. (1) Data collection: The AI vision recognition module collects images of the milk scale, and the data collection module collects the cow's breast structure information of the medicating robot and the operating status information of the turntable data panel. (2) Cup removal status recognition and judgment: The AI vision recognition module analyzes the collected images to identify whether the milk scale has removed the cup; the data processing module receives the recognition results of the AI vision recognition module and the data collected by the data collection module, automatically matches the breast structure, combines historical data to compare the fullness of the cow's breast, and at the same time analyzes the abnormal error message of the turntable data panel to comprehensively judge whether the cup removal is normal. (3) Abnormal reminder: If the data processing module determines that it is an abnormal cup removal, it sends a control signal to the indicator light module, and the indicator light module lights up the indicator light of the corresponding work station to remind the cup inspection personnel to handle it. (4) Data management: The data processing module summarizes and organizes the cup removal data and uploads it to the cloud data management module, and the cloud data management module stores, analyzes and manages the data.

[0022] Figure 1As shown, through the above technical solution, under the combined action of the camera and related software, the present invention can more accurately judge the state of milk cup detachment through AI vision recognition technology and multi-dimensional data collection and analysis, avoid milking problems caused by misjudgment or missed judgment, and ensure the health of dairy cows and the quality of milk production. Compared with the traditional sensor monitoring method, it has higher recognition accuracy and flexibility. The indicator light module can timely and prominently remind the cup inspection personnel to handle abnormal cup detachment situations, reduce the manual inspection time, improve the efficiency of abnormal situation handling, and reduce labor costs. The cloud data management module realizes the centralized storage and analysis of cup detachment data, provides data support for breeding personnel, facilitates the optimization of milking processes and equipment operation parameters, and promotes the intelligent and scientific development of breeding production. The present invention can be seamlessly docked with existing equipment such as automatic milking turntables and medicating robots, without the need for large-scale transformation of existing equipment, and has good compatibility and scalability, so it has good application prospects.

[0023] The foregoing has shown and described the basic principles, main features and advantages of the present invention. For those skilled in the art, it is obvious that the present invention is limited to the details of the above-described exemplary embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, in any regard, the embodiments should be regarded as exemplary and non-limiting, and the scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to embrace all changes falling within the meaning and scope of the equivalent elements of the claims in the present invention.

[0024] In addition, it should be understood that although this specification is described according to embodiments, not every embodiment only contains an independent technical solution. This narrative way of the specification is only for clarity. Those skilled in the art should regard the specification as a whole, and the technical solutions in the embodiments can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. Automatic milking turntable cup inspection operation auxiliary lighting system, including a PC and a camera, characterized in that, It also has an AI vision recognition module, a data acquisition module, a data processing module, an indicator light module, and a cloud data management module; the PC is installed in the control box, and the AI vision recognition module, the data acquisition module, the data processing module, and the indicator light module are application software installed in the PC. Multiple cameras are respectively installed on the side ends of the milking robots at multiple milking stations and the medicating station of the milking turntable; the AI vision recognition module is based on the cameras to perform real-time image acquisition on the milk scale, and through the AI vision recognition algorithm, analyze the acquired images to identify whether the milk scale has a cup-off situation. The data acquisition module can collect the cow breast structure information and related data obtained by the medicating robot based on the cameras. At the same time, the data acquisition module can read the operating status information of the turntable from the data panel of the milking turntable. The collected data will provide multi-dimensional data support for subsequent analysis and judgment; the data processing module: can receive the cup-off recognition result transmitted by the AI vision recognition module, the cow breast structure information collected by the data acquisition module, and the turntable operating status information. Then, the data processing module automatically matches the cow breast structure and combines historical data to compare the fullness degree of the cow breast, and automatically determines whether the cup-off of the milk cup is normal. At the same time, the data processing module analyzes the abnormal error message on the data panel of the turntable to further confirm whether the cup-off status of the milk cup is abnormal; there are multiple sets of the indicator light module, and multiple sets of indicator light modules are respectively installed at each station of the automatic milking turntable. The indicator light module is connected to the data processing module through a data cable. When the data processing module determines that there is an abnormal cup-off situation at a certain station, it sends a control signal to the indicator light module, and the indicator light module lights up the indicator light of the corresponding station to remind the cup inspection personnel to process this station; the cloud data management module: is connected to the data processing module wirelessly and receives the cup-off data summarized and sorted out by the data processing module.

2. The cup inspection operation auxiliary lighting system of the automatic milking turntable according to claim 1, wherein The AI vision recognition module can accurately capture the connection state details between the milk cup and the milk scale, that is, whether the milk cup is on the cow's teat, and has high recognition accuracy and flexibility.

3. The cup inspection operation auxiliary lighting system of the automatic milking turntable according to claim 1, wherein In the data acquisition module, the specific operating status information is whether there is an abnormal error message on the data panel.

4. The cup inspection operation auxiliary lighting system of the automatic milking turntable according to claim 1, characterized in that In the data processing module, an association model of the cow breast fullness degree, normal milking time, and cup-off status is established based on historical data. When the real-time collected data deviates from the model, it is judged as an abnormal cup-off.

5. The cup inspection operation auxiliary lighting system of the automatic milking turntable according to claim 1, characterized in that, In the indicator light module, the indicator lights can use different colors or blinking frequencies to distinguish different types of abnormal situations, which is convenient for the cup inspection personnel to quickly identify and process.

6. The cup inspection operation auxiliary lighting system of the automatic milking turntable according to claim 1, characterized in that In the cloud data management module, it can also store, analyze, and manage the cup-off data. The breeding personnel can access the cloud data management module through the network to view the historical cup-off data and analyze the rules of abnormal cup-off, providing a data basis for optimizing the milking process and improving the equipment operation efficiency.

7. The application method of the cup inspection operation auxiliary lighting system for the automatic milking turntable according to any one of claims 1 to 6, characterized in that, Application method of auxiliary lighting system for cup inspection operation of automatic milking turntable, including the following steps: S1: Data collection, the AI vision recognition module collects images of the milk scale, and the data collection module collects the cow udder structure information of the medicating robot and the operation status information of the turntable data panel; S2: Cup removal state recognition and judgment, the AI vision recognition module analyzes the collected images to identify whether the milk scale has removed the cup; the data processing module receives the recognition result of the AI vision recognition module and the data collected by the data collection module, matches the udder structure automatically, combines historical data to compare the fullness of the cow udder, and analyzes the abnormal error information of the turntable data panel at the same time, and comprehensively judges whether the cup removal of the milk cup is normal; S3: Data management: The data processing module summarizes and organizes the cup removal data and uploads it to the cloud data management module, and the cloud data management module stores, analyzes and manages the data.

8. According to the application method of the auxiliary lighting system for cup inspection operation of the automatic milking turntable according to any one of claims 7, in step S2, if the data processing module determines that it is an abnormal cup removal, it sends a control signal to the indicator light module, and the indicator light module lights up the indicator light of the corresponding work station to remind the cup inspection personnel to handle it.

9. According to the application method of the auxiliary lighting system for cup inspection operation of the automatic milking turntable according to any one of claims 7, in step S3, the cloud data management module can also optimize the milking process and equipment operation parameters to promote the intelligent and scientific development of breeding production.

Images