Intelligent identification and separation system, method and device for coal series kaolinite in coal gangue

Abstract

The invention provides an intelligent recognition and separation system, method and device for coal series kaolinite in coal gangue, and belongs to the field of coal gangue separation. The problems that in an existing coal gangue separation technology, coal series kaolinite recognition and separation efficiency is low, and precision is poor are solved. The system comprises a multi-modal data acquisition and processing module which is used for acquiring and processing a ray image, a texture image and the thickness of the coal gangue to obtain a comprehensive feature vector containing the gray scale, the texture and the thickness data of the coal gangue; the coal series kaolinite evaluation label establishing module is used for establishing an evaluation label; the image input classification algorithm module is used for taking the comprehensive feature vector as input, taking the evaluation label as a classification label, training a classification identification model and outputting a classification result; the control and separation execution module is used for generating a separation instruction according to the classification result and separating coal series kaolinite from the common coal gangue; the method is applied to separation of coal series kaolinite in coal gangue.

Description

Technical Field

[0001] The present application relates to the technical field of coal gangue sorting, and in particular to an intelligent identification and sorting system, method and device for coal-based kaolinite in coal gangue. Background Art

[0002] Gangue is solid waste generated during coal mining and washing. Its large accumulation not only occupies valuable land resources, but also causes serious pollution to the surrounding environment. With increasingly stringent environmental protection requirements and the urgent need for sustainable resource utilization, the effective treatment and resource utilization of gangue has become a key issue that needs to be urgently addressed in the coal industry.

[0003] In fact, as an important component of coal gangue, coal-bearing kaolinite has extremely high economic value. It is widely used in many fields such as ceramics, papermaking, rubber, plastics, etc. However, the current resource utilization of coal gangue is inefficient in industry, so it is very important to invent an intelligent identification and sorting system for coal-bearing kaolinite in coal gangue.

[0004] Existing intelligent sorting technologies mostly focus on the separation of coal and gangue, but there is a gap in the field of fine classification identification and sorting of coal-bearing kaolinite. In related research, it is difficult to accurately identify coal-bearing kaolinite by relying on a single technical means. For example, although X-ray technology can obtain internal structure information, it is not enough to capture surface texture features; industrial cameras can capture surface textures but cannot penetrate deep into the interior; laser radar can measure thickness but it is difficult to identify mineral types. Summary of the invention

[0005] In order to solve the problems of low efficiency, poor accuracy and insufficient intelligence in the identification and sorting of coal-bearing kaolinite in the existing coal gangue sorting technology, the present application proposes an intelligent identification and sorting system, method and device for coal-bearing kaolinite in coal gangue. Through multimodal data fusion technology, combined with X-ray image grayscale data, industrial camera texture information and lidar thickness data, and using advanced image input classification algorithms, intelligent identification and sorting of coal-bearing kaolinite in coal gangue are achieved, sorting efficiency and accuracy are improved, and the process of resource utilization of coal gangue is promoted.

[0006] The technical solution adopted in this application is: an intelligent identification and sorting system for coal-based kaolinite in coal gangue, comprising:

[0007] Multimodal data acquisition and processing module: used to collect the radiographic image, texture image and thickness of coal gangue, and process them to obtain a comprehensive feature vector containing the grayscale, texture and thickness data of coal gangue;

[0008] Coal-bearing kaolinite evaluation label establishment module: used to construct evaluation labels for coal-bearing kaolinite and ordinary coal gangue;

[0009] Image input classification algorithm module: It is used to take the comprehensive feature vector as the input, the evaluation label as the classification label, train the classification recognition model, and output the classification result;

[0010] Control and sorting execution module: It is used to generate a sorting instruction according to the classification result and separate the coal-series kaolinite from the ordinary coal gangue.

[0011] The ray image of the coal gangue is obtained by a dual-energy X-ray source and a ray source detector. The dual-energy X-ray source generates high- and low-energy rays to perform a penetrating scan on the coal gangue. The ray signal detector receives and records the transmitted X-ray signal, generates a digital ray image, which is used to obtain the internal structure information of the coal gangue and the gray-scale data related to the material properties. The collected ray image is processed by a filtering algorithm and an image contrast enhancement algorithm to obtain the gray-scale mean and gray-scale variance of the coal gangue.

[0012] Furthermore, the texture image of the coal gangue is obtained by an industrial camera. The image taken by the industrial camera is subjected to image enhancement and denoising processing, the texture features are extracted, and the texture direction and texture roughness are calculated.

[0013] Furthermore, the thickness data of the coal gangue is obtained by a lidar rangefinder.

[0014] Furthermore, the classification recognition model adopts a multi-layer perceptron. The comprehensive feature vector is input into the multi-layer perceptron. The signal is transmitted from the input layer to the hidden layer. The neurons in the hidden layer perform a weighted sum of the input features and add a bias, and complete the non-linear transformation through the ReLU activation function, and are passed layer by layer to the output layer. The output layer uses the Softmax function to convert the result into a prediction probability; then, the cross-entropy loss function is used to measure the difference between the prediction probability and the true label; then, the backpropagation algorithm is used to reverse-derive the gradients of each weight and bias from the output layer according to the loss value; then, the Adam optimizer is used for optimization, combining momentum gradient descent and RMSProp, adaptively adjusting the learning rate, updating the weights and biases according to the gradients, and after multiple rounds of iteration, until the loss function converges or reaches the preset number of times, the trained classification recognition model is obtained, enabling the model to accurately classify the coal-series kaolinite and ordinary coal gangue for the newly input features.

[0015] Furthermore, a high-pressure air pump and a gas ejection device are adopted in the control and sorting execution module to achieve the precise sorting of the coal-series kaolinite and the ordinary coal gangue

[0016] An intelligent identification and sorting method for coal-series kaolinite in coal gangue, which adopts an intelligent identification and sorting system for coal-series kaolinite in coal gangue, includes the following steps:

[0017] S1: Equipment startup and initialization;

[0018] S2: Collect the dual-energy X-ray images, texture images and coal gangue thickness data of coal gangue, and obtain the comprehensive feature vector after processing through the multi-modal data acquisition and processing module;

[0019] S2: Construct a binary classification evaluation label system;

[0020] S3: Input the comprehensive feature vector as input data into the image input classification algorithm module, and output the classification result;

[0021] S4: Generate a sorting instruction according to the classification result, and send the sorting instruction to the control and sorting execution module;

[0022] S5: The control and sorting execution module realizes the precise sorting of coal-series kaolin according to the sorting instruction.

[0023] A computer device includes a memory, a processor, and a computer program stored on the memory. The processor executes the computer program to implement the steps of the method.

[0024] A computer-readable storage medium stores a computer program / instructions thereon. When the computer program / instructions are executed by a processor, the steps of the method are implemented.

[0025] A computer program product includes a computer program / instructions. When the computer program / instructions are executed by a processor, the steps of the method are implemented.

[0026] The beneficial effects of the present application compared with the prior art are as follows:

[0027] (1) Multi-source data acquisition: Use three devices, namely an industrial camera, dual-energy X-ray, and lidar rangefinder, to collect multi-dimensional data of coal gangue, providing rich information for accurate classification;

[0028] (2) Fusion algorithm classification: Classify the input features through MLP and Adam optimizer to ensure the convergence speed and improve the training rate;

[0029] (3) Precise sorting execution: According to the algorithm result, the high-pressure air pump precisely operates to complete the sorting of coal-series kaolin in coal gangue;

[0030] (4) Strong environmental adaptability: The system combines multiple classification labels and input features, and can adapt to the sorting of coal gangue in different environments. Brief Description of the Drawings

[0031] The following further describes the present application with reference to the accompanying drawings:

[0032] Figure 1 It is a schematic diagram of the system module provided by the embodiment of the present application

[0033] Figure 2 Sorting flow chart provided for the embodiment of the present application

[0034] Figure 3 A schematic diagram of a three-dimensional structure of one of the implementable systems provided in the embodiment of the present application;

[0035] In the figure: 1 is the gangue inlet, 2 is the belt, 3 is the gas push-out device, 4 is the first gangue bin, 5 is the second gangue bin, 6 is the junction box, 7 is the computer workstation, 8 is the laser radar rangefinder, 9 is the industrial camera, 10 is the ray source, 11 is the detector, and 12 is the high-pressure air pump. DETAILED DESCRIPTION

[0036] like Figures 1 to 3 As shown, the present application provides an intelligent identification and sorting system for coal-based kaolinite in coal gangue, including a multimodal data acquisition and processing module, a coal-based kaolinite evaluation label establishment module, an image input classification algorithm module and a control and sorting execution module, wherein the multimodal data acquisition and processing module is used to collect the radiographic image, texture image and thickness of the coal gangue, and process them to obtain the grayscale, texture and thickness data of the coal gangue. The multimodal data acquisition and processing module obtains corresponding data according to a specific multimodal data acquisition device, wherein the radiographic image is obtained by a dual-energy X-ray source and a radiographic signal detector, the dual-energy X-ray source generates high and low energy rays to perform a penetrating scan on the coal gangue, the radiographic signal detector receives and records the transmitted X-ray signal, and generates a digital image for obtaining the internal structure information of the coal gangue and the grayscale data related to the material properties. For the collected radiographic image, a filtering algorithm is used to remove noise, enhance image contrast, extract key grayscale features, and calculate its grayscale mean and grayscale variance. The calculation process is as follows:

[0037]

[0038]

[0039] Where M and N are the number of rows and columns of the X-ray image, that is, the number of pixels in the vertical and horizontal directions of the image, and G lij is the gray value of the pixel in the i-th row and j-th column in the low-energy X-ray image, G hij is the gray value of the pixel in the i-th row and j-th column in the high-energy X-ray image, is the grayscale mean of the low-energy X-ray image, is the grayscale mean of the high-energy X-ray image, is the grayscale variance of the low-energy X-ray image, is the grayscale variance of the high energy X-ray image.

[0040] An industrial camera 9 is used to capture images of the coal gangue surface from multiple angles to obtain rich texture information. The images captured by the industrial camera 9 are subjected to image enhancement and denoising processing, texture features are extracted, and its texture direction and texture roughness are calculated as follows:

[0041]

[0042] Where G x and G y are the gradient values of the image in the x and y directions respectively, θ is the texture direction, L is the number of gray levels. Generally, the number of gray levels of an image is 256, P(i,j) is the element in the i-th row and j-th column of the gray-level co-occurrence matrix, and C is the texture roughness.

[0043] A lidar rangefinder 8 is used to measure the thickness data of the coal gangue in real time to provide accurate thickness parameters for subsequent analysis. The thickness data obtained by the lidar rangefinder 8 is smoothed to remove outliers to ensure the accuracy of the thickness data. The calculation process is as follows:

[0044]

[0045] Where m is the size of the moving average window, that is, the number of measurements participating in the average, k represents the measurement serial number, used to specify the measurement point for which the smoothed thickness is to be calculated currently, i is the index variable in the summation symbol ∑, used to traverse the measurement data within the moving average window, d 1i is the distance from the upper surface of the coal gangue to the radar obtained by the i-th measurement, d 2i is the distance from the lower surface of the coal gangue to the radar obtained by the i-th measurement, h smooth (k) represents the thickness of the coal gangue after smoothing processing at the k-th measurement.

[0046] The processed multi-modal data (gray scale, texture, thickness) is fused to form a comprehensive feature vector F feat :

[0047]

[0048] The coal-series kaolinite evaluation label establishment module adopts a binary classification evaluation label system, marking ordinary coal gangue as "0" and coal-series kaolinite as "1". The definition of the label depends on professional mineral identification methods to ensure the reliability and authority of the classification label, laying a solid foundation for subsequent intelligent recognition and classification work. The label assignment rule is:

[0049]

[0050] Where s i represents the sample set S = {s1, s2, …, s nThe i-th sample in {}. Here, i is the index of the sample, taking values 1, 2, …, n, where n is the total number of samples.

[0051] Image input classification algorithm module: Using a multi-layer perceptron as the classification and recognition model to obtain the comprehensive feature vector F feat After that, input it into a multi-layer perceptron (MLP). The signal is transmitted from the input layer to the hidden layer. The neurons in the hidden layer perform a weighted sum of the input features and add a bias, and complete a non-linear transformation through the ReLU activation function, and are passed layer by layer to the output layer. The output layer uses the Softmax function to convert the result into a classification probability. Then, use the cross-entropy loss function to measure the difference between the predicted probability and the true label. After that, use the backpropagation algorithm to reverse-derive the gradients of each weight and bias from the output layer according to the loss value. Then use the Adam optimizer for optimization, combining the advantages of momentum gradient descent and RMSProp, adaptively adjusting the learning rate, updating the weights and biases according to the gradients, and after multiple rounds of iteration until the loss function converges or reaches the preset number of times, enabling the model to accurately classify coal-measure kaolinite and ordinary coal gangue for new input features.

[0052] The calculation process of the image input classification algorithm module is as follows:

[0053]

[0054] m t = β1m t-1 + (1 - β1)g t ;

[0055]

[0056] In the formula, is the output value of the j-th neuron in the l-th hidden layer. l is the number of hidden layers, l = 1, 2, …, L, where L is the total number of hidden layers. j is the index of the neuron in the current hidden layer, j = 1, 2, …, ml (ml is the number of neurons in the l-th hidden layer). i is the index of the neuron in the previous layer, i = 1, 2, …, n. is the weight connecting the i-th neuron in the previous layer and the j-th neuron in the current layer. x i is the output of the i-th neuron in the previous layer. is the bias term of the j-th neuron in the l-th hidden layer. is the ReLU activation function;

[0057] o k is the input value of the k-th neuron in the output layer. k is the index of the neuron in the output layer, k = 1, 2, …, K, where K is the number of neurons in the output layer. m L is the number of neurons in the last hidden layer. is the weight connecting the j-th neuron in the last hidden layer to the k-th neuron in the output layer, is the output of the j-th neuron in the last hidden layer, is the bias term of the k-th neuron in the output layer;

[0058] yk is the probability value corresponding to the k-th neuron in the output layer after passing through the Softmax function;

[0059] L is the cross-entropy loss value, y true,k is the probability value of the k-th category in the true label, y pred,k is the probability value of the k-th category predicted by the model;

[0060] θ is the parameter (weight w and bias b) in the model, m t is the first moment estimate of the parameter θ at iteration t, v t is the second moment estimate of the parameter θ at iteration t, g t is the gradient of the parameter θ at iteration t, β1 is the hyperparameter controlling the first moment estimate in the Adam optimizer, β2 is the hyperparameter controlling the second moment estimate in the Adam optimizer, is the corrected first moment estimate, is the corrected second moment estimate, α is the learning rate, controlling the step size of each parameter update, ∈ is a very small constant used to avoid division by zero errors, and t is the number of iterations.

[0061] In this embodiment, an advanced and efficient multi-layer perceptron (MLP) architecture is adopted in the image input classification algorithm module to achieve accurate classification of coal-series kaolinite and ordinary coal gangue. After obtaining the comprehensive feature vector F feat it is passed into the input layer. The input layer is like the entrance of information, accurately passing F feat to the hidden layer. The hidden layer is the core processing part of the model. Each neuron will perform a weighted sum on the input and add a bias. Then, through the ReLU activation function for non-linear transformation, after passing through layer by layer, the signal reaches the output layer. The output layer uses the Softmax function to convert the result into classification probabilities, so as to clearly know the possibility of each sample belonging to coal-series kaolinite or ordinary coal gangue. To measure the prediction effect of the model, a cross-entropy loss function is used to calculate the difference between the predicted probability and the true label.

[0062] Subsequently, the backpropagation algorithm is used to reverse-derive the gradients of each weight and bias from the output layer. The Adam optimizer combines the advantages of momentum gradient descent and RMSProp, can adaptively adjust the learning rate, and updates the weights and biases according to the gradients. After multiple rounds of iteration, until the loss function converges or reaches the preset number of iterations, the model can accurately classify the newly input features, providing a reliable basis for the subsequent sorting work.

[0063] Control and sorting execution module: According to the recognition result of the image input classification algorithm module, the control system generates a sorting instruction. The sorting execution device uses a high-pressure air pump 12 as the sorting execution device to accurately separate the identified coal-series kaolinite from other minerals and convey them to the corresponding collection containers respectively.

[0064] Based on the above system, the present application also proposes an intelligent recognition and sorting method for coal-series kaolinite in coal gangue, including the following steps:

[0065] S1: Start and initialize the equipment, and turn on the multi-modal data acquisition device;

[0066] S2: Obtain the dual-energy X-ray image of the coal gangue, the texture image collected by the industrial camera, and the coal gangue thickness data through the multi-modal data acquisition device, and obtain a comprehensive feature vector after processing through the multi-modal data acquisition and processing module;

[0067] S2: Construct a binary classification evaluation label system;

[0068] S3: Input the comprehensive feature vector as input data into the image input classification algorithm module, use the binary classification evaluation label as the classification label, train the classification recognition model through machine learning, and recognize the newly input features through the trained classification recognition model to output the classification result;

[0069] S4: Generate a sorting instruction according to the classification result and send the sorting instruction to the control and sorting execution module;

[0070] S5: The control and sorting execution module controls the action of the sorting device according to the sorting instruction to achieve accurate sorting of coal-series kaolinite.

[0071] Such as Figure 3As shown, the embodiment of the present application provides an implementable intelligent identification and sorting system for coal-based kaolinite in coal gangue, including a transport component, the inlet of the transport component is a gangue feed port 1, and the outlet of the transport component is provided with a first gangue bin 4 and a second gangue bin 5 placed in parallel, wherein the first gangue bin 4 is close to the outlet of the transport component, and a gas push device 3 is also provided at the outlet of the transport component, and the gas push device 3 is connected to the high-pressure air pump 12 through an air path to provide power to the gas push device 3, and the gas push device 3 and the high-pressure air pump 12 constitute a sorting execution device. A laser radar rangefinder 8, an industrial camera 9 and a ray source 10 are arranged above the transport component, and a detector 11 is installed below the transport component, and the detector 11 is arranged opposite to the ray source 10. The laser radar rangefinder 8, industrial camera 9, radiation source 10 and detector 11 constitute a multimodal data acquisition device, in which the laser radar rangefinder 8 is used to measure the thickness of coal gangue in real time; the industrial camera 9 is used to take images of the surface of coal gangue from multiple angles to obtain rich texture information; the radiation source 10 is used to collect radiation images of coal gangue and extract grayscale features.

[0072] A junction box 6 and a computer workstation 7 are also provided on one side of the transport component. The computer workstation 7 is electrically connected to the control end of the transport component, the laser radar rangefinder 8, the industrial camera 9, the radiation source 10, the detector 11 and the high-pressure air pump 12 through the junction box 6.

[0073] In this embodiment, the transport component may be a belt.

[0074] The computer workstation 7 is provided with an industrial computer and a display. The industrial computer is connected to the sorting execution device and the multimodal data acquisition device through the junction box 6. The industrial computer has built-in computer programs of a multimodal data acquisition and processing module, a coal-based kaolinite evaluation label establishment module, an image input classification algorithm module, and a control and sorting execution module.

[0075] When in use, the system is set at the gangue discharge port of the gangue sorting device. The gangue coming out of the gangue sorting device directly falls into the gangue inlet 1 of the gangue conveying device. After being transported by the belt 2, the dual-energy X-ray image, the image taken by the camera and the thickness data of the gangue on the belt 2 are collected in real time. Finally, the gangue and coal-bearing kaolinite are blown into the first gangue bin 4 and the second gangue bin 5 respectively through the gas pushing device 3, thereby realizing the accurate distinction of coal-bearing kaolinite.

[0076] The high-pressure air pump 12 and the gas ejection device 3, according to the instructions of the control system, precisely separate the identified coal-series kaolin from the coal gangue by the difference in gas ejection intensity and fall into the corresponding coal gangue bins. For example, for the coal-series kaolin with a classification label of 1, the gas ejection intensity is controlled to be weak, and it is ejected into a coal gangue bin about 1 meter away. For the ordinary coal gangue with a classification label of 0, the gas ejection intensity is controlled to be strong, and it is ejected into a coal gangue bin about 5 meters away, so as to achieve the precise separation of the coal-series kaolin.

[0077] This application is based on dual-energy X-ray technology, industrial cameras, lidar rangefinders, and machine learning algorithms to achieve intelligent identification and separation of coal gangue for quality-based and classification-based resource utilization, and to achieve high-precision identification and separation of coal-series kaolin. By using a multi-layer perceptron (MLP) combined with a cross-entropy loss function, a backpropagation algorithm, and an Adam optimizer, etc., an efficient and accurate image input classification algorithm is constructed, which can automatically learn and optimize model parameters to achieve precise classification of coal-series kaolin and ordinary coal gangue.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. An intelligent identification and separation system for coal-measure kaolinite in coal gangue, characterized in that: Including: Multi-modal data acquisition and processing module: It is used to acquire the ray image, texture image and thickness of coal gangue, and process them to obtain a comprehensive feature vector containing the gray scale, texture and thickness data of coal gangue; Evaluation label establishment module for coal-series kaolin: It is used to construct evaluation labels for coal-series kaolin and ordinary coal gangue; Image input classification algorithm module: It is used to take the comprehensive feature vector as the input, take the evaluation label as the classification label, train the classification recognition model, and output the classification result; Control and sorting execution module: It is used to generate a sorting instruction according to the classification result and separate the coal-series kaolin from the ordinary coal gangue.

2. The intelligent identification and separation system for coal measure kaolinite in coal gangue according to claim 1, characterized in that: The ray image of coal gangue is obtained through a dual-energy X-ray source and a ray source detector. The dual-energy X-ray source generates high- and low-energy rays to perform a penetrating scan on the coal gangue. The ray signal detector receives and records the transmitted X-ray signal to generate a digital ray image, which is used to obtain the gray scale data related to the internal structure information and material properties of the coal gangue. The collected ray image is processed using a filtering algorithm and an image contrast enhancement algorithm to obtain the gray scale mean and gray scale variance of the coal gangue.

3. The intelligent identification and separation system for coal measure kaolinite in coal gangue according to claim 1, characterized in that: The texture image of coal gangue is obtained through an industrial camera. The image taken by the industrial camera is subjected to image enhancement and denoising processing, texture features are extracted, and the texture direction and texture roughness are calculated.

4. The intelligent identification and separation system for coal measure kaolinite in coal gangue according to claim 1, wherein: The thickness data of coal gangue is obtained through a lidar rangefinder.

5. An intelligent identification and separation system for coal measure kaolinite in coal gangue according to any one of claims 1-4, characterized in that: The classification recognition model adopts a multi-layer perceptron. The comprehensive feature vector is input into the multi-layer perceptron. The signal is transmitted from the input layer to the hidden layer. The neurons in the hidden layer perform weighted summation on the input features and add a bias, and complete the non-linear transformation through the ReLU activation function, and are passed layer by layer to the output layer. The output layer uses the Softmax function to convert the result into a prediction probability; then, the cross-entropy loss function is used to measure the difference between the prediction probability and the true label; then, the backpropagation algorithm is used to reverse-derive the gradients of each weight and bias from the output layer according to the loss value; then, the Adam optimizer is used for optimization, combining momentum gradient descent and RMSProp, adaptively adjusting the learning rate, updating the weights and biases according to the gradients, and after multiple rounds of iteration, until the loss function converges or reaches the preset number of times, to obtain the trained classification recognition model.

6. The intelligent identification and separation system for coal measure kaolinite in coal gangue according to claim 1, wherein: In the control and sorting execution module, a high-pressure air pump and a gas ejection device are used to achieve precise sorting of coal-series kaolin and ordinary coal gangue.

7. An intelligent identification and separation method for coal-series kaolinite in coal gangue, which uses the intelligent identification and separation system for coal-series kaolinite in coal gangue as described in any one of claims 1-6, is characterized in that: Including the following steps: S1: Equipment startup and initialization; S2: Acquire the dual-energy X-ray image, texture image and coal gangue thickness data of the coal gangue, and obtain a comprehensive feature vector after processing through the multi-modal data acquisition and processing module; S2: Construct a binary classification evaluation label system; S3: Input the comprehensive feature vector as input data into the image input classification algorithm module and output the classification result; S4: Generate a sorting instruction according to the classification result and send the sorting instruction to the control and sorting execution module; S5: The control and sorting execution module realizes the precise sorting of coal-series kaolin according to the sorting instruction.

8. A computer device, comprising a memory, a processor, and a computer program stored on the memory, characterized in that: The processor executes the computer program to implement the steps of the method described in claim 7.

9. A computer-readable storage medium having computer programs / instructions stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to claim 7 are implemented.

10. A computer program product, comprising a computer program / instructions, characterized in that: When the computer program / instructions are executed by a processor, the steps of the method according to claim 7 are implemented.

Images