A method, device, controller and system for weighing and self-cleaning

Abstract

The embodiment of the application discloses a kind of self-cleaning methods of weighing, which comprises: obtaining the weight detection information of M weighing units;If the weight detection information corresponding to m1 weighing units in M weighing units indicates that there is a weighing object at m1 weighing units, process the weight detection information corresponding to m1 weighing units, and obtain the weight of the weighing object;If the weight detection information corresponding to m2 weighing units in M weighing units indicates that there is no weighing object at m2 weighing units, control the cleaning unit corresponding to m2 weighing units to clean m2 weighing units;Wherein, 1 < m1 < M, 1 < m2 < M, and M is an integer greater than or equal to 2.The embodiment of the application also discloses a kind of self-cleaning device of weighing, controller and self-cleaning system of weighing.The weighing self-cleaning method provided by the application realizes the combined operation of weighing and self-cleaning, and realizes the effect of automatically weighing the individual weighing object flexibly.

Description

Technical Field

[0001] This application relates to, but is not limited to, the field of information technology, and particularly relates to a weighing and self-cleaning method, a weighing and self-cleaning device, a controller, and a weighing and self-cleaning system. Background Art

[0002] At present, during the breeding process in livestock farms, it is necessary to obtain the weight information of livestock at different growth stages in order to timely understand the growth situation of livestock, and generally the livestock are raised in a confined manner. However, in large livestock farms, due to the excessive number of livestock, at present, manual guidance is used for weighing, and the labor cost of manually weighing livestock is too high, and the weighing frequency of this method is relatively low (for example, once every 10 days). That is to say, in the traditional weighing method, there are problems of high labor cost and inability to accurately obtain the weight of each livestock in real time. At the same time, after traditional farms are used for a period of time, the environment is often poor due to the failure to clean the manure and urine excreted by livestock in time.

[0003] Application Content

[0004] Embodiments of this application are expected to provide a weighing and self-cleaning method, a weighing and self-cleaning device, a controller, and a weighing and self-cleaning system.

[0005] The technical solution of this application is implemented as follows:

[0006] A weighing and self-cleaning method, the method includes:

[0007] Obtain the weight detection information of M weighing units;

[0008] If the weight detection information corresponding to m1 weighing units among the M weighing units indicates that there are weighing objects at the m1 weighing units, process the weight detection information corresponding to the m1 weighing units to obtain the weight of the weighing objects;

[0009] If the weight detection information corresponding to m2 weighing units among the M weighing units indicates that there are no weighing objects at the m2 weighing units, control the cleaning units corresponding to the m2 weighing units to clean the m2 weighing units; where 1 < m1 < M, 1 < m2 < M, and M is an integer greater than or equal to 2.

[0010] A weighing and self-cleaning device, the weighing and self-cleaning device includes:

[0011] An obtaining unit, configured to obtain the weight detection information of M weighing units;

[0012] A processing unit, configured to, if the weight detection information corresponding to m1 weighing units among the M weighing units indicates that there are weighing objects at the m1 weighing units, process the weight detection information corresponding to the m1 weighing units to obtain the weight of the weighing objects;

[0013] A control unit, configured to, if the weight detection information corresponding to m2 of the M weighing units indicates that there is no weighing object at the m2 weighing units, control the cleaning units corresponding to the m2 weighing units to clean the m2 weighing units; wherein, 1 < m1 < M, 1 < m2 < M, and M is an integer greater than or equal to 2.

[0014] A controller, the controller comprising:

[0015] The controller is configured to obtain the weight detection information of M weighing units; if the weight detection information corresponding to m1 of the M weighing units indicates that there is a weighing object at the m1 weighing units, process the weight detection information corresponding to the m1 weighing units to obtain the weight of the weighing object; if the weight detection information corresponding to m2 of the M weighing units indicates that there is no weighing object at the m2 weighing units, control the cleaning units corresponding to the m2 weighing units to clean the m2 weighing units; wherein, 1 < m1 < M, 1 < m2 < M, and M is an integer greater than or equal to 2.

[0016] A weighing and self-cleaning system, the weighing and self-cleaning system comprising:

[0017] M weighing units, a controller, and cleaning units;

[0018] The M weighing units are configured to sense the weight change of the weighing units to obtain detection information;

[0019] The controller is configured to obtain the detection information, if the weight detection information corresponding to m1 of the M weighing units indicates that there is a weighing object at the m1 weighing units, process the weight detection information corresponding to the m1 weighing units to obtain the weight of the weighing object;

[0020] The controller is further configured to, if the weight detection information corresponding to m2 of the M weighing units indicates that there is no weighing object at the m2 weighing units, control the cleaning units corresponding to the m2 weighing units to clean the m2 weighing units; wherein, 1 < m1 < M, 1 < m2 < M, and M is an integer greater than or equal to 2.

[0021] A weighing and self-cleaning method, a weighing and self-cleaning device, a controller, and a weighing and self-cleaning system provided by an embodiment of the present application. The method includes: obtaining weight detection information of M weighing units; if the weight detection information corresponding to m1 weighing units among the M weighing units indicates that there are weighing objects at the m1 weighing units, processing the weight detection information corresponding to the m1 weighing units to obtain the weight of the weighing objects; if the weight detection information corresponding to m2 weighing units among the M weighing units indicates that there are no weighing objects at the m2 weighing units, controlling the cleaning units corresponding to the m2 weighing units to clean the m2 weighing units; where 1 < m1 < M, 1 < m2 < M, and M is an integer greater than or equal to 2. That is to say, in the area covered by the M weighing units in the present application, for the weighing units with weighing objects, the weight of the weighing objects is calculated, realizing the automation of weighing; for the weighing units without weighing objects, cleaning is carried out, realizing the automation of cleaning. It can be seen that in the weighing and self-cleaning method provided by the present application, weighing and self-cleaning can be carried out jointly, and for the entire pen, the effect of automatically weighing individual weighing objects flexibly is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic flowchart of the weighing and self-cleaning method provided by an embodiment of the present application;

[0023] Figure 2 It is a schematic diagram of converting weight into a weight value matrix provided by an embodiment of the present application;

[0024] Figure 3 It is a schematic diagram of the result obtained by processing an image with an intensity diffusion algorithm of Gaussian hypothesis provided by an embodiment of the present application;

[0025] Figure 4 It is a schematic structural diagram of a weighing and self-cleaning system provided by an embodiment of the present application;

[0026] Figure 5 It is a schematic structural diagram of the rolling curtain movement and high-pressure nozzle cleaning of a weighing and self-cleaning system provided by an embodiment of the present application;

[0027] Figure 6 It is a schematic structural diagram of the bottom view inside a weighing and self-cleaning system provided by an embodiment of the present application;

[0028] Figure 7 It is a schematic structural diagram of the left front view inside a weighing and self-cleaning system without a sliding rolling curtain provided by an embodiment of the present application;

[0029] Figure 8 It is a schematic structural diagram of the obliquely upper view inside a weighing and self-cleaning system with a sliding rolling curtain provided by an embodiment of the present application;

[0030] Figure 9 A top view of a weighing self-cleaning system deployment provided for an embodiment of this application;

[0031] Figure 10 A schematic diagram of the structure of a controller provided for an embodiment of this application;

[0032] Figure 11 This is a schematic diagram of a weighing self-cleaning device provided for an embodiment of this application. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations on this application. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0034] In the following description, references are made to “some embodiments,” which describe a subset of all possible embodiments. However, it is understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0035] In the following description, the terms "first, second, third" are used merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first, second, third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of this application described herein can be implemented in a sequence other than that illustrated or described herein.

[0036] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0037] Embodiments of this application provide a weighing self-cleaning method applied to a weighing self-cleaning control device, also known as a weighing self-cleaning controller, or simply a controller; wherein the controller is included in a weighing self-cleaning control system, and the weighing self-cleaning control system further includes M weighing units and a cleaning unit; refer to Figure 1 As shown, the method includes steps 101 to 102; or steps 101 and 103:

[0038] Step 101: Obtain the weight detection information of M weighing units.

[0039] In this embodiment, the weight detection information includes the weight information of the object being weighed detected by the weighing unit, and the identification information of the weighing unit. The identification information of the weighing unit can be used to determine the location of the object being weighed. The controller can determine the area where the object being weighed is located among the M weighing units, and the area where there is no object being weighed, based on the weight detection information obtained from the M weighing units.

[0040] In some embodiments, the weighing array of M weighing units may include a hydraulic gravity sensor, a capacitive gravity sensor that senses the oscillation frequency of the sensing circuit, or an electromagnetic gravity sensor. This application does not specifically limit the specific type of gravity sensor.

[0041] In some embodiments, the controller can be a server that connects to the gateway and communicates with M weighing units and cleaning units. It can run systems such as Ubuntu and deploy a set of service code to process the data sent by the weighing units.

[0042] In this embodiment of the application, taking the weighing array as a piezoelectric device as an example, each weighing unit reads the data of its own piezoelectric device and uploads the read piezoelectric data to the controller. After receiving the piezoelectric data, the controller can also determine which weighing units have weighing objects based on the piezoelectric data, and can also calculate the weight information corresponding to the weighing objects.

[0043] Step 102: If the weight detection information corresponding to m1 of the M weighing units indicates that there is a weighing object at m1 weighing units, process the weight detection information corresponding to m1 weighing units to obtain the weight of the weighing object.

[0044] The regions containing the m1 weighing units can be connected or disconnected, and this application does not impose specific limitations on this. It is understood that, taking a pig as an example, if the region containing the m1 weighing units is connected, the pig's posture is lying down; if the region containing the m1 weighing units is disconnected, the pig's posture is standing on all four hooves.

[0045] In some embodiments, processing the weight detection information corresponding to m1 weighing units can be achieved by obtaining the weight information of the weighing objects corresponding to m1 weighing units and the identification information of the weighing units, and then using a target detection algorithm based on a deep neural network to model and train the identification information of the weighing units according to the image to obtain an RGB image. The position of the weighing object obtained by recognizing the RGB image is used to perform local region weighted cumulative calculation of the weight information of the weighing object according to the position of the weighing object to obtain the corrected weight.

[0046] Step 103: If the weight detection information corresponding to m2 of the M weighing units indicates that there is no weighing object at the m2 weighing units, control the cleaning units corresponding to the m2 weighing units to clean the m2 weighing units.

[0047] Where 1 < m1 < M, 1 < m2 < M, and M is an integer greater than or equal to 2.

[0048] In the embodiment of the present application, taking the weighing object as a pig as an example, during the actual pig raising process, the m2 weighing units are areas where no pigs stay. The controller can control the corresponding cleaning units to clean manure regularly or as needed. In this way, traditional manual manure cleaning is not required. Since the manure cleaning process is automated, it is not necessary for all pigs to be removed from the pen. The cleaning method is flexible, without the need to manually drive the pigs, reducing manual operations.

[0049] It can be understood that as the distribution of pigs in the farm during the actual pig raising process, if the pigs stay in the area where the m1 weighing units are located, the weights of the pigs at the m1 weighing units can be obtained; then, if the pigs leave the area where the m1 weighing units are located, the area where the m1 weighing units are located can be controlled to be cleaned. That is to say, for different areas in the farm, when there is a weighing object, weighing is performed; when there is no weighing object, cleaning is performed, realizing synchronous operation of weighing and cleaning, and flexibly changing the areas to be weighed and cleaned as the distribution of pigs changes.

[0050] A weighing and self-cleaning method, a weighing and self-cleaning device, a controller, and a weighing and self-cleaning system provided by an embodiment of the present application. The method includes: obtaining weight detection information of M weighing units; if the weight detection information corresponding to m1 of the M weighing units indicates that there is a weighing object at the m1 weighing units, processing the weight detection information corresponding to the m1 weighing units to obtain the weight of the weighing object; if the weight detection information corresponding to m2 of the M weighing units indicates that there is no weighing object at the m2 weighing units, controlling the cleaning units corresponding to the m2 weighing units to clean the m2 weighing units; where 1 < m1 < M, 1 < m2 < M, and M is an integer greater than or equal to 2. That is to say, in the area where the M weighing units are laid in the present application, for the weighing units with weighing objects, the weights of the weighing objects are calculated, realizing automatic weighing; for the weighing units without weighing objects, cleaning is performed, realizing automatic cleaning. It can be seen that in the weighing and self-cleaning method provided by the present application, weighing and self-cleaning can be jointly operated, and for the whole pen, the effect of automatically weighing individual weighing objects flexibly is achieved.

[0051] In some embodiments, in step 102, the weight detection information corresponding to the m1 weighing units can be obtained through the following steps:

[0052] Step A1: Obtain the voltage output values ​​of m1 weighing units and the calibration correction coefficients of the weighing units.

[0053] Step A2: Obtain the fusion weights of m1 weighing units under different interval division granularities.

[0054] Step A3: Perform nonlinear processing on the voltage output values ​​of m1 weighing units, the calibration correction coefficients of the weighing units, and the fusion weights to obtain the reference weights corresponding to m1 weighing units.

[0055] Step A4: Correct the reference weights corresponding to the m1 weighing units to obtain the corrected weights corresponding to the m1 weighing units.

[0056] The weight detection information corresponding to m1 weighing units includes the corrected weights corresponding to m1 weighing units.

[0057] In this embodiment, taking a weighing unit containing four piezoelectric modules as an example, each array unit will obtain four piezoelectric data points. Before deploying the weighing unit, each weighing unit will be calibrated. This is because each piezoelectric module has different piezoelectric coefficient nonlinearity, initial values ​​(weight, internal force), and the cleanliness of the sliding roller shutter above the module, so it is necessary to perform factory calibration and periodic correction for each piezoelectric module.

[0058] In this embodiment of the application, the fusion weight under different interval division granularities refers to the fusion weight corresponding to the weight value calculated by different interval division granularities.

[0059] In this embodiment, the reference weights corresponding to m1 weighing units can be modeled and trained by a target detection algorithm based on a deep neural network, which models the weight value matrix as an image.

[0060] In this embodiment of the application, the weight values ​​obtained from each weighing unit can also be combined to obtain a weight matrix.

[0061] Figure 2 A schematic diagram illustrating the conversion of weight into a weight value matrix is ​​provided for embodiments of this application, with reference to... Figure 2 As shown, taking pigs as the weighing object as an example, the weight of the pig is represented as W(x,y), where x and y are the position coordinates of the weighing unit in the overall area of ​​the pigsty, respectively. The controller monitors the pig's position within the pigsty. Figure 2 The weights measured by different weighing units in the entire pigsty area shown on the left side of the middle section are spliced ​​together to obtain... Figure 2 The weight matrix is ​​presented on the right side of the middle section.

[0062] In this way, this application calculates the weight of the weighing object for the weighing unit where the object is to be weighed, and can obtain the weight of livestock in real time and accurately, saving the cost of individual weight measurement and facilitating the information management of the farm; in addition, the weight detection information is optimized by using calibration correction coefficients, fusion weights and correction methods, so that the obtained weight detection information data is accurate, stable and robust.

[0063] In some embodiments, the calibration correction coefficient of the weighing unit obtained in step A1 can be acquired through the following steps:

[0064] When multiple counterweights of different weights are placed on m1 weighing units, obtain multiple sets of original voltage output values ​​of m1 weighing units;

[0065] Obtain the normalized variance of multiple sets of raw voltage output values;

[0066] Based on the no-load voltage of m1 weighing units, multiple sets of original voltage output values ​​and normalized variance, the normalized voltage output values ​​of m1 weighing units are obtained.

[0067] Nonlinear fitting was performed on the normalized voltage output values ​​of m1 weighing units and the weight of the counterweight to obtain the calibration correction coefficients of m1 weighing units under different interval division granularities.

[0068] In this embodiment, the multiple counterweights of different weights can be multiple standard weights ranging from 1 kg to 50 kg, and the placement and statistical process can be carried out using automated testing devices such as robotic arms.

[0069] In this embodiment, the no-load voltage refers to the original voltage output value of the weighing unit when no counterweight is placed on it.

[0070] In this embodiment, the normalized variance of the original voltage output value refers to first normalizing the variance of the original voltage output value, and then normalizing the result of the normalization process. The normalization process is defined as: New value = (Original value - Minimum value) / (Maximum value - Minimum value).

[0071] In this embodiment of the application, the normalized voltage output value refers to the voltage value obtained by summing (original voltage output value - no-load voltage) × normalized variance.

[0072] In this embodiment, the nonlinear fitting can preset the linear interval index value n and the linear interval granularity nl. First, the fusion weights corresponding to different interval division granularities are calculated, and then the normalized voltage output value and the weight of the counterweight are nonlinearly fitted.

[0073] For example, for any one of the m1 weighing units, taking a unit containing 4 gravity sensors as an example, the method for obtaining the calibration correction coefficient is as follows:

[0074] First, using multiple standard weights ranging from 1 kg to 50 kg as counterweights, multiple sets of original voltage output values ​​v can be obtained during the process from having no standard weights (to having a 50 kg standard weight) on the ground to having them on the ground. i w ; where the superscript indicates that the weight of the counterweight is w, and the subscript i indicates the i-th gravity sensor;

[0075] Calculate the variance of multiple sets of original voltage output values Calculate the maximum value again and minimum value The variance is normalized using formula (1):

[0076]

[0077] Secondly, the results of standardization After normalization, the normalized variance is obtained:

[0078]

[0079] Where w_v is the preset variance weighting coefficient; w_var i Normalized variance;

[0080] Next, the normalized voltage output value v' of the weighing unit is calculated using formula (3). w :

[0081]

[0082] Where v0 is the no-load voltage of the weighing unit.

[0083] Finally, assuming nl = [1, 10, 30, 50], combine this with the standard weight W_fm used. w Calculate the calibration correction coefficients:

[0084]

[0085] MAXV is the maximum piezoelectric voltage output setting.

[0086] In this way, the original voltage output value is corrected by using the normalized variance, and then the normalized voltage output value is corrected by the weight correction method of multilinear intervals, so as to obtain a calibration correction coefficient with higher accuracy.

[0087] In some embodiments, obtaining the voltage output values ​​of m1 weighing units in step A1 can be achieved through the following steps:

[0088] Obtain multiple groups of original voltage output values of m1 weighing units within a preset time period;

[0089] Obtain the normalized variance of the variance of multiple groups of original voltage output values at the current moment;

[0090] Determine the fusion coefficient according to the normalized variance and a preset variance weight coefficient;

[0091] Based on the no-load voltages, multiple groups of original voltage output values, and the fusion coefficient of the m1 weighing units, obtain the voltage output values of the m1 weighing units.

[0092] In the embodiments of the present application, the normalized variance refers to the voltage value obtained by normalizing the variance of the original voltage output values.

[0093] In the embodiments of the present application, the fusion coefficient can be obtained by normalizing the normalized contrast with a preset variance weight coefficient.

[0094] It should be noted that the descriptions of the same steps and the same content in this embodiment and other embodiments can be referred to the descriptions in other embodiments, and will not be elaborated here.

[0095] Exemplarily, for any one of the m1 weighing units, the method of obtaining the voltage output value v'(x, y) is similar to obtaining v', w and will not be elaborated here.

[0096]

[0097]

[0098]

[0099] where t represents the current moment, T - T_s < t < T represents the preset time period, (x, y) represents the position of the weighing unit, var i (x, y, T) represents the variance at the current moment, var' i (x, y, T) represents the normalized variance at the current moment, w_var i (x, y) represents the fusion coefficient.

[0100] In this way, by normalizing multiple groups of original voltage output values within a preset time period and combining with the fusion coefficient, a higher-accuracy voltage output value can be obtained.

[0101] In some embodiments, step A2 of obtaining the fusion weights of the m1 weighing units under different interval division granularities can be obtained through the following steps:

[0102] The fusion weights of the m1 weighing units under different interval division granularities are set based on the gradient descent algorithm;

[0103] or,

[0104] From the preset database, based on the interval division granularity index value of the m1 weighing units, the fusion weight of the m1 weighing units is found.

[0105] In some embodiments, step A3 performs nonlinear processing on the voltage output values ​​of m1 weighing units, the calibration correction coefficients of the weighing units, and the fusion weights to obtain the reference weights corresponding to m1 weighing units. A weight matrix can be obtained by calculating the reference weight of each weighing unit and splicing the weight values ​​obtained from each weighing unit.

[0106] For example, for any one of the m1 weighing units, taking nl = [1, 10, 30, 50] as an example, its corresponding reference weight is calculated as follows:

[0107]

[0108] Among them, the linear interval index value n and the linear interval granularity nl can be adjusted according to the actual situation. Generally speaking, the smaller the granularity of nl (e.g., nl=50), the larger wnl is, and the easier it is to overfit.

[0109] In this way, reference weights corresponding to m1 weighing units are obtained through fitting, which facilitates subsequent correction of the weight of the weighing object. Taking pigs as an example, real-time measurement and estimation of the weight of the entire pen can be automated, greatly saving the cost of measuring the weight of the entire pen, and real-time weight acquisition improves the control of the overall physical condition of pigs by farm managers.

[0110] In other embodiments, the weight value matrix is ​​modeled and trained as an image using a target detection algorithm based on a deep neural network. Taking pigs as an example, the weight of pigs in a pen produces a significant weight response (reflected in pixel value changes in the weight value matrix). This response may exhibit a high-intensity response distributed across all four hooves, or a uniform response in a prone position. Therefore, the task of target detection on the weight value matrix is ​​relatively simple. Compared to target detection in RGB images, the data obtained in this embodiment is not affected by factors such as lighting or pig coat color. The current step obtains the pig's position in the weight value matrix. If the pig is standing, the positions of its four hooves are identified; if it is prone, the prone position of its body is identified.

[0111] In some embodiments, step A4 corrects the reference weights corresponding to the m1 weighing units to obtain the corrected weights corresponding to the m1 weighing units, which can be obtained through the following steps:

[0112] Obtain the binarized weight matrix associated with the reference weights of the m1 weighing units;

[0113] Perform connected component detection on the binary weight matrix corresponding to the m1 weighing units, and determine the coordinates of the center point of the detected connected component;

[0114] Obtain the direction and aspect ratio of the connected component;

[0115] Based on the center point coordinates, the direction, and the aspect ratio, a Gaussian convolution kernel is generated that diverges outward from the center point coordinates according to the aspect ratio.

[0116] The reference weights corresponding to the m1 weighing units are diffused using the Gaussian convolution kernel to obtain the corrected weights corresponding to the m1 weighing units.

[0117] In this embodiment of the application, since the excrement of the object being weighed on the weighing unit will cause the weighing result to be inaccurate, the reference weights corresponding to the m1 weighing units can be binarized according to a preset threshold to obtain a binarized matrix.

[0118] For example, the threshold for binarization can be set to 2 kg. Objects weighing less than 2 kg can be considered as feces of the object being weighed, and the error caused by this can be removed by setting the threshold.

[0119] In this way, binarization can remove errors caused by livestock excrement, reduce the dependence of the weighing unit on cleanliness, and improve the accuracy of the reference weight.

[0120] In this embodiment of the application, after obtaining the connected component corresponding to the binarized weight matrix, the coordinates c of the center point of the connected component can be obtained by the following formula. i Direction d i and aspect ratio λ i :

[0121]

[0122] Where i represents a point in the connected domain, (x,y) represents the coordinates of the point in the connected domain, and l i Represents the set of points in a connected domain;

[0123]

[0124] Where l and k represent the two points with the largest width in the connected region, (x l ,y l ), (x k ,y k ) represent the positions of the two points with the largest widths, dist((x) l,y l ),(x k ,y k () represents the distance between the two points with the largest width;

[0125]

[0126] Where, x k x l The x-coordinates of the two points with the greatest width are represented by y. k y l The ordinates of the two points with the greatest width;

[0127]

[0128] Wherein, λ_w i Indicates the width of the connected component;

[0129]

[0130] Wherein, λ_h i The height of the connected component is represented by line((x) k ,y k ),(x l ,y l )) represents a line connecting two points;

[0131]

[0132] In this embodiment, a Gaussian convolution kernel that diverges outward from the center point coordinates according to the aspect ratio can be generated using the following formula based on the center point coordinates, the direction, and the aspect ratio:

[0133]

[0134]

[0135] Where Σ is the covariance matrix corresponding to the connected domain. This represents the number of connected component points, f'(x,y). i Let f(x,y) be the bivariate Gaussian function corresponding to the connected component. i It is a Gaussian convolution kernel.

[0136] In this embodiment of the application, the reference weights corresponding to the m1 weighing units are diffused using the Gaussian convolution kernel to obtain the corrected weights corresponding to the m1 weighing units. The weight values ​​within the connected domain can be weighted and summed by using the Gaussian convolution kernel as a weight.

[0137] For example, this can be achieved using the following formula:

[0138]

[0139] Among them, W(x, y) represents the weight value within the connected region, and W i represents the weight corresponding to the corrected weighing unit.

[0140] Figure 3 It is a schematic diagram of the result obtained by processing an image with an intensity diffusion algorithm based on a Gaussian hypothesis provided by an embodiment of the present application. The result includes Figure 3 the pig detection result presented in the left image in Figure 3 and the diffusion schematic diagram presented in the right image in . Taking the weighing object as a pig as an example, among them Figure 3 the large frame in the lower right corner of the left image is the pig body detection result, and the four small frames in the upper left corner are the pig's hoof detection results; Figure 3 In the right image, the color depth gradually changes from light to deep, representing the change of intensity from weak to strong. Regarding the intensity, it can be understood that the closer to the central position of the weight distribution of the detection object, the greater the intensity and the deeper the color depth; at the edge position, the smaller the intensity and the shallower the color depth.

[0141] In this way, the obtained reference weight is subjected to diffusion processing, so that the obtained weight is closer to the actual value, and the result is more accurate and has strong anti-interference ability.

[0142] Figure 4 It is a schematic diagram of the structure of a weighing and self-cleaning system provided by an embodiment of the present application; referring to Figure 4 as shown, the weighing and self-cleaning system 400 includes M weighing units 401, a controller 402, and a cleaning unit 403, where:

[0143] The M weighing units 401 are used to sense the weight change of the weighing units to obtain detection information;

[0144] The controller 402 is used to obtain the detection information. If the weight detection information corresponding to m1 weighing units among the M weighing units indicates that there is a weighing object at the m1 weighing units, the weight detection information corresponding to the m1 weighing units is processed to obtain the weight of the weighing object;

[0145] The controller 402 is further used to control the cleaning unit corresponding to the m2 weighing units to clean the m2 weighing units if the weight detection information corresponding to the m2 weighing units among the M weighing units indicates that there is no weighing object at the m2 weighing units; where 1 < m1 < M, 1 < m2 < M, and M is an integer greater than or equal to 2.

[0146] The weighing and self-cleaning system provided by the present application can realize the combined operation of weighing and self-cleaning. For the entire pen, it can achieve the effect of automatically weighing the individual weighing objects flexibly, as well as the effect of timely cleaning.

[0147] Furthermore, combined Figure 5 , Figure 7 right Figure 4 The components included in the weighing self-cleaning system 400 will be further described below. Figure 4 The M weighing units 401 include Figure 7 The weighing array 4011 shown Figure 5 The sliding roller shutter 4012 shown and Figure 7 The multiple support columns 4013 shown, among which,

[0148] The sliding roller shutter 4012 is laid on the weighing array 4011;

[0149] Multiple support columns 4013 are used to support the weighing array 4011 and the sliding roller shutter 4012;

[0150] Figure 4 The cleaning unit 403 includes Figure 7 The water storage tank 4031, high-pressure nozzle module 4032, and sewer module 4033 shown are, among which,

[0151] The water storage tank 4031 is housed in the space formed by multiple support columns 4013. The water storage tank 4031 is connected to the high-pressure nozzle module 4032 and is used to supply water to the high-pressure nozzle module 4032.

[0152] The sewer module 4033 is used to discharge water sprayed by the high-pressure nozzle module 4032;

[0153] During the cleaning process, Figure 4 The controller 402 is used to control the sliding roller shutter of m2 weighing units to move the area to be cleaned to the position of the high-pressure nozzle module of the cleaning unit corresponding to the m2 weighing units, and to control the high-pressure nozzle module of the cleaning unit corresponding to the m2 weighing units to spray water onto the area to be cleaned.

[0154] Furthermore, Figure 5 A schematic diagram of the roller shutter movement and high-pressure nozzle cleaning of a weighing self-cleaning system provided for embodiments of this application is shown below. Figure 5 As shown, the weighing self-cleaning system also includes an active adsorption outlet 404.

[0155] For example, Figure 4 The controller 402 controls the sliding roller shutters of m2 weighing units to move the area to be cleaned to the position of the high-pressure nozzle module of the corresponding cleaning unit of the m2 weighing units, and controls the high-pressure nozzle module of the corresponding cleaning unit of the m2 weighing units to spray water onto the area to be cleaned. This can be achieved in the following way:

[0156] The sprayed water passes through Figure 5The active adsorption outlet of the 404 is oriented Figure 7 The storage tank 4031 is pressurized with water, and the weighing unit on the left includes strips made of hard metal. Figure 5 The sliding roller blind 4012 is connected to the drive belt, so that the sliding roller blind 4012 follows the... Figure 5 Rotate in the direction indicated by the middle arrow. When it rotates to the right side of the weighing unit, it is... Figure 5 The high-pressure nozzle 4032 of the cleaning unit on the right sprays water from the water storage tank 4031 to clean the area to be cleaned.

[0157] In this way, not only is automatic cleaning achieved without the need for manual manure removal, but since the cleaning unit only performs automatic manure removal on areas without weighing objects, the automatic manure removal process also has minimal impact on weighing objects.

[0158] Figure 6 A structural schematic diagram of the internal bottom of a weighing self-cleaning system provided for embodiments of this application, with reference to... Figure 6 As shown, the weighing self-cleaning system provided in this application also includes:

[0159] Inlet 405, used for supplying water Figure 5 Water is being added to the storage tank 4031.

[0160] In some embodiments, the water in the storage tank is obtained by high-pressure injection through the inlet, and the inlet is connected to the outlet installed on the ground (i.e., Figure 5 The system can be coupled to the ground outlet (404), which can have an adsorption device similar to an adsorption cap. When the storage tank needs to be filled with water, the controller sends a signal to the ground outlet, which opens the gate and uses the adsorption outlet to inject water under high pressure.

[0161] In some other embodiments, the water inlet 405 is a waterproof power inlet. Correspondingly, each weighing unit is independently powered by the unit's on-board battery, and the battery is charged through the waterproof power inlet at the same location as the water inlet.

[0162] In some embodiments, taking pigs as an example, a track can be first installed on the ground of the pigsty, and a weighing self-cleaning system can be installed on the track. The weighing self-cleaning system consists of an array of multiple weighing units, each weighing unit being approximately 20mm × 10mm × 20mm in size.

[0163] This allows for automated manure removal from pig pens, and the use of weighing units enhances the overall installation flexibility, enabling installation on pens of any shape.

[0164] In some embodiments, the weighing unit includes a piezoelectric device connected to the weighing support, the suction cap, and the sliding roller shutter. When pressure is applied above the weighing unit, the piezoelectric device outputs a change in electrical signal, which the weighing unit then sends to the controller for processing. Each weighing unit is wirelessly controlled by the controller for cleaning and returns a piezoelectric signal to the controller. The controller then analyzes and calculates the data to obtain the weight of each pig.

[0165] Figure 7 This application provides a schematic diagram of the internal left front view of a weighing self-cleaning system without a sliding roller shutter, as part of an embodiment of this application. Figure 4 , Figure 5 and Figure 7 As shown, the weighing unit 401 also includes:

[0166] The roller shutter suction cap 4014 is used to suction the sliding transmission belt 4017 and to drive the transmission belt in the sliding transmission belt 4017 in the suction state.

[0167] Sliding bearing 4015 is used to attach to a sliding belt for sliding;

[0168] The magnetic field generating coil 4016 is used to release or attract sliding roller blinds.

[0169] The sliding transmission belt 4017 drives the sliding roller shutter 4012 to move.

[0170] The movable hub 4018 and track 4019 are used for moving the weighing unit.

[0171] Among them, the roller shutter adsorption cap 4014 is connected to the sliding transmission belt 4017 by magnetic attraction control, the sliding transmission belt 4017 is connected to the sliding bearing 4015 by belt drive; the movable hub 4018 is connected to the track 4019; the magnetic field generating coil 4016 is connected to the weighing support column 4013; and the weighing array 4011 is connected to the weighing support column 4013 by stacking them vertically.

[0172] See Figure 7 The self-cleaning weighing system of this application adopts a sliding roller shutter and electromagnetic adsorption scheme to achieve independent deployment of weighing units. The weighing units are deployed in an array, and after deployment, they work together, are automatically cleaned, and have piezoelectric data automatically acquired. This achieves the technical effect of real-time automatic acquisition of piezoelectric array data of the weighing object and real-time calculation of the weight of the weighing object.

[0173] In some embodiments, the controller controls the sliding roller shutters of m2 weighing units to move the area to be cleaned to a position facing the high-pressure nozzle module of the cleaning unit corresponding to the m2 weighing units, and controls the high-pressure nozzle module of the cleaning unit corresponding to the m2 weighing units to spray water onto the area to be cleaned. This can be achieved in the following way:

[0174] When the magnetic field generating coil 4016 is de-energized, the sliding roller shutter 4012, which is adsorbed on the roller shutter adsorption cap 4014, is released.

[0175] The rotation of the sliding transmission belt 4017 drives the transmission belt to rotate, which in turn moves the sliding roller shutter 4012 to the right side of the weighing unit.

[0176] The adjacent units of the weighing unit that need cleaning are sprayed with their own high-pressure nozzles 4032 to clean the area to be cleaned.

[0177] The reverse rotation of the sliding transmission belt 4017 causes the sliding roller shutter 4012 to return to its original position.

[0178] When the magnetic field generating coil 4016 is energized, the sliding roller shutter is attracted to the roller shutter adsorption cap 4014.

[0179] In this way, not only is automatic cleaning achieved without the need for manual manure removal, but since the cleaning unit only performs automatic manure removal on areas without weighing objects, the automatic manure removal process also has minimal impact on weighing objects.

[0180] Figure 8 A structural schematic diagram of a weighing self-cleaning system with an added sliding roller shutter, provided as an embodiment of this application, viewed from an oblique upper angle. Figure 8 As shown, the weighing self-cleaning system provided in this application also includes:

[0181] Status indicator light 406 is used to indicate whether the weighing self-cleaning system is malfunctioning or operating normally.

[0182] Due to the harsh environment of farms, the weighing self-cleaning system often gets damaged due to erosion by manure and urine after a period of use. In this way, according to the status indicator light, only the damaged unit can be replaced after the weighing unit is damaged, without spending time on repairs. This achieves automated maintenance and reduces the overall maintenance cost of the system.

[0183] Figure 9 A top view schematic diagram of the deployment of a weighing self-cleaning system provided for an embodiment of this application, with reference to... Figure 9 Area A is the array area where self-cleaning is not enabled, and area B is the array area where self-cleaning is enabled.

[0184] Embodiments of this application provide a controller 402, which can be applied to... Figure 1In a weighing and self-cleaning method provided by a corresponding embodiment, refer to Figure 10 As shown in Figure 10 , the controller 402 includes: a processor 4021, a memory 4022, and a communication bus 4023, where:

[0185] The communication bus 4023 is used to implement a communication connection between the processor 4021 and the memory 4022.

[0186] The processor 4021 is configured to obtain weight detection information of M weighing units; if the weight detection information corresponding to m1 weighing units among the M weighing units indicates that there are weighing objects at the m1 weighing units, process the weight detection information corresponding to the m1 weighing units to obtain the weight of the weighing object; if the weight detection information corresponding to m2 weighing units among the M weighing units indicates that there are no weighing objects at the m2 weighing units, control the cleaning units corresponding to the m2 weighing units to clean the m2 weighing units; where 1 < m1 < M, 1 < m2 < M, and M is an integer greater than or equal to 2.

[0187] In other embodiments of the present application, the processor 4021 is further configured to:

[0188] Obtain the voltage output values of the m1 weighing units and the calibration correction coefficients of the weighing units;

[0189] Obtain the fusion weights of the m1 weighing units under different interval division granularities;

[0190] Perform non-linear processing on the voltage output values of the m1 weighing units, the calibration correction coefficients of the weighing units, and the fusion weights to obtain the reference weights corresponding to the m1 weighing units;

[0191] Correct the reference weights corresponding to the m1 weighing units to obtain the weights corresponding to the m1 weighing units after correction; where the weight detection information corresponding to the m1 weighing units includes the weights corresponding to the m1 weighing units after correction.

[0192] In other embodiments of the present application, the processor 4021 is further configured to:

[0193] Obtain multiple sets of original voltage output values of the m1 weighing units when counterweights of multiple different weights are placed on the m1 weighing units;

[0194] Obtain the normalized variance of the multiple sets of original voltage output values;

[0195] Based on the no-load voltages of the m1 weighing units, the multiple sets of original voltage output values, and the normalized variance, obtain the normalized voltage output values of the m1 weighing units;

[0196] The normalized voltage output values ​​of the m1 weighing units and the weight of the counterweight are nonlinearly fitted to obtain the calibration correction coefficients of the m1 weighing units under different interval division granularities.

[0197] In other embodiments of this application, the processor 4021 is further configured to:

[0198] Obtain the binarized weight matrix associated with the reference weights of the m1 weighing units;

[0199] Perform connected component detection on the binary weight matrix corresponding to the m1 weighing units, and determine the coordinates of the center point of the detected connected component;

[0200] Obtain the direction and aspect ratio of the connected component;

[0201] Based on the center point coordinates, the direction, and the aspect ratio, a Gaussian convolution kernel is generated that diverges outward from the center point coordinates according to the aspect ratio.

[0202] The reference weights corresponding to the m1 weighing units are diffused using the Gaussian convolution kernel to obtain the corrected weights corresponding to the m1 weighing units.

[0203] In other embodiments of this application, the processor 4021 is further configured to:

[0204] Obtain multiple sets of original voltage output values ​​of the m1 weighing units within a preset time period;

[0205] Obtain the normalized variance of the variance of the multiple sets of original voltage output values ​​at the current moment;

[0206] The fusion coefficient is determined based on the normalized variance and the preset variance weighting coefficient.

[0207] The voltage output values ​​of the m1 weighing units are obtained based on the no-load voltage of the m1 weighing units, the multiple sets of original voltage output values, and the fusion coefficient.

[0208] In other embodiments of this application, the processor 4021 is further configured to:

[0209] The cleaning area of ​​the sliding roller shutter of the m2 weighing units is moved to a position facing the high-pressure nozzle module of the cleaning unit corresponding to the m2 weighing units, and the high-pressure nozzle module of the cleaning unit corresponding to the m2 weighing units is controlled to spray water onto the cleaning area.

[0210] In the area covered by M weighing units, for the weighing units with weighing objects, the weights of the weighing objects are calculated, realizing the automation of weighing; for the weighing units without weighing objects, cleaning is carried out, realizing the automation of cleaning. It can be seen that in the weighing and self-cleaning method provided in this application, weighing and self-cleaning can be carried out jointly, and for the entire pen, the effect of flexibly and automatically weighing individual weighing objects is achieved.

[0211] Figure 11 An embodiment of this application provides a weighing and self-cleaning device, which can be applied to Figure 1 the method provided in the corresponding embodiment, refer to Figure 11 As shown, the weighing and self-cleaning device 11 includes:

[0212] An acquisition unit 1101, configured to acquire the weight detection information of M weighing units;

[0213] A processing unit 1102, configured to, if the weight detection information corresponding to m1 weighing units among the M weighing units indicates that there are weighing objects at the m1 weighing units, process the weight detection information corresponding to the m1 weighing units to obtain the weights of the weighing objects;

[0214] A control unit 1103, configured to, if the weight detection information corresponding to m2 weighing units among the M weighing units indicates that there are no weighing objects at the m2 weighing units, control the cleaning units corresponding to the m2 weighing units to clean the m2 weighing units; where 1 < m1 < M, 1 < m2 < M, and M is an integer greater than or equal to 2.

[0215] In other embodiments of this application, the processing unit 1102 is further configured to:

[0216] Obtain the voltage output values of the m1 weighing units and the calibration correction coefficients of the weighing units;

[0217] Obtain the fusion weights of the m1 weighing units under different interval division granularities;

[0218] Perform non-linear processing on the voltage output values, calibration correction coefficients, and fusion weights of the m1 weighing units to obtain the reference weights corresponding to the m1 weighing units;

[0219] Correct the reference weights corresponding to the m1 weighing units to obtain the weights corresponding to the corrected m1 weighing units; where the weight detection information corresponding to the m1 weighing units includes the weights corresponding to the corrected m1 weighing units.

[0220] In other embodiments of this application, the processing unit 1102 is further configured to:

[0221] When multiple counterweights of different weights are placed on m1 weighing units, obtain multiple sets of original voltage output values ​​of m1 weighing units;

[0222] Obtain the normalized variance of multiple sets of raw voltage output values;

[0223] Based on the no-load voltage of m1 weighing units, multiple sets of original voltage output values ​​and normalized variance, the normalized voltage output values ​​of m1 weighing units are obtained.

[0224] Nonlinear fitting was performed on the normalized voltage output values ​​of m1 weighing units and the weight of the counterweight to obtain the calibration correction coefficients of m1 weighing units under different interval division granularities.

[0225] In other embodiments of this application, the processing unit 1102 is further configured to:

[0226] Obtain the binary weight matrix associated with the reference weights of m1 weighing units;

[0227] Perform connected component detection on the binary weight matrix corresponding to m1 weighing units, and determine the coordinates of the center point of the detected connected component;

[0228] Get the direction and aspect ratio of the connected components;

[0229] Based on the center point coordinates, direction, and aspect ratio, a Gaussian convolution kernel is generated that diverges outward from the center point coordinates according to the aspect ratio.

[0230] The reference weights corresponding to m1 weighing units are diffused using Gaussian convolution kernels to obtain the corrected weights corresponding to m1 weighing units.

[0231] In other embodiments of this application, the processing unit 1102 is further configured to:

[0232] Obtain multiple sets of raw voltage output values ​​of m1 weighing units within a preset time period;

[0233] Obtain the normalized variance of the variance of multiple sets of raw voltage output values ​​at the current moment;

[0234] The fusion coefficient is determined based on the normalized variance and the preset variance weighting coefficients;

[0235] Based on the no-load voltage of m1 weighing units, multiple sets of original voltage output values, and fusion coefficients, the voltage output values ​​of m1 weighing units are obtained.

[0236] In other embodiments of this application, the control unit 1103 is further configured to:

[0237] Control the sliding roller shutter of m2 weighing units to move the area to be cleaned to the position of the high-pressure nozzle module of the cleaning unit corresponding to the m2 weighing units, and control the high-pressure nozzle module of the cleaning unit corresponding to the m2 weighing units to spray water onto the area to be cleaned.

[0238] It should be noted that the descriptions of the same steps and contents as in other embodiments in this embodiment can be found in the descriptions in other embodiments, and will not be repeated here.

[0239] Embodiments of this application provide a computer-readable storage medium storing one or more programs that can be executed by one or more processors to perform, as follows: Figure 1 The implementation process of the weighing self-cleaning method provided in the corresponding embodiment will not be described in detail here.

[0240] It should be noted that the descriptions of the storage medium and device embodiments above are similar to the descriptions of the method embodiments above, and have similar beneficial effects. For technical details not disclosed in the storage medium and device embodiments of this application, please refer to the descriptions of the method embodiments of this application for understanding.

[0241] The aforementioned computer storage media / memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), magnetic random access memory (FRAM), flash memory, magnetic surface memory, optical disc, or compact disc read-only memory (CD-ROM), etc.; or it can be various terminals that include one or any combination of the above-mentioned memories, such as mobile phones, computers, tablet devices, personal digital assistants, etc.

[0242] It should be understood that the terms "an embodiment," "an embodiment," "an embodiment of this application," "the foregoing embodiment," "some embodiments," or "some implementations" mentioned throughout the specification mean that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, the phrases "an embodiment," "an embodiment," "an embodiment of this application," "the foregoing embodiment," "some embodiments," or "some implementations" appearing throughout the specification do not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. It should be understood that in the various embodiments of this application, the sequence numbers of the above processes do not imply a sequential order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application. The sequence numbers of the above embodiments of this application are merely descriptive and do not represent the superiority or inferiority of the embodiments.

[0243] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0244] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0245] In addition, each functional unit in the various embodiments of this application can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0246] The methods disclosed in the several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0247] The features disclosed in the several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0248] The features disclosed in the several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method or device embodiments.

[0249] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media that can store program code, such as mobile storage devices, read-only memory (ROM), magnetic disks, or optical disks.

[0250] Alternatively, if the integrated units described above are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of this application, or the parts that contribute to related technologies, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0251] It is worth noting that the accompanying drawings in this application are only for illustrating the schematic positions of various devices on the terminal device and do not represent their actual positions in the terminal device. The actual positions of each device or area may be changed or shifted according to the actual situation (e.g., the structure of the terminal device). Furthermore, the proportions of different parts in the terminal device in the drawings do not represent the actual proportions.

[0252] The above description is merely an embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A weighing self-cleaning control method, characterized in that, The method includes: Obtain weight detection information from M weighing units; the M weighing units include a weighing array; If the weight detection information corresponding to m1 of the M weighing units indicates that there is a weighing object at m1 weighing units, process the weight detection information corresponding to m1 weighing units to obtain the weight of the weighing object; If the weight detection information of m² weighing units out of the M weighing units indicates that there is no object to be weighed at those m² weighing units, then the cleaning units corresponding to those m² weighing units are controlled to clean the m² weighing units; wherein... , And M is an integer greater than or equal to 2; The weight detection information corresponding to the m1 weighing units is obtained through the following steps: Obtain the voltage output values ​​of the m1 weighing units and the calibration correction coefficients of the weighing units; Obtain the fusion weights of the m1 weighing units under different interval division granularities; The voltage output values ​​of the m1 weighing units, the calibration correction coefficients of the weighing units, and the fusion weights are subjected to nonlinear processing to obtain the reference weights corresponding to the m1 weighing units. Obtain the binarized weight matrix associated with the reference weights of the m1 weighing units; Perform connected component detection on the binary weight matrix corresponding to the m1 weighing units, and determine the coordinates of the center point of the detected connected component; Obtain the direction and aspect ratio of the connected component; Based on the center point coordinates, the direction, and the aspect ratio, a Gaussian convolution kernel is generated that diverges outward from the center point coordinates according to the aspect ratio. The reference weights corresponding to the m1 weighing units are diffused using the Gaussian convolution kernel to obtain the corrected weights corresponding to the m1 weighing units; wherein, the weight detection information corresponding to the m1 weighing units includes the corrected weights corresponding to the m1 weighing units.

2. The method according to claim 1, characterized in that, Obtaining the calibration correction coefficients of the m1 weighing units includes: When multiple counterweights of different weights are placed on the m1 weighing units, obtain multiple sets of original voltage output values ​​of the m1 weighing units; Obtain the normalized variance of the multiple sets of original voltage output values; Based on the no-load voltage of the m1 weighing units, the multiple sets of original voltage output values ​​and the normalized variance, the normalized voltage output values ​​of the m1 weighing units are obtained. The normalized voltage output values ​​of the m1 weighing units and the weight of the counterweight are nonlinearly fitted to obtain the calibration correction coefficients of the m1 weighing units under different interval division granularities.

3. The method according to claim 1, characterized in that, Obtaining the voltage output values ​​of the m1 weighing units includes: Obtain multiple sets of original voltage output values ​​of the m1 weighing units within a preset time period; Obtain the normalized variance of the variance of the multiple sets of original voltage output values ​​at the current moment; The fusion coefficient is determined based on the normalized variance and the preset variance weighting coefficient. The voltage output values ​​of the m1 weighing units are obtained based on the no-load voltage of the m1 weighing units, the multiple sets of original voltage output values, and the fusion coefficient.

4. The method according to claim 1, characterized in that, The step of controlling the cleaning units corresponding to the m2 weighing units to clean the m2 weighing units includes: The cleaning area of ​​the sliding roller shutter of the m2 weighing units is moved to a position facing the high-pressure nozzle module of the cleaning unit corresponding to the m2 weighing units, and the high-pressure nozzle module of the cleaning unit corresponding to the m2 weighing units is controlled to spray water onto the cleaning area.

5. A weighing self-cleaning control device, characterized in that, The weighing self-cleaning control device includes: An acquisition unit is used to acquire weight detection information from M weighing units; the M weighing units include a weighing array. The processing unit is configured to, if the weight detection information corresponding to m1 of the M weighing units indicates that there is a weighing object at the m1 weighing units, process the weight detection information corresponding to the m1 weighing units to obtain the weight of the weighing object. The control unit is configured to, if the weight detection information of m² weighing units out of the M weighing units indicates that there is no object to be weighed at the m² weighing units, control the cleaning unit corresponding to the m² weighing units to clean the m² weighing units; wherein... , And M is an integer greater than or equal to 2; The processing unit is further configured to: obtain the voltage output values ​​of the m1 weighing units and the calibration correction coefficients of the weighing units; obtain the fusion weights of the m1 weighing units under different interval division granularities; perform nonlinear processing on the voltage output values ​​of the m1 weighing units, the calibration correction coefficients of the weighing units, and the fusion weights to obtain the reference weights corresponding to the m1 weighing units; obtain the binarized weight matrix associated with the reference weights corresponding to the m1 weighing units; perform connected component detection on the binarized weight matrix corresponding to the m1 weighing units and determine the coordinates of the center point of the detected connected component; obtain the direction and aspect ratio of the connected component; generate a Gaussian convolution kernel that diverges outward from the center point coordinates according to the aspect ratio based on the center point coordinates, the direction, and the aspect ratio; and use the Gaussian convolution kernel to diffuse the reference weights corresponding to the m1 weighing units to obtain the corrected weights corresponding to the m1 weighing units; wherein, the weight detection information corresponding to the m1 weighing units includes the corrected weights corresponding to the m1 weighing units.

6. A controller, characterized in that, The controller is used to acquire weight detection information from M weighing units; the M weighing units include a weighing array; if the weight detection information corresponding to m1 weighing units among the M weighing units indicates that there is a weighing object at the m1 weighing units, the controller processes the weight detection information corresponding to the m1 weighing units to obtain the weight of the weighing object; if the weight detection information corresponding to m2 weighing units among the M weighing units indicates that there is no weighing object at the m2 weighing units, the controller controls the cleaning units corresponding to the m2 weighing units to clean the m2 weighing units; wherein... , And M is an integer greater than or equal to 2; The controller is further configured to: obtain the voltage output values ​​of the m1 weighing units and the calibration correction coefficients of the weighing units; obtain the fusion weights of the m1 weighing units under different interval division granularities; perform nonlinear processing on the voltage output values ​​of the m1 weighing units, the calibration correction coefficients of the weighing units, and the fusion weights to obtain the reference weights corresponding to the m1 weighing units; obtain the binarized weight matrix associated with the reference weights corresponding to the m1 weighing units; perform connected component detection on the binarized weight matrix corresponding to the m1 weighing units and determine the coordinates of the center point of the detected connected component; obtain the direction and aspect ratio of the connected component; generate a Gaussian convolution kernel that diverges outward from the center point coordinates according to the aspect ratio based on the center point coordinates, the direction, and the aspect ratio; and use the Gaussian convolution kernel to diffuse the reference weights corresponding to the m1 weighing units to obtain the corrected weights corresponding to the m1 weighing units; wherein, the weight detection information corresponding to the m1 weighing units includes the corrected weights corresponding to the m1 weighing units.

7. A weighing self-cleaning system, characterized in that, The weighing self-cleaning system includes M weighing units, a controller, and a cleaning unit; The M weighing units are used to sense changes in the weight of the weighing unit to obtain detection information; the M weighing units include a weighing array; The controller is used to acquire the detection information. If the weight detection information corresponding to m1 weighing units out of the M weighing units indicates that there is a weighing object at the m1 weighing units, the controller processes the weight detection information corresponding to the m1 weighing units to obtain the weight of the weighing object. The controller is further configured to, if the weight detection information corresponding to m² of the M weighing units indicates that there is no object to be weighed at the m² weighing units, control the cleaning unit corresponding to the m² weighing units to clean the m² weighing units; wherein... , And M is an integer greater than or equal to 2; The controller is further configured to: obtain the voltage output values ​​of the m1 weighing units and the calibration correction coefficients of the weighing units; obtain the fusion weights of the m1 weighing units under different interval division granularities; perform nonlinear processing on the voltage output values ​​of the m1 weighing units, the calibration correction coefficients of the weighing units, and the fusion weights to obtain the reference weights corresponding to the m1 weighing units; obtain the binarized weight matrix associated with the reference weights corresponding to the m1 weighing units; perform connected component detection on the binarized weight matrix corresponding to the m1 weighing units and determine the coordinates of the center point of the detected connected component; obtain the direction and aspect ratio of the connected component; generate a Gaussian convolution kernel that diverges outward from the center point coordinates according to the aspect ratio based on the center point coordinates, the direction, and the aspect ratio; and use the Gaussian convolution kernel to diffuse the reference weights corresponding to the m1 weighing units to obtain the corrected weights corresponding to the m1 weighing units; wherein, the weight detection information corresponding to the m1 weighing units includes the corrected weights corresponding to the m1 weighing units.

8. The system according to claim 7, characterized in that, The M weighing units also include sliding roller shutters and multiple support columns, wherein, The sliding roller shutter is laid on the weighing array; The plurality of support columns are used to support the weighing array and the sliding roller shutter; The cleaning unit includes a water storage tank, a high-pressure nozzle module, and a sewer module, wherein... The water storage tank is housed in the space formed by the plurality of support columns, and the water storage tank is connected to the high-pressure nozzle module to supply water to the high-pressure nozzle module. The sewer module is used to discharge the water sprayed by the high-pressure nozzle module; The controller is used to control the area to be cleaned of the sliding roller shutter of the m2 weighing units to move toward the position of the high-pressure nozzle module of the cleaning unit corresponding to the m2 weighing units, and to control the high-pressure nozzle module of the cleaning unit corresponding to the m2 weighing units to spray water onto the area to be cleaned.

Images