Product cleaning equipment and method thereof

Abstract

The invention discloses product cleaning equipment and a method thereof, and relates to the technical field of cleaning equipment, the product cleaning equipment comprises an information acquisition module, the information acquisition module comprises an image acquisition unit, and the image acquisition unit obtains water body image information; the cleaning monitoring module is used for receiving the water body image information, preprocessing the water body image information, carrying out gray gradient co-occurrence matrix calculation according to the water body image preprocessing information, calculating water body characteristic parameters based on a gray gradient co-occurrence matrix calculation result, and carrying out turbidity grade matching based on the water body characteristic parameters; and performing water body replacement judgment on the turbidity grade matching result, and generating a water body replacement control signal according to a water body replacement judgment result. According to the invention, the turbidity of the water body of the cleaning machine can be monitored in real time, and the water body replacement of the cleaning module is controlled by judging the turbidity of the water body, so that the automatic control of the water body replacement of the cleaning module is realized, and the labor intensity is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of cleaning equipment, and in particular to a product cleaning equipment and a method thereof. Background Art

[0002] Sauerkraut, as a traditional fermented food deeply loved by people, occupies an important position in the market due to its unique taste and rich nutritional value. In the production process of sauerkraut, cleaning of raw materials is one of the key steps to ensure product quality and safety. Cleaning can not only effectively remove dirt, impurities and residual pesticides on the surface of raw materials, but also provide a good sanitary environment for the subsequent fermentation process, prevent the growth of harmful microorganisms, thereby ensuring the stability of sauerkraut during the fermentation process and the quality of the final product.

[0003] At present, in the sauerkraut production industry, the commonly used cleaning method is to clean the raw materials through a spray system. The spray system uses high-pressure water to wash away the dirt on the surface of the raw materials. This cleaning method can meet production needs to a certain extent. However, with the continuous deepening of practical applications, this cleaning method has also exposed some significant technical defects.

[0004] As the cleaning process proceeds, dirt on the surface of the raw materials is continuously washed off and enters the cleaning water, causing the water to gradually become turbid. The turbidity of the water not only affects the cleaning effect, but is also directly related to the sanitary conditions of the subsequent fermentation process. When the water is turbid to a certain extent, in order to ensure the cleaning quality and the smooth progress of subsequent production, the water needs to be replaced. However, in the prior art, the turbidity of the water is usually judged by naked eye observation, which is not only labor-intensive, but also greatly affected by human factors and has low judgment accuracy. Summary of the invention

[0005] Therefore, in order to solve the above-mentioned shortcomings, on the one hand, the present invention provides a product cleaning device, comprising: An information acquisition module, the information acquisition module includes an image acquisition unit, the image acquisition unit acquires water body image information; a cleaning monitoring module, the cleaning monitoring module receiving water body image information, preprocessing the water body image information, obtaining water body image preprocessing information, calculating a grayscale gradient co-occurrence matrix according to the water body image preprocessing information, calculating water body characteristic parameters based on the grayscale gradient co-occurrence matrix calculation result, performing turbidity level matching based on the water body characteristic parameters, performing water body replacement judgment on the turbidity level matching result, and generating a water body replacement control signal according to the water body replacement judgment result; The cleaning module obtains a water body replacement control signal and executes an action corresponding to the water body replacement control signal based on the water body replacement control signal.

[0006] Furthermore, the grayscale gradient symbiosis calculation includes: Process the water body image preprocessing information to obtain a grayscale matrix; The Sobel operator is used to perform convolution operation on the gray matrix to obtain the gradient matrix of the water body image preprocessing information; The element values ​​of the grayscale gradient co-occurrence matrix are counted according to the grayscale matrix and the gradient matrix, and the element value statistical results are normalized to obtain the grayscale gradient co-occurrence matrix.

[0007] Furthermore, the water body characteristic parameters include grayscale energy and gradient energy; The calculation of water body characteristic parameters based on the grayscale gradient co-occurrence matrix calculation results includes: The grayscale energy of the water body image preprocessing information is calculated as follows: ; in, U F It is the gray energy calculation result; N g is the selected gray level; x , y are the horizontal and vertical coordinates of the pixels in the water body image information; N g is the selected gradient level; h ( x,y ) is the grayscale gradient co-occurrence matrix; The gradient energy of the water body image preprocessing information is calculated as follows: ; Among them, U G Grayscale energy calculation result.

[0008] Furthermore, before using the Sobel operator to perform convolution operation on the grayscale matrix, the grayscale matrix is ​​corrected and calculated as follows: ; in, F ( x,y ) is the correction calculation result of the gray matrix in the water body image preprocessing information; G min is the maximum grayscale value in the grayscale matrix; G max is the minimum gray value in the gray matrix; d is the control coefficient; s ( x,y ) is the illumination amount of the pixel; h ( x,y ) is the incident light amount.

[0009] Furthermore, the turbidity level matching based on the water body characteristic parameters is to input the water body characteristic parameters into the turbidity level model for matching, and determine the turbidity level according to the matching result.

[0010] Furthermore, the data sample set is obtained in the following manner: Obtain sample water body images and turbidity levels, process the sample water body images, and obtain sample water body parameters; The sample water parameters and turbidity values ​​are input into a neural network model based on the BP neural network framework for training to obtain a turbidity grade matching model.

[0011] Furthermore, the cleaning module comprises: a water tank, which collects the oil inlet and outlet; A lifting conveyor belt, the lower part of which is arranged in the water tank; A spray pipe, which is arranged above the lower part of the lifting conveyor belt and fixedly installed in the water tank; A filter box, wherein the filter box is arranged on one side of the water tank, and an upper portion of one side of the filter box is connected to an upper portion of one side of the water tank; A circulation pump is used to connect the filter box to the spray pipe.

[0012] On the other hand, the present invention also provides a product cleaning method, which uses the above-mentioned product cleaning device, and the cleaning method comprises: Receiving water body image information, preprocessing the water body image information, and obtaining water body image preprocessing information; The grayscale gradient co-occurrence matrix is ​​calculated based on the pre-processing information of the water body image, the water body characteristic parameters are calculated based on the calculation results of the grayscale gradient co-occurrence matrix, and the turbidity level matching is performed based on the water body characteristic parameters; A water body replacement judgment is made based on the turbidity level matching result, and a water body replacement control signal is generated based on the water body replacement judgment result.

[0013] Furthermore, the grayscale gradient symbiosis calculation includes: Process the water body image preprocessing information to obtain a grayscale matrix; The Sobel operator is used to perform convolution operation on the gray matrix to obtain the gradient matrix of the water body image preprocessing information; The element values ​​of the grayscale gradient co-occurrence matrix are counted according to the grayscale matrix and the gradient matrix, and the element value statistical results are normalized to obtain the grayscale gradient co-occurrence matrix.

[0014] The present invention has the following advantages: The present invention can realize real-time monitoring of the turbidity of the water in the cleaning machine, and control the water replacement of the cleaning module by judging the turbidity of the water, thereby realizing automatic control of the water replacement of the cleaning module and reducing labor intensity. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the logical structure of the cleaning equipment; Figure 2 yes Figure 1 A schematic diagram of the structure of the information collection module in the cleaning device shown; Figure 3 yes Figure 1 A schematic diagram of the logical structure of the cleaning monitoring module in the cleaning device shown; Figure 4 yes Figure 1 A schematic diagram of the structure of the cleaning module in the cleaning device shown.

[0016] In the figure: 100, information acquisition module; 110, sensor unit; 120, image acquisition unit; 200, cleaning monitoring module; 210, data acquisition unit; 220, turbidity monitoring unit; 230, cleaning control unit; 300. Remote monitoring module; 400, cleaning module; 410, water tank; 420, spray pipe; 430, water inlet; 440, circulation pump; 450, filter tank. DETAILED DESCRIPTION

[0017] Embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present application, and should not be construed as limiting the present application.

[0018] In this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises", or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or apparatus that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article, or apparatus.

[0019] As described in the background technology, existing cleaning equipment mainly relies on manual operation for flow control. Operators need to manually adjust the flow of the spray equipment based on experience or observed cleaning conditions. This method is not only inefficient, but also difficult to control accurately. Due to the intervention of human factors, the flow adjustment often has lags and inaccuracies, and it is impossible to make timely and effective adjustments based on actual flow conditions.

[0020] Embodiment 1: Therefore, in order to solve the above technical problems of the prior art, this embodiment provides a product cleaning device, such as Figure 1 As shown, the cleaning device comprises: Information collection module 100, such as Figure 2 As shown, the information acquisition module includes an image acquisition unit 120, which acquires water body image information. Specifically, the image acquisition unit is a high-resolution camera or image acquisition device, and the water body image is acquired through the camera or image acquisition device; A cleaning monitoring module 200, which receives water body image information, pre-processes the water body image information, obtains water body image pre-processing information, calculates a grayscale gradient co-occurrence matrix based on the water body image pre-processing information, calculates water body characteristic parameters based on the grayscale gradient co-occurrence matrix calculation result, performs turbidity level matching based on the water body characteristic parameters, performs water body replacement judgment on the turbidity level matching result, and generates a water body replacement control signal according to the water body replacement judgment result; The cleaning module 400 obtains a water body replacement control signal, and executes an action corresponding to the water body replacement control signal based on the water body replacement control signal.

[0021] For example, Figure 3 As shown, the cleaning monitoring module may include: A data acquisition unit 210 receives water body image information; The turbidity monitoring unit 220 pre-processes the water body image information to obtain the water body image pre-processing information, calculates the gray gradient co-occurrence matrix according to the water body image pre-processing information, calculates the water body characteristic parameters based on the gray gradient co-occurrence matrix calculation result, performs turbidity level matching based on the water body characteristic parameters, performs water body replacement judgment on the turbidity level matching result, and obtains the water body replacement judgment result; The cleaning control unit 230 obtains a water body replacement judgment result, generates a water body replacement control signal according to the water body replacement judgment result, and sends the water body replacement control signal to the cleaning module so that the cleaning module performs an action corresponding to the water body replacement control signal based on the water body replacement control signal.

[0022] Specifically, the preprocessing of the water body image information includes converting the size and grayscale of the water body image.

[0023] In this embodiment, the grayscale gradient symbiosis calculation includes: Processing the water body image preprocessing information to obtain a grayscale matrix, wherein the grayscale matrix is ​​a matrix composed of grayscale values ​​of image pixels after the image is converted into a grayscale image; The Sobel operator is used to perform convolution operation on the gray matrix to obtain the gradient matrix of the water body image preprocessing information; The element values ​​of the grayscale gradient co-occurrence matrix are counted according to the grayscale matrix and the gradient matrix, and the element value statistical results are normalized to obtain the grayscale gradient co-occurrence matrix.

[0024] Specifically, by setting a grayscale level range and a gradient level range, the elements in the grayscale matrix and the gradient matrix are quantized into these levels. For each pixel in the image, its grayscale level and gradient level are considered at the same time, and the frequency of the grayscale level and gradient level appearing at the same time in the specified direction and distance is counted. These frequencies form a two-dimensional matrix, namely the grayscale gradient co-occurrence matrix. The rows of the matrix represent the grayscale level, the columns represent the gradient level, and the element value represents the number of times the corresponding grayscale level and gradient level appear at the same time.

[0025] In this embodiment, the water body characteristic parameters include grayscale energy and gradient energy; The calculation of water body characteristic parameters based on the grayscale gradient co-occurrence matrix calculation results includes: The grayscale energy of the water body image preprocessing information is calculated as follows: ; in, U F It is the gray energy calculation result; N g is the selected gray level; x , y are the horizontal and vertical coordinates of the pixels in the water body image information; N g is the selected gradient level; h ( x,y ) is the grayscale gradient co-occurrence matrix; The gradient energy of the water body image preprocessing information is calculated as follows: ; Among them, U G Grayscale energy calculation result.

[0026] In this embodiment, before using the Sobel operator to perform convolution operation on the gray matrix, a correction calculation is performed on the gray matrix, and the calculation method is as follows: ; in, F ( x,y ) is the correction calculation result of the gray matrix in the water body image preprocessing information; G min is the maximum grayscale value in the grayscale matrix; G max is the minimum gray value in the gray matrix; d is the control coefficient; s ( x,y ) is the illumination amount of the pixel; h ( x,y ) is the incident light amount.

[0027] In this embodiment, the turbidity level matching based on the water body characteristic parameters is to input the water body characteristic parameters into the turbidity level model for matching, and determine the turbidity level according to the matching result.

[0028] Through the above method, the gray value of each pixel can be corrected based on the lighting factor, thereby improving the turbidity monitoring accuracy.

[0029] Exemplarily, the data sample set is obtained in the following manner: Obtain sample water body images and turbidity levels, process the sample water body images, and obtain sample water body parameters; The sample water parameters and turbidity values ​​are input into a neural network model based on the BP neural network framework for training to obtain a turbidity grade matching model.

[0030] Specifically, first add the sample water image characteristic parameters and turbidity values ​​to the sample library, randomly allocate the sample library so that some samples belong to training samples and other samples belong to test samples, and then train and test the neural network. If qualified, the process ends; if failed, reallocate the sample library for training and testing again. If valid results cannot be obtained within the specified number of times, start the image characteristic parameter extraction module, expand the sample library, and repeat the above operations.

[0031] This embodiment provides a structure of a cleaning module, such as Figure 4 As shown, the cleaning module includes: A water tank 410, which collects oil inlet 430 and outlet; A lifting conveyor belt, the lower part of which is arranged in the water tank; A spray pipe 420, which is arranged above the lower part of the lifting conveyor belt and fixedly installed in the water tank; A filter box 450, the filter box is arranged on one side of the water tank, and the upper part of one side of the filter box is connected to one side of the water tank; A circulation pump 440 is used to connect the filter box to the spray pipe.

[0032] In this embodiment, the information acquisition module may further include a sensor unit (110, the sensor unit includes but is not limited to a temperature sensor, a flow sensor, a liquid level sensor, etc., and the flow, temperature, and liquid level of the water body may be monitored by the sensor unit.

[0033] In addition, the cleaning device may further include a remote monitoring module (300, which can obtain water characteristic parameters, turbidity level matching results and water parameters collected by the sensor unit, and output the above data to a designated display window. In addition, the remote monitoring module can also input instructions to the cleaning control module, so that the cleaning control module generates a control signal corresponding to the instruction, thereby realizing remote control of the cleaning machine. The remote monitoring module includes but is not limited to a mobile terminal, a remote terminal, a computer, etc.

[0034] Embodiment 2: This embodiment provides a product cleaning method based on Embodiment 1. The cleaning method uses the above-mentioned product cleaning device. The cleaning method includes: Receiving water body image information, preprocessing the water body image information, and obtaining water body image preprocessing information; Grayscale gradient co-occurrence matrix is ​​calculated based on the pre-processing information of water body image, water body characteristic parameters are calculated based on the calculation results of grayscale gradient co-occurrence matrix, and turbidity level matching is performed based on the water body characteristic parameters; A water body replacement judgment is made based on the turbidity level matching result, and a water body replacement control signal is generated based on the water body replacement judgment result.

[0035] In this embodiment, the grayscale gradient symbiosis calculation includes: Process the water body image preprocessing information to obtain a grayscale matrix; The Sobel operator is used to perform convolution operation on the gray matrix to obtain the gradient matrix of the water body image preprocessing information; The element values ​​of the grayscale gradient co-occurrence matrix are counted according to the grayscale matrix and the gradient matrix, and the element value statistical results are normalized to obtain the grayscale gradient co-occurrence matrix.

[0036] Specifically, by setting a grayscale level range and a gradient level range, the elements in the grayscale matrix and the gradient matrix are quantized into these levels. For each pixel in the image, its grayscale level and gradient level are considered at the same time, and the frequency of the grayscale level and gradient level appearing at the same time in the specified direction and distance is counted. These frequencies form a two-dimensional matrix, namely the grayscale gradient co-occurrence matrix. The rows of the matrix represent the grayscale level, the columns represent the gradient level, and the element value represents the number of times the corresponding grayscale level and gradient level appear at the same time.

[0037] In this embodiment, the water body characteristic parameters include grayscale energy and gradient energy; The calculation of water body characteristic parameters based on the grayscale gradient co-occurrence matrix calculation results includes: The grayscale energy of the water body image preprocessing information is calculated as follows: ; in, U F is the grayscale energy calculation result; N g is the selected gray level; x , y are the horizontal and vertical coordinates of the pixels in the water body image information; N g is the selected gradient level; h ( x,y ) is the grayscale gradient co-occurrence matrix; The gradient energy of the water body image preprocessing information is calculated as follows: ; Among them, U G Grayscale energy calculation result.

[0038] In this embodiment, before using the Sobel operator to perform convolution operation on the gray matrix, a correction calculation is performed on the gray matrix, and the calculation method is as follows: ; in, F ( x,y ) is the correction calculation result of the gray matrix in the water body image preprocessing information; G min is the maximum grayscale value in the grayscale matrix; G max is the minimum gray value in the gray matrix; d is the control coefficient; s ( x,y ) is the illumination amount of the pixel; h ( x,y ) is the incident light amount.

[0039] In this embodiment, the turbidity level matching based on the water body characteristic parameters is to input the water body characteristic parameters into the turbidity level model for matching, and determine the turbidity level according to the matching result.

[0040] Through the above method, the gray value of each pixel can be corrected based on the lighting factor, thereby improving the turbidity monitoring accuracy.

[0041] Exemplarily, the data sample set is obtained in the following manner: Obtain sample water body images and turbidity levels, process the sample water body images, and obtain sample water body parameters; The sample water parameters and turbidity values ​​are input into a neural network model based on the BP neural network framework for training to obtain a turbidity grade matching model.

[0042] Specifically, first add the sample water image characteristic parameters and turbidity values ​​to the sample library, randomly allocate the sample library so that some samples belong to training samples and other samples belong to test samples, and then train and test the neural network. If qualified, the process ends; if failed, reallocate the sample library for training and testing again. If valid results cannot be obtained within the specified number of times, start the image characteristic parameter extraction module, expand the sample library, and repeat the above operations.

[0043] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A product cleaning device, characterized in that: include: An information acquisition module, the information acquisition module includes an image acquisition unit, the image acquisition unit acquires water body image information; a cleaning monitoring module, the cleaning monitoring module receiving water body image information, preprocessing the water body image information, obtaining water body image preprocessing information, calculating a grayscale gradient co-occurrence matrix according to the water body image preprocessing information, calculating water body characteristic parameters based on the grayscale gradient co-occurrence matrix calculation result, performing turbidity level matching based on the water body characteristic parameters, performing water body replacement judgment on the turbidity level matching result, and generating a water body replacement control signal according to the water body replacement judgment result; The cleaning module obtains a water body replacement control signal and executes an action corresponding to the water body replacement control signal based on the water body replacement control signal.

2. A product cleaning device according to claim 1, characterized in that: The grayscale gradient symbiosis calculation includes: Process the water body image preprocessing information to obtain a grayscale matrix; The Sobel operator is used to perform convolution operation on the gray matrix to obtain the gradient matrix of the water body image preprocessing information; The element values ​​of the grayscale gradient co-occurrence matrix are counted according to the grayscale matrix and the gradient matrix, and the element value statistical results are normalized to obtain the grayscale gradient co-occurrence matrix.

3. A product cleaning device according to claim 1, characterized in that: The water body characteristic parameters include grayscale energy and gradient energy; The calculation of water body characteristic parameters based on the grayscale gradient co-occurrence matrix calculation results includes: The grayscale energy of the water body image preprocessing information is calculated as follows: ； in, U F It is the gray energy calculation result; N g is the selected gray level; x , y are the horizontal and vertical coordinates of the pixels in the water body image information; N g is the selected gradient level; h ( x,y ) is the grayscale gradient co-occurrence matrix; The gradient energy of the water body image preprocessing information is calculated as follows: ； Among them, U G Grayscale energy calculation result.

4. A product cleaning device according to claim 2, characterized in that: Before using the Sobel operator to perform convolution operation on the grayscale matrix, the grayscale matrix is ​​corrected and calculated as follows: ； in, F ( x,y ) is the correction calculation result of the gray matrix in the water body image preprocessing information; G min is the maximum grayscale value in the grayscale matrix; G max is the minimum gray value in the gray matrix; d is the control coefficient; s ( x,y ) is the illumination amount of the pixel; h ( x,y ) is the incident light amount.

5. A product cleaning device according to claim 1, characterized in that: The turbidity level matching based on the water body characteristic parameters is to input the water body characteristic parameters into the turbidity level model for matching, and determine the turbidity level according to the matching result.

6. A product cleaning device according to claim 5, characterized in that: The data sample set is obtained in the following way: Obtain sample water body images and turbidity levels, process the sample water body images, and obtain sample water body parameters; The sample water parameters and turbidity values ​​are input into a neural network model based on the BP neural network framework for training to obtain a turbidity grade matching model.

7. A product cleaning device according to claim 1, characterized in that: The cleaning module comprises: a water tank, which collects the oil inlet and outlet; A lifting conveyor belt, the lower part of which is arranged in the water tank; A spray pipe, which is arranged above the lower part of the lifting conveyor belt and fixedly installed in the water tank; A filter box, wherein the filter box is arranged on one side of the water tank, and an upper portion of one side of the filter box is connected to an upper portion of one side of the water tank; A circulation pump is used to connect the filter box to the spray pipe.

8. A product cleaning method, characterized in that: The cleaning method uses a product cleaning device as described in any one of claims 1 to 7 above, and the cleaning method comprises: Receiving water body image information, preprocessing the water body image information, and obtaining water body image preprocessing information; The grayscale gradient co-occurrence matrix is ​​calculated based on the pre-processing information of the water body image, the water body characteristic parameters are calculated based on the calculation results of the grayscale gradient co-occurrence matrix, and the turbidity level matching is performed based on the water body characteristic parameters; A water body replacement judgment is made based on the turbidity level matching result, and a water body replacement control signal is generated based on the water body replacement judgment result.

9. A product cleaning method according to claim 8, characterized in that: The grayscale gradient symbiosis calculation includes: Process the water body image preprocessing information to obtain a grayscale matrix; The Sobel operator is used to perform convolution operation on the gray matrix to obtain the gradient matrix of the water body image preprocessing information; The element values ​​of the grayscale gradient co-occurrence matrix are counted according to the grayscale matrix and the gradient matrix, and the element value statistical results are normalized to obtain the grayscale gradient co-occurrence matrix.

Images