Quartz sand washing production line and method

Abstract

The present application relates to the field of quartz sand washing, and specifically to a quartz sand washing production line and method thereof. The quartz sand washing production line includes: a preliminary cleaning module, a hydraulic classification module, a screening module and a quartz sand particle size quality detection module. Among them, the quartz sand particle size quality detection module collects the scattered light signal of the quartz sand after washing by a laser particle size analyzer, and uses the signal analysis and processing technology of computer vision to perform signal enhancement and semantic feature extraction on the scattered light signal, so as to automatically judge whether the particle size quality of the quartz sand after washing meets the requirements according to the semantic feature information of the key frame in the scattered light signal. In this way, the quartz sand can be quickly washed and the particle size can be detected to meet the needs of mass production and ensure that the quality of the quartz sand product can meet the high standards of specific industrial applications.

Description

Technical Field

[0001] The present application relates to the field of quartz sand washing, and specifically to a quartz sand washing production line and method thereof. Background Art

[0002] As a common non-metallic mineral raw material, quartz sand has important applications in many industrial fields, including the construction industry, glass manufacturing, ceramic industry and foundry industry. With the advancement of science and technology and the increasing demand for high-tech silicon, the requirements for the quality of quartz sand are getting higher and higher, especially the demand for high-purity quartz sand, which has promoted the development of quartz sand beneficiation and purification technology.

[0003] Water washing is an important step in the quartz sand purification process. It helps to remove impurities in quartz sand, such as clay and fine muddy particles, which will affect the purity of quartz sand and the quality of subsequent applications. In addition, water washing can also be used as a pretreatment method before ore selection to improve purification efficiency.

[0004] In many industrial applications, such as glass manufacturing, electronics, coatings, papermaking, etc., the particle size distribution of quartz sand directly affects the quality of the final product. Therefore, it is essential to accurately detect the particle size quality of washed quartz sand. However, most traditional quartz sand particle size detection methods use manual screening or sedimentation methods, which is not only time-consuming and labor-intensive, but also difficult to meet the needs of rapid detection of large quantities of quartz sand. In addition, most traditional detection methods require manual operation, have a low degree of automation, and are difficult to achieve continuous online detection and data collection.

[0005] Therefore, an optimized quartz sand washing production line is desired, which can use automated detection technology to detect the particle size quality of quartz sand. Summary of the invention

[0006] This application is made in view of the above problems. One object of this application is to provide a quartz sand washing production line and method thereof.

[0007] The embodiment of the present application provides a quartz sand washing production line, which includes: a preliminary cleaning module, which is used to pour raw sand into a sand washing tank for soaking, and adopt a hydraulic feeding method to cooperate with a disc sand washing machine in the sand washing tank for preliminary cleaning to obtain quartz sand after preliminary washing; a hydraulic classification module, which is used to transport the quartz sand after preliminary washing to a first hydraulic classifier for washing, so as to use the rotary vane in the first hydraulic classifier to lift and dehydrate the quartz sand to obtain dehydrated quartz sand; a screening module, which is used to transport the dehydrated quartz sand to a cylindrical roller screen for screening to obtain screened quartz sand, and send the screened quartz sand to a second hydraulic classifier for classification treatment to obtain washed quartz sand; a quartz sand particle size quality detection module, which is used to perform quality detection on the washed quartz sand to determine whether the particle size quality of the washed quartz sand meets the requirements.

[0008] For example, according to the quartz sand washing production line of the embodiment of the present application, the quartz sand particle size quality detection module includes: a scattered light signal acquisition unit for quartz sand after washing, which is used to obtain the scattered light signal of the detected washed quartz sand collected by the laser particle size analyzer; a scattered light signal denoising and enhancement processing unit, which is used to denoise and enhance the scattered light signal to obtain an enhanced scattered light signal; a scattered light signal semantic extraction unit, which is used to pass the enhanced scattered light signal through a scattered light signal semantic feature extractor to obtain a scattered light signal semantic feature map; a scattered light signal semantic feature enhancement unit, which is used to pass the scattered light signal semantic feature map through a scattered light signal key semantic feature enhancer based on a channel pruning optimization mechanism to obtain an enhanced scattered light signal semantic feature map; a quartz sand particle size quality detection unit, which is used to obtain a detection result based on the enhanced scattered light signal semantic feature, and the detection result is used to indicate whether the particle size quality of the detected washed quartz sand meets the requirements.

[0009] For example, according to the quartz sand washing production line of the embodiment of the present application, the scattered light signal denoising and enhancement processing unit includes: a grayscale histogram calculation subunit, used to determine the grayscale histogram of the scattered light signal to obtain a scattered light signal grayscale histogram; a grayscale range confirmation subunit, used to determine the grayscale range of the scattered light signal that needs to be enhanced according to the scattered light signal grayscale histogram; a mapping subunit, used to map the grayscale range of the scattered light signal that needs to be enhanced to a wider grayscale range to obtain a grayscale image of the scattered light signal after mapping; a diffusion process control subunit, used to control the diffusion process according to the local gradient information of the grayscale image of the scattered light signal after mapping; an iterative update subunit, used to iteratively update the pixel values ​​of the grayscale image of the scattered light signal after mapping based on the diffusion process to obtain the enhanced scattered light signal.

[0010] For example, according to the quartz sand washing production line of an embodiment of the present application, the scattered light signal semantic extraction unit is used to: pass the enhanced scattered light signal through a scattered light signal semantic feature extractor based on a deep separable convolutional neural network to obtain the scattered light signal semantic feature map.

[0011] For example, according to the quartz sand washing production line of the embodiment of the present application, the scattered light signal semantic feature enhancement unit includes: a global mean pooling subunit, which is used to perform global mean pooling processing on each feature matrix of the scattered light signal semantic feature map along the channel dimension to obtain a scattered light signal semantic channel feature vector; a scattered light signal semantic feature nonlinear activation processing subunit, which is used to perform nonlinear activation processing on the scattered light signal semantic channel feature vector to obtain a scattered light signal semantic channel weight vector; a mask subunit, which is used to set the eigenvalues ​​greater than or equal to a predetermined threshold in each position of the scattered light signal semantic channel weight vector to the original value, and the rest to zero to obtain a scattered light signal semantic channel mask weight vector; a position point multiplication subunit, which is used to use the eigenvalues ​​at each position in the scattered light signal semantic channel mask weight vector as weight values, and perform position point multiplication on each feature matrix along the channel dimension of the scattered light signal semantic feature map to obtain the enhanced scattered light signal semantic feature map.

[0012] For example, according to the quartz sand washing production line of an embodiment of the present application, the scattered light signal semantic feature nonlinear activation processing subunit is used to: first multiply the scattered light signal semantic channel feature vector with the parameter matrix to obtain a parameter feature vector, and then add the parameter feature vector and the bias vector by position to obtain a scattered light signal semantic channel bias feature vector; input the scattered light signal semantic channel bias feature vector into an activation function for activation processing to obtain the scattered light signal semantic channel weight vector.

[0013] For example, according to the quartz sand washing production line of an embodiment of the present application, the quartz sand particle size quality detection unit is used to: pass the enhanced scattered light signal semantic feature map through a classifier-based quartz sand particle size quality detector to obtain a detection result, and the detection result is used to indicate whether the particle size quality of the quartz sand after the detected water washing meets the requirements.

[0014] For example, according to the quartz sand washing production line of an embodiment of the present application, the quartz sand particle size quality detection unit includes: an optimization subunit, used to optimize the enhanced scattered light signal semantic feature map to obtain an optimized fused enhanced scattered light signal semantic feature vector; a fully connected encoding subunit, used to use the fully connected layer of the classifier-based quartz sand particle size quality detector to fully connect encode the optimized fused enhanced scattered light signal semantic feature vector to obtain an encoded classification feature vector; a classification subunit, used to input the encoded classification feature vector into the Softmax classification function of the classifier-based quartz sand particle size quality detector to obtain the detection result.

[0015] The embodiment of the present application also provides a quartz sand washing method, which includes: pouring raw sand into a sand washing tank for soaking, and using hydraulic feeding to cooperate with a disc sand washing machine in the sand washing tank for preliminary cleaning to obtain quartz sand after preliminary washing; conveying the quartz sand after preliminary washing to a first hydraulic classifier for washing, and using the rotor in the first hydraulic classifier to lift and dehydrate the quartz sand to obtain dehydrated quartz sand; conveying the dehydrated quartz sand to a cylindrical roller screen for screening to obtain screened quartz sand, and conveying the screened quartz sand to a second hydraulic classifier for grading treatment to obtain washed quartz sand; and performing quality inspection on the washed quartz sand to determine whether the particle size quality of the washed quartz sand meets the requirements.

[0016] For example, according to the quartz sand washing method of the embodiment of the present application, the quality of the washed quartz sand is inspected to determine whether the particle size quality of the washed quartz sand meets the requirements, including: obtaining the scattered light signal of the washed quartz sand collected by a laser particle size analyzer; denoising and enhancing the scattered light signal to obtain an enhanced scattered light signal; passing the enhanced scattered light signal through a scattered light signal semantic feature extractor to obtain a scattered light signal semantic feature map; passing the scattered light signal semantic feature map through a scattered light signal key semantic feature enhancer based on a channel pruning optimization mechanism to obtain an enhanced scattered light signal semantic feature map; based on the enhanced scattered light signal semantic feature, obtaining a detection result, the detection result is used to indicate whether the particle size quality of the washed quartz sand meets the requirements.

[0017] According to the quartz sand washing production line and method of the embodiments of the present application, quartz sand can be quickly washed and the particle size detected to meet the needs of mass production and ensure that the quality of quartz sand products can meet the high standards of specific industrial applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0019] Figure 1 A schematic diagram of the application architecture of a quartz sand washing production line in an embodiment of the present application is shown.

[0020] Figure 2 A schematic structural diagram of a quartz sand washing production line in an embodiment of the present application is shown.

[0021] Figure 3 A structural schematic diagram of a quartz sand particle size quality detection module of a quartz sand washing production line in an embodiment of the present application is shown.

[0022] Figure 4 A flow chart of a quartz sand washing method in an embodiment of the present application is shown.

[0023] Figure 5 The application scenario diagram of the quartz sand washing production line in the embodiment of the present application is shown. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0025] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intentions of those skilled in the art, precedents, or new technologies in the art. In addition, specific terms may be selected, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0026] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0027] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0028] Figure 1 A schematic diagram of the application architecture of a quartz sand washing production line in an embodiment of the present application is shown, including a server 100 and a terminal device 200.

[0029] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a private network or a virtual private network. In some embodiments, the data exchanged through the network can be represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec), etc. can be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.

[0030] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.

[0031] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In the embodiment of the present application, the memory 120 may be used to store the corresponding method of the quartz sand washing production line in the embodiment of the present application.

[0032] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of any one of the quartz sand washing methods in the embodiments of the present application according to the obtained program instructions.

[0033] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0034] The following is a non-restrictive description of the quartz sand washing production line provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present application.

[0035] Based on this, this application proposes a quartz sand washing production line, such as Figure 2As shown, it includes: a preliminary cleaning module 810, which is used to pour the raw sand into the sand washing tank for soaking, and adopt a hydraulic feeding method to cooperate with the disc sand washing machine in the sand washing tank for preliminary cleaning to obtain quartz sand after preliminary water washing; a hydraulic classification module 820, which is used to transport the quartz sand after preliminary water washing to the first hydraulic classifier for washing, so as to use the rotary vane in the first hydraulic classifier to lift and dehydrate the quartz sand to obtain dehydrated quartz sand; a screening module 830, which is used to transport the dehydrated quartz sand to the cylindrical roller screen for screening to obtain screened quartz sand, and send the screened quartz sand to the second hydraulic classifier for classification treatment to obtain washed quartz sand; a quartz sand particle size quality detection module 840, which is used to perform quality detection on the washed quartz sand to determine whether the particle size quality of the washed quartz sand meets the requirements.

[0036] Correspondingly, in the above-mentioned quartz sand particle size quality detection module 840, the scattered light signal of the washed quartz sand to be detected is collected by a laser particle size analyzer, and the signal analysis and processing technology of computer vision is used to perform signal enhancement and semantic feature extraction on the scattered light signal, so as to automatically judge whether the particle size quality of the washed quartz sand to be detected meets the requirements according to the semantic feature information of the key frame in the scattered light signal. In this way, the scattered light signal can be collected in real time by using a laser particle size analyzer, and the automatic analysis and detection of the quartz sand particle size can be combined with computer vision technology, which reduces the errors caused by human factors, thereby improving the intelligent level of quartz sand quality detection and water washing process. At the same time, the automated production line can quickly wash and detect the particle size of quartz sand, meet the needs of mass production, and ensure that the quality of quartz sand products can meet the high standards of specific industrial applications.

[0037] Accordingly, if Figure 3 As shown, the quartz sand particle size quality detection module 840 includes: a quartz sand scattered light signal acquisition unit 841 after washing, which is used to obtain the scattered light signal of the washed quartz sand to be detected collected by the laser particle size analyzer; a scattered light signal denoising and enhancement processing unit 842, which is used to denoise and enhance the scattered light signal to obtain an enhanced scattered light signal; a scattered light signal semantic extraction unit 843, which is used to pass the enhanced scattered light signal through a scattered light signal semantic feature extractor to obtain a scattered light signal semantic feature map; a scattered light signal semantic feature enhancement unit 844, which is used to pass the scattered light signal semantic feature map through a scattered light signal key semantic feature enhancer based on a channel pruning optimization mechanism to obtain an enhanced scattered light signal semantic feature map; a quartz sand particle size quality detection unit 845, which is used to obtain a detection result based on the enhanced scattered light signal semantic feature, and the detection result is used to indicate whether the particle size quality of the washed quartz sand to be detected meets the requirements.

[0038] Specifically, in the technical solution of the present application, first, the scattered light signal of the washed quartz sand detected by the laser particle size analyzer is obtained. Next, considering that the scattered light signal is emitted by the laser particle size analyzer to irradiate the washed quartz sand with a laser beam, these signals are related to the particle size distribution of the quartz sand. In detail, there may be complex textures and noise on the surface of the quartz sand, and these factors will affect the quality of the scattered light signal. And due to the color and surface characteristics of the quartz sand, the scattered light signal may have a low contrast, making the details difficult to discern. In addition, the grayscale distribution of the original quartz sand image may be relatively concentrated, resulting in quartz sand of different particle sizes being difficult to distinguish in the image. Based on this, in order to improve the key information in the scattered light signal and reduce the interference of background and noise, in the technical solution of the present application, the scattered light signal is denoised and enhanced to obtain an enhanced scattered light signal. That is, the contrast of the scattered light signal is first improved by using a piecewise linear transformation method, and then anisotropic diffusion filtering is used to weaken the texture noise in the background of the scattered light signal. Specifically, the piecewise linear transformation first determines the grayscale histogram in the scattered light signal, and determines the grayscale range (low contrast part) that needs to be enhanced based on the grayscale histogram. Then, the piecewise linear transformation function is used to map the selected grayscale range to a grayscale range wider than that in the original signal to improve the contrast dynamic range of the scattered light signal. Next, a diffusion function is defined to retain the edge information of the scattered light signal based on the local gradient information in the grayscale image. Finally, the value of each pixel is iteratively updated to ensure that the background noise is effectively suppressed, while important features (such as the quartz sand grain boundary) are retained to obtain an enhanced scattered light signal.

[0039] Correspondingly, the scattered light signal denoising and enhancement processing unit 842 includes: a grayscale histogram calculation subunit, used to determine the grayscale histogram of the scattered light signal to obtain a scattered light signal grayscale histogram; a grayscale range confirmation subunit, used to determine the grayscale range of the scattered light signal that needs to be enhanced according to the scattered light signal grayscale histogram; a mapping subunit, used to map the grayscale range of the scattered light signal that needs to be enhanced to a wider grayscale range to obtain a grayscale image of the scattered light signal after mapping; a diffusion process control subunit, used to control the diffusion process according to the local gradient information of the grayscale image of the scattered light signal after mapping; an iterative update subunit, used to iteratively update the pixel values ​​of the grayscale image of the scattered light signal after mapping based on the diffusion process to obtain the enhanced scattered light signal.

[0040] Then, considering that the enhanced scattered light signal contains semantic information about the particle size and distribution of the quartz sand material, these semantic information plays an important role in the subsequent judgment of the quartz sand particle quality. The traditional standard convolution layer usually contains a large number of parameters, because each convolution kernel is independently learned, including the weight and bias of the convolution kernel, and the overall computational complexity is high. Based on this, in order to reduce the amount of calculation in the feature processing process and speed up the detection speed, in the technical solution of the present application, the enhanced scattered light signal is passed through a scattered light signal semantic feature extractor based on a deep separable convolutional neural network to obtain a scattered light signal semantic feature map. In particular, the deep separable convolution layer includes a 3x3 deep convolution kernel, batch normalization, a 1x1 point-by-point convolution kernel and a nonlinear activation function (ReLU6), wherein ReLU6 is obtained by limiting the maximum output of the ReLU nonlinear activation function to 6, and its main function is to allow the network to learn sparse features earlier and prevent numerical explosion, so that the semantic features of the scattered light signal can be effectively extracted while significantly reducing the amount of calculation.

[0041] Correspondingly, the scattered light signal semantic extraction unit 843 is used to: pass the enhanced scattered light signal through a scattered light signal semantic feature extractor based on a deep separable convolutional neural network to obtain the scattered light signal semantic feature map.

[0042] Furthermore, it is considered that the semantic feature information of the particle size characteristics such as the size and shape of the quartz sand particles is expressed in the semantic feature map of the scattered light signal, and these semantic information are the key information in the analysis of the particle size quality of the quartz sand. It also contains some background redundant feature information or irrelevant information that is not related to the particle size of the quartz sand. Therefore, in the technical solution of the present application, the semantic feature map of the scattered light signal is passed through a scattered light signal key semantic feature enhancer based on a channel pruning optimization mechanism to obtain an enhanced semantic feature map of the scattered light signal. In other words, by optimizing the channel pruning of the semantic feature map of the scattered light signal, it is possible to remove redundant channels in the deep neural network that have little impact on the quality assessment of the quartz sand particle size, reduce the number of parameters and the computational complexity of the model, and achieve lightweight model. At the same time, the key semantic feature information of the scattered light signal is retained. Specifically, the key semantic feature enhancer of the scattered light signal based on the channel pruning optimization mechanism performs global mean pooling on the semantic feature map of the scattered light signal to reduce the dimension of the entire feature map, while retaining the key semantic features in the semantic feature map of the scattered light signal to obtain the semantic channel feature vector of the scattered light signal. Next, the channel feature vector is nonlinearly activated and masked to filter out scattered light channel information below a predetermined threshold. Finally, the vector obtained after masking is used to weight the scattered light signal semantic feature map so that useful key information is retained and enhanced, thereby obtaining a more representative enhanced scattered light signal semantic feature map.

[0043] Correspondingly, the scattered light signal semantic feature enhancement unit 844 includes: a global mean pooling subunit, which is used to perform global mean pooling processing on each feature matrix of the scattered light signal semantic feature map along the channel dimension to obtain a scattered light signal semantic channel feature vector; a scattered light signal semantic feature nonlinear activation processing subunit, which is used to perform nonlinear activation processing on the scattered light signal semantic channel feature vector to obtain a scattered light signal semantic channel weight vector; a mask subunit, which is used to set the eigenvalues ​​greater than or equal to a predetermined threshold in each position of the scattered light signal semantic channel weight vector to the original value, and set the rest to zero to obtain a scattered light signal semantic channel mask weight vector; a position point multiplication subunit, which is used to use the eigenvalues ​​at each position in the scattered light signal semantic channel mask weight vector as weight values, and perform position point multiplication on each feature matrix along the channel dimension of the scattered light signal semantic feature map to obtain the enhanced scattered light signal semantic feature map.

[0044] Among them, the scattered light signal semantic feature nonlinear activation processing subunit is used to: first multiply the scattered light signal semantic channel feature vector with the parameter matrix to obtain a parameter feature vector, and then add the parameter feature vector and the bias vector by position to obtain the scattered light signal semantic channel bias feature vector; input the scattered light signal semantic channel bias feature vector into the activation function for activation processing to obtain the scattered light signal semantic channel weight vector.

[0045] In a specific example, the scattered light signal semantic feature enhancement unit 844 is used to: perform a scattered light signal key semantic feature enhancer based on a channel pruning optimization mechanism on the scattered light signal semantic feature map, and process it with the following channel pruning optimization formula to obtain the enhanced scattered light signal semantic feature map; wherein the channel pruning optimization formula is: ;in, is the semantic feature map of the scattered light signal, It is to perform global mean pooling on each feature matrix along the channel dimension of the feature map. is the semantic channel feature vector of the scattered light signal, is the parameter matrix, is the bias vector, is the activation function, is the semantic channel weight vector of the scattered light signal, For masking operation, is the semantic channel mask weight vector of the scattered light signal, It is the point multiplication by position. It is the semantic feature map of the enhanced scattered light signal.

[0046] Then, the enhanced scattered light signal semantic feature map is passed through a quartz sand particle size quality detector based on a classifier to obtain a test result, and the test result is used to indicate whether the particle size quality of the quartz sand after water washing meets the requirements. That is, the enhanced scattered light signal semantic features obtained after pruning and optimization of the scattered light signal semantic feature map are classified and processed to automatically determine whether the particle size quality of the quartz sand after water washing meets the requirements. In this way, a laser particle size analyzer can be used to collect scattered light signals in real time, and combined with computer vision technology, the particle size of quartz sand can be automatically analyzed and detected, thereby reducing errors caused by human factors and improving the accuracy of the test results. At the same time, the automated production line can quickly wash and detect the particle size of quartz sand, meeting the needs of mass production and ensuring that the quality of quartz sand products meets the high standards of specific industrial applications.

[0047] Correspondingly, the quartz sand particle size quality detection unit 845 is used to: pass the enhanced scattered light signal semantic feature map through a quartz sand particle size quality detector based on a classifier to obtain a detection result, and the detection result is used to indicate whether the particle size quality of the quartz sand after being tested and washed meets the requirements.

[0048] Among them, the quartz sand particle size quality detection unit 845 includes: an optimization subunit, used to optimize the enhanced scattered light signal semantic feature map to obtain an optimized fused enhanced scattered light signal semantic feature vector; a fully connected encoding subunit, used to use the fully connected layer of the classifier-based quartz sand particle size quality detector to fully connect encode the optimized fused enhanced scattered light signal semantic feature vector to obtain an encoded classification feature vector; a classification subunit, used to input the encoded classification feature vector into the Softmax classification function of the classifier-based quartz sand particle size quality detector to obtain the detection result.

[0049] It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules, and then classify (or predict) unknown data. Logistic regression, SVM, etc. are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification, but this is prone to errors and is not efficient. Commonly used multi-classification methods include the Softmax classification function.

[0050] Here, the present application takes into account that the scattered light signal semantic feature map expresses the image semantic features of the scattered light signal. After the spatial distribution of the local channel image semantic features is enhanced based on the channel pruning optimization mechanism along the channel dimension, the enhanced scattered light signal semantic feature map obtained will also have an expression deviation relative to the source image feature semantics of the scattered light signal, thereby causing category deviation.

[0051] Therefore, the present application hopes to optimize the enhanced scattered light signal semantic feature map by fusing the scattered light signal semantic feature map and the enhanced scattered light signal semantic feature map to compensate for the category shift caused by the imbalance of feature correspondence ratio under the fine-grained distribution of image semantics. At the same time, considering that the difference in image semantic feature distribution between the scattered light signal semantic feature map and the enhanced scattered light signal semantic feature map will affect the overall expression effect of the enhanced scattered light signal semantic feature map.

[0052] In a preferred embodiment, the enhanced scattered light signal semantic feature map is optimized to obtain an optimized fused enhanced scattered light signal semantic feature vector, comprising the following steps: weighting the scattered light signal semantic feature map and the enhanced scattered light signal semantic feature map to obtain a fused enhanced scattered light signal semantic feature map, and expanding the fused enhanced scattered light signal semantic feature map into a fused enhanced scattered light signal semantic feature vector, calculating the mean and variance of all eigenvalues ​​of the fused enhanced scattered light signal semantic feature vector, adding the fused enhanced scattered light signal semantic feature vector to the mean and the variance respectively, taking the inverse of each eigenvalue and further adding it to the fused enhanced scattered light signal semantic feature vector. The fused enhanced scattered light signal semantic feature vector is point multiplied to obtain a first intermediate feature vector and a second intermediate feature vector. Then, the fused enhanced scattered light signal semantic feature vector is transposed to a column vector and then matrix multiplied with the first intermediate feature vector as a row vector to obtain a first intermediate feature matrix. The second intermediate feature vector is transposed to a column vector and then matrix multiplied with the fused enhanced scattered light signal semantic feature vector to obtain a second intermediate feature matrix. Finally, the first intermediate feature matrix, the transpose of the second intermediate feature matrix and the transpose of the fused enhanced scattered light signal semantic feature vector are matrix multiplied to obtain an optimized fused enhanced scattered light signal semantic feature vector.

[0053] The semantic feature vector of the fused enhanced scattered light signal is expressed as , then the optimization process is expressed as: ;in, represents the semantic feature vector of the fused enhanced scattered light signal, and represent the mean and variance of all eigenvalues ​​of the semantic feature vector of the fused enhanced scattered light signal, respectively, It means adding by position. It means point multiplication by position. It means to calculate the inverse of each eigenvalue of the vector. and are the first intermediate eigenvector and the second intermediate eigenvector respectively, represents matrix multiplication, represents the transpose of a vector, Represents the optimized fusion enhanced scattered light signal semantic feature vector.

[0054] That is, in view of the collapse of the tail feature distribution caused by the loss of tail feature information in the synthetic feature distribution ecology of the fused enhanced scattered light signal semantic feature vector, the descriptive ecological discovery of the feature set of the fused enhanced scattered light signal semantic feature vector is realized by the stable diffusion of the distribution probability of the global statistical characteristics of the fused enhanced scattered light signal semantic feature vector, so as to form a hybrid intuitive feature ecology through the variational self-encoding based on the stable diffusion probability distribution, so as to improve the universality of the feature real interactive value acquisition of the fused enhanced scattered light signal semantic feature vector as a synthetic feature, improve the overall expression effect of the fused enhanced scattered light signal semantic feature vector, and improve the accuracy of the detection results obtained by the fused enhanced scattered light signal semantic feature vector through the quartz sand particle size quality detector based on the classifier. In this way, the scattered light signal can be collected in real time by using a laser particle size analyzer, and the automatic analysis and detection of the quartz sand particle size can be combined with computer vision technology, which reduces the error caused by human factors, thereby improving the accuracy of the detection results. At the same time, the automated production line can quickly wash and detect the particle size of quartz sand, meet the needs of mass production, and ensure that the quality of quartz sand products meets the high standards of specific industrial applications.

[0055] Based on the above embodiments, see Figure 4 , which is a flow chart of a quartz sand washing method in an embodiment of the present application. For example, the quartz sand washing method can be executed by a server, which can be Figure 1 The server 100 shown in FIG. Figure 4As shown, the quartz sand washing method according to the embodiment of the present application includes the following steps: S510, pouring the raw sand into a sand washing tank for soaking, and using hydraulic feeding to cooperate with the disc sand washing machine in the sand washing tank for preliminary cleaning to obtain quartz sand after preliminary washing; S520, conveying the quartz sand after preliminary washing to a first hydraulic classifier for washing, so as to use the rotary vane in the first hydraulic classifier to lift and dehydrate the quartz sand to obtain dehydrated quartz sand; S530, conveying the dehydrated quartz sand to a cylindrical roller screen for screening to obtain screened quartz sand, and sending the screened quartz sand to a second hydraulic classifier for grading treatment to obtain washed quartz sand; S540, performing quality inspection on the washed quartz sand to determine whether the particle size quality of the washed quartz sand meets the requirements.

[0056] In a specific example, in step S540, the quality of the washed quartz sand is inspected to determine whether the particle size quality of the washed quartz sand meets the requirements, including: obtaining a scattered light signal of the washed quartz sand collected by a laser particle size analyzer; denoising and enhancing the scattered light signal to obtain an enhanced scattered light signal; passing the enhanced scattered light signal through a scattered light signal semantic feature extractor to obtain a scattered light signal semantic feature map; passing the scattered light signal semantic feature map through a scattered light signal key semantic feature enhancer based on a channel pruning optimization mechanism to obtain an enhanced scattered light signal semantic feature map; and obtaining a detection result based on the enhanced scattered light signal semantic feature, wherein the detection result is used to indicate whether the particle size quality of the washed quartz sand meets the requirements.

[0057] Here, those skilled in the art will appreciate that the specific operations of each step in the above-mentioned quartz sand washing method have been described in detail above. Figure 2 to Figure 3 The quartz sand washing production line 800 is described in detail, and therefore, its repeated description will be omitted.

[0058] Figure 5 FIG. 1 is an application scenario diagram of a quartz sand washing production line according to an embodiment of the present application. Figure 5 As shown, in this application scenario, first, a laser particle size analyzer (e.g., Figure 5 The scattered light signal of the washed quartz sand (e.g., Figure 5 D), and then, the scattered light signal is input to a server (for example, Figure 5 In S) as shown in , the server can use the quartz sand washing algorithm to process the scattered light signal to obtain a detection result indicating whether the particle size quality of the detected quartz sand after washing meets the requirements.

[0059] Based on the above embodiment, another exemplary embodiment of an electronic device is also provided in the embodiments of the present application. In some possible implementations, the electronic device in the embodiments of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the steps of the quartz sand washing method in the above embodiment can be implemented when the processor executes the program.

[0060] For example, in the case of electronic equipment Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.

[0061] The embodiment of the present application also provides a computer-readable storage medium, on which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, the quartz sand washing method according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may include, for example, a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may include, for example, a read-only memory (ROM), a hard disk, a flash memory, etc.

[0062] The embodiment of the present application also provides a computer program product or a computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer readable storage medium. The processor of the computer device reads the computer executable instructions from the computer readable storage medium, and the processor executes the computer executable instructions, so that the computer device executes the quartz sand washing method according to the embodiment of the present application.

[0063] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various variations and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0064] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.

[0065] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0066] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0067] The above is an explanation of the present application and should not be considered as limiting thereof. Although several exemplary embodiments of the present application have been described, those skilled in the art will readily appreciate that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application.

Claims

1. A quartz sand washing production line, characterized in that: include: The preliminary cleaning module is used to pour the raw sand into the sand washing tank for soaking, and adopts a hydraulic feeding method to cooperate with the disc sand washing machine in the sand washing tank for preliminary cleaning to obtain quartz sand after preliminary water washing; A hydraulic classification module, used for conveying the quartz sand after the preliminary water washing to the first hydraulic classifier for washing, so as to utilize the rotary vanes in the first hydraulic classifier to lift and dehydrate the quartz sand to obtain dehydrated quartz sand; A screening module, used for conveying the dehydrated quartz sand to a drum screen for screening to obtain screened quartz sand, and conveying the screened quartz sand to a second hydraulic classifier for classification to obtain washed quartz sand; A quartz sand particle size quality detection module is used to perform quality detection on the washed quartz sand to determine whether the particle size quality of the washed quartz sand meets the requirements; Wherein, the quartz sand particle size quality detection module includes: The washed quartz sand scattered light signal acquisition unit is used to acquire the scattered light signal of the washed quartz sand collected by the laser particle size analyzer; A scattered light signal denoising and enhancement processing unit, used for performing denoising and enhancement processing on the scattered light signal to obtain an enhanced scattered light signal; A scattered light signal semantic extraction unit, used for passing the enhanced scattered light signal through a scattered light signal semantic feature extractor to obtain a scattered light signal semantic feature map; A scattered light signal semantic feature enhancement unit, used for passing the scattered light signal semantic feature map through a scattered light signal key semantic feature enhancer based on a channel pruning optimization mechanism to obtain an enhanced scattered light signal semantic feature map; A quartz sand particle size quality detection unit, used to obtain a detection result based on the semantic features of the enhanced scattered light signal, wherein the detection result is used to indicate whether the particle size quality of the detected quartz sand after washing meets the requirements; Wherein, the scattered light signal denoising and enhancement processing unit comprises: A grayscale histogram calculation subunit, used to determine the grayscale histogram of the scattered light signal to obtain a scattered light signal grayscale histogram; A grayscale range confirmation subunit, used to determine the grayscale range of the scattered light signal that needs to be enhanced according to the grayscale histogram of the scattered light signal; A mapping subunit, used for mapping the grayscale range of the scattered light signal that needs to be enhanced to a wider grayscale range, so as to obtain a grayscale image of the scattered light signal after mapping; A diffusion process control subunit, used for controlling the diffusion process according to the local gradient information of the grayscale image of the scattered light signal after the mapping; The iterative updating subunit is used to iteratively update the pixel values ​​of the grayscale image of the mapped scattered light signal step by step based on the diffusion process to obtain the enhanced scattered light signal.

2. The quartz sand washing production line according to claim 1 is characterized in that: The scattered light signal semantic extraction unit is used to: The enhanced scattered light signal is passed through a scattered light signal semantic feature extractor based on a deep separable convolutional neural network to obtain the scattered light signal semantic feature map.

3. The quartz sand washing production line according to claim 2 is characterized in that: The scattered light signal semantic feature enhancement unit comprises: A global mean pooling subunit, used for performing global mean pooling processing on each feature matrix of the scattered light signal semantic feature map along the channel dimension to obtain a scattered light signal semantic channel feature vector; A scattered light signal semantic feature nonlinear activation processing subunit, used for performing nonlinear activation processing on the scattered light signal semantic channel feature vector to obtain a scattered light signal semantic channel weight vector; A mask subunit, used for setting the feature values ​​greater than or equal to a predetermined threshold in each position of the scattered light signal semantic channel weight vector to the original value, and setting the rest to zero to obtain a scattered light signal semantic channel mask weight vector; The position point multiplication subunit is used to use each position feature value in the scattered light signal semantic channel mask weight vector as a weight value, and perform position point multiplication on each feature matrix along the channel dimension of the scattered light signal semantic feature map to obtain the enhanced scattered light signal semantic feature map.

4. The quartz sand washing production line according to claim 3 is characterized in that: The scattered light signal semantic feature nonlinear activation processing subunit is used to: The scattered light signal semantic channel feature vector is first multiplied by a parameter matrix to obtain a parameter feature vector, and then the parameter feature vector is added to a bias vector by position to obtain a scattered light signal semantic channel bias feature vector; The scattered light signal semantic channel bias feature vector is input into an activation function for activation processing to obtain the scattered light signal semantic channel weight vector.

5. The quartz sand washing production line according to claim 4 is characterized in that: The quartz sand particle size quality detection unit is used for: The enhanced scattered light signal semantic feature map is passed through a quartz sand particle size quality detector based on a classifier to obtain a detection result, and the detection result is used to indicate whether the particle size quality of the detected washed quartz sand meets the requirements.

6. The quartz sand washing production line according to claim 5, characterized in that: The quartz sand particle size quality detection unit comprises: An optimization subunit, used for optimizing the enhanced scattered light signal semantic feature map to obtain an optimized fused enhanced scattered light signal semantic feature vector; A fully connected encoding subunit, used for performing fully connected encoding on the optimized fused enhanced scattered light signal semantic feature vector using the fully connected layer of the classifier-based quartz sand particle size quality detector to obtain an encoded classification feature vector; The classification subunit is used to input the encoded classification feature vector into the Softmax classification function of the classifier-based quartz sand particle size quality detector to obtain the detection result.

7. A method for washing quartz sand, characterized in that: include: The raw sand is poured into the sand washing tank for soaking, and the raw sand is preliminarily washed by the disc sand washing machine in the sand washing tank in a hydraulic feeding manner to obtain preliminarily washed quartz sand; The quartz sand after the preliminary water washing is transported to the first hydraulic classifier for washing, so as to use the rotary vanes in the first hydraulic classifier to lift and dehydrate the quartz sand to obtain dehydrated quartz sand; The dehydrated quartz sand is conveyed to a drum screen for screening to obtain screened quartz sand, and the screened quartz sand is conveyed to a second hydraulic classifier for classification to obtain washed quartz sand; Performing a quality test on the washed quartz sand to determine whether the particle size quality of the washed quartz sand meets the requirements; The washed quartz sand is subjected to quality inspection to determine whether the particle size quality of the washed quartz sand meets the requirements, including: Obtaining scattered light signals of the washed quartz sand collected by a laser particle size analyzer; Performing denoising and enhancement processing on the scattered light signal to obtain an enhanced scattered light signal; Passing the enhanced scattered light signal through a scattered light signal semantic feature extractor to obtain a scattered light signal semantic feature map; The scattered light signal semantic feature map is passed through a scattered light signal key semantic feature enhancer based on a channel pruning optimization mechanism to obtain an enhanced scattered light signal semantic feature map; Based on the semantic features of the enhanced scattered light signal, a detection result is obtained, wherein the detection result is used to indicate whether the particle size quality of the detected washed quartz sand meets the requirements; The step of performing denoising and enhancement processing on the scattered light signal to obtain an enhanced scattered light signal comprises: Determining a grayscale histogram of the scattered light signal to obtain a scattered light signal grayscale histogram; Determining a grayscale range of the scattered light signal that needs to be enhanced according to the grayscale histogram of the scattered light signal; Mapping the grayscale range of the scattered light signal that needs to be enhanced to a wider grayscale range to obtain a grayscale image of the scattered light signal after mapping; Controlling the diffusion process according to the local gradient information of the grayscale image of the scattered light signal after the mapping; Based on the diffusion process, the pixel values ​​of the grayscale image of the mapped scattered light signal are updated step by step and iteratively to obtain the enhanced scattered light signal.

Images