Deep Learning-Based Concrete Mix Design Adjustment Method, Device, and Readable Medium

Abstract

The present invention discloses a method, device and readable medium for adjusting concrete formulation based on deep learning. By using an object detection model, the concrete area image in the mixing image collected in real time is extracted and preprocessed to obtain the processed concrete area image, and the image gray histogram is calculated to obtain the gray mean change curve. A first relationship between the gray mean change curve and the workability of concrete is established, and the predicted value of the workability of concrete is determined. An instance segmentation model is used to segment the aggregate image to obtain the segmentation result, and the grading of the aggregate is determined based on the segmentation result. A second relationship between the change value of the workability of concrete and the change value of the water consumption and / or the change value of the aggregate dosage is established, and the water consumption and / or the aggregate dosage are adjusted based on the predicted value of the workability, the grading of the aggregate and the second relationship. The above steps are repeated to make the workability meet the requirements. The present invention can adjust the concrete formulation in real time and improve the efficiency.

Description

Technical Field

[0001] The present invention relates to the field of industrial intelligent control, and particularly relates to a method, device and readable medium for adjusting concrete formula based on deep learning. Background Art

[0002] The design of concrete mix ratio is usually carried out under laboratory conditions and then fine-tuned during construction site use. However, this mix ratio is determined under one kind of sand and stone conditions, and different sources of sand and stone raw materials are often used during production. The particle shape parameters, moisture content, etc. of sand and stone have a greater impact on the workability of concrete (including slump, slump flow, etc.). Therefore, it is necessary to adjust the concrete mix ratio by measuring the parameters of sand and stone.

[0003] The workability of concrete is measured by relevant testers using testing instruments after the concrete is discharged from the mixer. Usually, the workability test must be carried out after adjusting the sand and stone materials. This brings three problems: First, it may take multiple rounds of repeated adjustment to make the workability of concrete meet the requirements by adjusting the sand and stone dosage, wasting labor and time; Second, once the test is unqualified after the concrete is discharged from the mixer, the whole batch of concrete can only be discarded, resulting in waste of resources. Third, there will be certain differences in the moisture content inside and outside the same batch of sand and stone raw materials, and the workability of concrete will be different. Therefore, there is an urgent need for a method that can detect the parameters of sand and stone and the workability of concrete in real time during the concrete production process. Summary of the Invention

[0004] In view of the above-mentioned technical problems, the purpose of the embodiments of the present application is to provide a method, device and readable medium for adjusting concrete formula based on deep learning to solve the technical problems mentioned in the above background art section.

[0005] In a first aspect, the embodiments of the present application provide a method for adjusting concrete formula based on deep learning, including the following steps:

[0006] S1, obtaining the mixing image during the concrete mixing process, extracting the concrete region image in the mixing image through an object detection model, and preprocessing the concrete region image to obtain a processed concrete region image;

[0007] S2, calculating the image gray histogram based on the processed concrete region image, and calculating the gray mean value according to the image gray histogram to obtain a gray mean value change curve;

[0008] S3, establishing a first relationship between the gray mean value change curve and the workability of concrete, and determining the predicted value of the workability of concrete in the mixing image according to the gray mean value change curve and the first relationship;

[0009] S4. Obtain the aggregate image before concrete mixing, segment the aggregate image using an instance segmentation model to obtain a segmentation result, and determine the gradation of the aggregates during the concrete mixing process based on the segmentation result;

[0010] S5. Establish a second relationship between the change value of the workability of the concrete corresponding to the gradation of different aggregates and the change value of the water consumption and / or the change value of the aggregate dosage. Determine whether it is necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process according to the predicted value of the workability of the concrete in the mixing image, and adjust the water consumption and / or the aggregate dosage during the concrete mixing process based on the predicted value of the workability of the concrete, the gradation of the aggregates during the concrete mixing process, and the second relationship. Repeat steps S1 - S5 to make the workability of the concrete meet the requirements.

[0011] In some embodiments, determining whether it is necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process according to the predicted value of the workability of the concrete in the mixing image in step S5 specifically includes:

[0012] Determine whether the predicted value of the workability of the concrete in the mixing image exceeds the preset threshold range. If so, it is necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process; otherwise, it is not necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process.

[0013] In some embodiments, the workability of the concrete is the slump of the concrete. In the first relationship, the fluctuation interval of the grayscale mean in the slump - grayscale mean change curve corresponds one - to - one. In the second relationship, the change value of the slump corresponds one - to - one with the change value of the water consumption and / or the change value of the aggregate dosage. Adjusting the water consumption and / or the aggregate dosage during the concrete mixing process based on the predicted value of the workability of the concrete, the gradation of the aggregates during the concrete mixing process, and the second relationship in step S5 specifically includes:

[0014] If the predicted value of the slump of the concrete is lower than the preset threshold range, increase the water consumption during the concrete mixing process according to the relationship between the increase value of the slump and the increase value of the water consumption in the second relationship;

[0015] If the predicted value of the slump of the concrete is higher than the preset threshold range, adjust the increase amount of the aggregate dosage during the concrete mixing process according to the relationship between the decrease value of the slump and the increase amount of the aggregate dosage in the second relationship corresponding to the gradation of the aggregates during the concrete mixing process.

[0016] In some embodiments, the target detection model is a trained first Mask - Rcnn neural network, the instance segmentation model is a trained second Mask - Rcnn neural network, and the backbone network of the first Mask - Rcnn neural network and the second Mask - Rcnn neural network is Resnet50.

[0017] In some embodiments, in step S1, preprocessing is performed on the concrete area image, which specifically includes:

[0018] Perform binarization processing on the concrete area image to obtain the binarized concrete area image;

[0019] Filter out the background part in the binarized concrete area image to obtain the processed concrete area image.

[0020] In some embodiments, in step S2, the calcHist function in OpenCV is used to calculate the grayscale histogram of the image. The grayscale histogram of the image records the number of pixels corresponding to different grayscale values in the processed concrete area image, and the grayscale mean value is the ratio between the sum of the grayscale values of all pixel points in the processed concrete area image and the number of pixels.

[0021] In some embodiments, in step S4, based on the segmentation result, the gradation of the aggregate during the concrete mixing process is determined, which specifically includes:

[0022] Obtain the contour of each particle in the aggregate image according to the segmentation result;

[0023] Based on the contour of each particle, use the fitEllipse function in OpenCV to calculate the fitted ellipse of the contour of each particle and the corresponding minor axis;

[0024] Judge the particle size range to which each particle belongs according to the size of the minor axis;

[0025] Statistically analyze the particle size range to which each particle belongs to obtain the gradation of the aggregate.

[0026] In a second aspect, an embodiment of the present application provides a device for adjusting the concrete formula based on deep learning, including:

[0027] A stirring image acquisition module, configured to acquire a stirring image during the concrete mixing process, extract the concrete area image in the stirring image through a target detection model, and perform preprocessing on the concrete area image to obtain the processed concrete area image;

[0028] A grayscale mean value calculation module, configured to calculate the grayscale histogram of the image based on the processed concrete area image, and calculate the grayscale mean value according to the grayscale histogram to obtain a grayscale mean value change curve;

[0029] A working performance prediction module, configured to establish a first relationship between the grayscale mean value change curve and the working performance of the concrete, and determine the predicted value of the working performance of the concrete in the stirring image according to the grayscale mean value change curve and the first relationship;

[0030] The aggregate gradation calculation module is configured to obtain the aggregate image before concrete mixing, segment the aggregate image using an instance segmentation model to obtain a segmentation result, and determine the gradation of the aggregate during the concrete mixing process based on the segmentation result;

[0031] The adjustment module is configured to establish a second relationship between the change value of the workability of the concrete corresponding to the gradation of different aggregates and the change value of the water consumption and / or the change value of the aggregate dosage. Determine whether it is necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process according to the predicted value of the workability of the concrete in the mixing image, and adjust the water consumption and / or the aggregate dosage during the concrete mixing process based on the predicted value of the workability of the concrete, the gradation of the aggregate during the concrete mixing process, and the second relationship. Repeat the execution of the mixing image acquisition module to the adjustment module to make the workability of the concrete meet the requirements.

[0032] In a third aspect, an embodiment of the present application provides an electronic device, including one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, enabling the one or more processors to implement the method described in any implementation manner of the first aspect.

[0033] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the method described in any implementation manner of the first aspect.

[0034] Compared with the prior art, the present invention has the following beneficial effects:

[0035] (1) By collecting the concrete mixing process images, establishing a target detection model, and judging in real time whether the workability of the concrete meets the requirements. Through the collection of sand and stone images on the belt, an instance segmentation model is established to predict particle shape parameters online. And when the workability of the concrete does not meet the requirements, the dosage can be adjusted according to the aggregate gradation and particle shape parameters at that moment, and the workability of the concrete can be verified.

[0036] (2) By real-time detecting the workability of the concrete and particle shape parameters such as sand and stone aggregates, the aggregate dosage is intelligently adjusted to meet the workability requirements of the concrete, reducing the necessary performance detection time during the production process and improving the production efficiency.

[0037] (3) When the workability of the concrete in this batch does not meet the requirements, the amount of aggregate to be supplemented can be calculated based on the aggregate gradation and particle shape at the current moment and adjusted in real time to ensure that the workability of the concrete meets the requirements when it leaves the machine, reducing resource waste and improving the efficiency of adjusting the formula. Description of the Drawings

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0039] Figure 1 is an exemplary device architecture diagram to which an embodiment of the present application can be applied;

[0040] Figure 2 is a schematic flowchart of the method for adjusting concrete formulation based on deep learning according to the embodiment of the present invention;

[0041] Figure 3 is a schematic diagram of the overall equipment of the method for adjusting concrete formulation based on deep learning according to the embodiment of the present invention;

[0042] Figure 4 is a schematic flowchart of the prediction of the working performance of concrete in the method for adjusting concrete formulation based on deep learning according to the embodiment of the present invention;

[0043] Figure 5 is a result diagram of the prediction of the working performance of concrete in the method for adjusting concrete formulation based on deep learning according to the embodiment of the present invention;

[0044] Figure 6 is an input and segmentation result diagram of the instance segmentation model in the method for adjusting concrete formulation based on deep learning according to the embodiment of the present invention;

[0045] Figure 7 is a schematic diagram of the device for adjusting concrete formulation based on deep learning according to the embodiment of the present invention;

[0046] Figure 8 is a schematic structural diagram of a computer device of an electronic device suitable for implementing the embodiments of the present application. Detailed implementation manners

[0047] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0048] Figure 1 Shows an exemplary device architecture 100 to which the method for adjusting concrete formulation based on deep learning or the device for adjusting concrete formulation based on deep learning according to the embodiments of the present application can be applied.

[0049] As Figure 1 shown, the device architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. The network 104 is used to provide a medium for communication links between the terminal devices 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0050] Users can use the terminal devices 101, 102, 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various applications can be installed on the terminal devices 101, 102, 103, such as data processing applications, file processing applications, etc.

[0051] The terminal devices 101, 102, 103 can be hardware or software. When the terminal devices 101, 102, 103 are hardware, they can be various electronic devices, including but not limited to smartphones, tablets, laptop computers, and desktop computers, etc. When the terminal devices 101, 102, 103 are software, they can be installed in the above-listed electronic devices. It can be implemented as multiple software or software modules (such as software or software modules for providing distributed services), or it can be implemented as a single software or software module. No specific limitation is made here.

[0052] The server 105 can be a server that provides various services, such as a background data processing server that processes files or data uploaded by the terminal devices 101, 102, 103. The background data processing server can process the obtained files or data and generate processing results.

[0053] It should be noted that the method for adjusting the concrete formula based on deep learning provided by the embodiments of the present application can be executed by the server 105, or can be executed by the terminal devices 101, 102, 103. Correspondingly, the device for adjusting the concrete formula based on deep learning can be set in the server 105, or can be set in the terminal devices 101, 102, 103.

[0054] It should be understood that Figure 1 the numbers of the terminal devices, the network, and the server in

[0055] Figure 2 are merely illustrative. According to the implementation requirements, there can be any number of terminal devices, networks, and servers. In the case where the data to be processed does not need to be obtained remotely, the above device architecture may not include a network, but only a server or a terminal device.

[0056] S1. Obtain the mixing image during the concrete mixing process, extract the concrete area image in the mixing image through the target detection model, and preprocess the concrete area image to obtain the processed concrete area image.

[0057] In a specific embodiment, refer to Figure 3 , set an image acquisition device above the mixer, set light sources on both sides of the image acquisition device, and collect the image information during the concrete mixing process in real time. Set a bracket 2 beside the concrete mixer 1. Fix the light source 3 and the image acquisition device 4 on the bracket. The image acquisition device 4 can capture the mixing image during the concrete mixing process through the feeding port of the mixer and transmit the mixing image to the computer 5 or the server side. Therefore, the mixing image during the concrete mixing process can be collected and obtained in real time. Use the target detection model to extract the concrete area (ROI area) in the mixing image to obtain the concrete area image. In a preferred embodiment, the target detection model is the trained first Mask-Rcnn neural network, and the backbone network of the first Mask-Rcnn neural network is Resnet50. Resnet50 is an existing neural network model, so its structure will not be elaborated here. This target detection model is constructed based on the deep learning neural network model, identifies and extracts the concrete area image, and predicts the working performance of the concrete according to steps S2 and S3.

[0058] In a specific embodiment, the preprocessing of the concrete area image in step S1 specifically includes:

[0059] Perform binarization processing on the concrete area image to obtain the binarized concrete area image;

[0060] Filter out the background part in the binarized concrete area image to obtain the processed concrete area image.

[0061] Specifically, the value of each pixel point in the binarized concrete area image is 0 or 1. By setting the pixel value of the background part to 0, the remaining background interference can be filtered out, which is convenient for subsequent calculation of the gray mean value.

[0062] S2. Calculate the image gray histogram based on the processed concrete area image, and calculate the gray mean value according to the image gray histogram to obtain the gray mean value change curve.

[0063] In a specific embodiment, the calcHist function in OpenCV is used to calculate the image gray histogram in step S2. Refer to Figure 4, the image grayscale histogram records the number of pixels corresponding to different grayscale values in the processed concrete area image. The abscissa is the grayscale value, and the ordinate is the number of pixels. The image grayscale histogram can be extracted by calculating using OpenCV from the processed concrete area image. The grayscale mean is the ratio between the sum of the grayscale values of all pixel points in the processed concrete area image and the number of pixels, that is:

[0064] Grayscale mean = sum of the grayscale values of all pixel points in the processed concrete area image / number of pixels.

[0065] Based on the change of the grayscale mean over time, a grayscale mean change curve is established, as Figure 5 shown. The grayscale mean change curve represents the fluctuation of the grayscale mean within a certain time range, and the fluctuation interval of the grayscale mean can be obtained.

[0066] S3. Establish a first relationship between the grayscale mean change curve and the workability of the concrete, and determine the predicted value of the workability of the concrete in the mixing image according to the grayscale mean change curve and the first relationship.

[0067] In a specific embodiment, the workability of the concrete is the slump of the concrete, and in the first relationship, the slump corresponds one-to-one with the fluctuation interval of the grayscale mean in the grayscale mean change curve.

[0068] Specifically, when determining the first relationship, the slump of the concrete is adjusted respectively, and according to the fluctuation interval of the grayscale mean in the grayscale mean change curve obtained under the numerical range of the slump of this concrete. For example: the grayscale mean change curve corresponding to a slump of 180 fluctuates between 164 and 166; the grayscale mean change curve corresponding to a slump of 150 fluctuates between 162 and 164; the grayscale mean change curve corresponding to a slump of 120 fluctuates between 160 and 162. Different workability (slump) concretes are distinguished according to the size of the grayscale mean during the mixing process, and thus it can be judged whether the workability of the concrete at this moment meets the requirements.

[0069] S4. Obtain the aggregate image before concrete mixing, segment the aggregate image using an instance segmentation model to obtain the segmentation result, and determine the aggregate gradation during the concrete mixing process based on the segmentation result.

[0070] In a specific embodiment, the instance segmentation model is a trained second Mask-Rcnn neural network, and the backbone network of the second Mask-Rcnn neural network is Resnet50. Resnet50 is an existing neural network model, so its structure will not be described in detail here. Determining the aggregate gradation during the concrete mixing process based on the segmentation result in step S4 specifically includes:

[0071] Obtain the contour of each particle in the aggregate image according to the segmentation result;

[0072] Based on the contour of each particle, use the fitEllipse function in OpenCV to calculate the fitted ellipse of the contour of each particle and the corresponding minor axis;

[0073] Judge the particle size range to which each particle belongs according to the size of the minor axis;

[0074] Statistically analyze the particle size range to which each particle belongs to obtain the gradation of the aggregate.

[0075] Specifically, the aggregate gradation is the proportional relationship of different particle size particles that make up the aggregate; the aggregate gradation is mainly divided into continuous gradation and discontinuous gradation (single-grain gradation). Continuous gradation mainly means that below the maximum particle size, there are other corresponding particle sizes in sequence without interruption, in order to fully fill the voids between the aggregates. Discontinuous gradation means that one or several intermediate particle sizes are missing in the continuous gradation. Concrete aggregates are an important part of concrete and play a role of skeleton and filling in concrete. They are usually divided into two categories: fine aggregates and coarse aggregates. In concrete, those with a particle size between 0.155 and 5 mm are generally called fine aggregates; those with a particle size greater than 5 mm are called coarse aggregates. Coarse aggregates are classified into pebbles, crushed stones, crushed pebbles, and mixtures of pebbles and crushed stones according to their types. If the aggregates are all too fine, it may cause the slump loss of the concrete mixture to increase and bleeding to increase, while if they are too coarse, it will cause the cohesion of the concrete mixture to become poor and the slump to decrease. Therefore, in the actual production process, it is necessary to handle the optimal combination of the aggregate gradation well. According to the requirements of the aggregate gradation, the proportion of aggregates that meet the slump can be calculated, and then the amount of aggregates can be calculated. In the embodiments of the present application, an instance segmentation model can be used to realize the online prediction of the aggregate gradation and particle shape parameters. Refer to Figure 3 , convey the aggregate 8 above the conveyor belt 9, set image acquisition devices 6 on both sides of the conveyor belt 9, set light sources 7 on both sides of the image acquisition devices, collect the aggregate image in real time, and transmit the aggregate image to the computer 5 or the server side. Refer to Figure 6 , Figure 6 (a) is the aggregate image input into the instance segmentation model, Figure 6 (b) is the segmentation result output after the instance segmentation model performs segmentation. According to the segmentation result, the outer contour of each particle can be obtained. Using the fitEllipse function in OpenCV, the fitted ellipse of the particle contour and the corresponding minor axis can be obtained. According to the size of the minor axis, it can be judged which grade of aggregate this particle belongs to. By judging all the particles, the gradation of all the aggregates in this batch can be obtained.

[0076] S5. Establish a second relationship between the change value of the workability of the concrete corresponding to the gradation of different aggregates and the change value of the water consumption and / or the change value of the aggregate dosage. Determine whether it is necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process according to the predicted value of the workability of the concrete in the mixing image, and adjust the water consumption and / or the aggregate dosage during the concrete mixing process based on the predicted value of the workability of the concrete, the gradation of the aggregates during the concrete mixing process, and the second relationship. Repeat steps S1 - S5 to make the workability of the concrete meet the requirements.

[0077] Specifically, there is a second relationship corresponding to each gradation of the aggregates. In the second relationship, the change value of the slump corresponds one-to-one with the change value of the water consumption and / or the aggregate dosage. Specifically, the second relationship is determined based on multiple experiments. When determining the second relationship, a model of the gradation, the aggregate dosage, and the slump influence is established through multiple experiments. This model is a corresponding relationship established based on experimental data, reflecting the relationship between the gradation of the aggregates, the increase amount (percentage) of the aggregate dosage and the decrease value of the slump, or the increase amount of the water consumption and the increase value of the slump. The increase amount of the aggregate dosage or the increase amount of the water consumption is the increase amount relative to the original mixing amount.

[0078] In a specific embodiment, in step S5, determining whether it is necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process according to the predicted value of the workability of the concrete in the mixing image specifically includes:

[0079] Determine whether the predicted value of the workability of the concrete in the mixing image exceeds the preset threshold range. If so, it is necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process; otherwise, it is not necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process.

[0080] Therefore, through the embodiments of the present application, not only can the workability of the concrete be predicted in real time, but also the water consumption and / or the aggregate dosage during the concrete mixing process can be adjusted according to the relationships such as the workability of the concrete and the gradation of the aggregates, realizing the intelligent adjustment of the concrete formula.

[0081] In step S5, adjusting the water consumption and / or the aggregate dosage during the concrete mixing process based on the predicted value of the workability of the concrete, the gradation of the aggregates during the concrete mixing process, and the second relationship specifically includes:

[0082] If the predicted value of the slump of the concrete is lower than the preset threshold range, increase the water consumption during the concrete mixing process according to the relationship between the increase value of the slump and the increase value of the water consumption in the second relationship;

[0083] If the predicted value of the slump of the concrete is higher than the preset threshold range, the increase amount of the aggregate used in the concrete mixing process is adjusted according to the relationship between the decrease value of the slump and the increase amount of the aggregate used corresponding to the aggregate gradation during the concrete mixing process.

[0084] Specifically, for each aggregate gradation, there is a corresponding relationship between the decrease value of the slump and the increase amount of the aggregate used. If the required slump of the concrete is 150±20, and the predicted value of the slump of the concrete is 110, then the amount of water used needs to be increased accordingly. If the predicted value of the slump of the concrete is 190, according to the aggregate gradation measured in step S4, a certain amount of aggregate is supplemented, and this amount needs to be obtained by establishing a model of the relationship between the gradation, the aggregate dosage, and the slump influence through multiple experiments.

[0085] Within a slump range, the increased or decreased value does not fluctuate greatly. As long as the finally adjusted slump is within the error tolerance range, for example, the required slump is 150 and the error is ±20, it only means that according to this corresponding relationship of the aggregate gradation, the slump can be adjusted to be closer to 150.

[0086] The present invention can effectively predict the working performance during the concrete mixing process in real time, detect the gradation and particle shape parameters of the sand and gravel aggregate, reduce the necessary performance detection time during the production process, and improve the production efficiency. When the working performance of this batch of concrete does not meet the requirements, the amount of aggregate to be supplemented is calculated based on the aggregate gradation and particle shape at the current moment, and adjusted in real time to ensure that the working performance of the concrete meets the requirements when it leaves the machine, reduce resource waste, and improve the efficiency of adjusting the formula.

[0087] Further referring to Figure 7 As an implementation of the methods shown in the above figures, an embodiment of a concrete formula adjustment device based on deep learning is provided in this application. This device embodiment corresponds to Figure 2 the method embodiment shown, and this device can be specifically applied to various electronic devices.

[0088] An embodiment of the application provides a concrete formula adjustment device based on deep learning, including:

[0089] A mixing image acquisition module 1, configured to acquire the mixing image during the concrete mixing process, extract the concrete area image in the mixing image through a target detection model, and preprocess the concrete area image to obtain the processed concrete area image;

[0090] A grayscale mean value calculation module 2, configured to calculate the image grayscale histogram based on the processed concrete area image, and calculate the grayscale mean value according to the image grayscale histogram to obtain the grayscale mean value change curve;

[0091] The working performance prediction module 3 is configured to establish a first relationship between the grayscale mean change curve and the working performance of the concrete, and determine the predicted value of the working performance of the concrete in the mixing image according to the grayscale mean change curve and the first relationship;

[0092] The aggregate gradation calculation module 4 is configured to obtain the aggregate image before concrete mixing, segment the aggregate image by using an instance segmentation model to obtain a segmentation result, and determine the gradation of the aggregate in the concrete mixing process based on the segmentation result;

[0093] The adjustment module 5 is configured to establish a second relationship between the change value of the working performance of the concrete corresponding to the gradation of different aggregates and the change value of the water consumption and / or the change value of the aggregate dosage, judge whether it is necessary to adjust the water consumption and / or the aggregate dosage in the concrete mixing process according to the predicted value of the working performance of the concrete in the mixing image, and adjust the water consumption and / or the aggregate dosage in the concrete mixing process based on the predicted value of the working performance of the concrete, the gradation of the aggregate in the concrete mixing process and the second relationship, and repeatedly execute the mixing image acquisition module to the adjustment module to make the working performance of the concrete meet the requirements.

[0094] Next, refer to Figure 8 , which shows a schematic structural diagram of a computer device 800 suitable for implementing the embodiments of the present application (such as Figure 1 the server or terminal device shown). Figure 8 The electronic device shown is only an example and should not bring any limitation to the functions and usage scope of the embodiments of the present application.

[0095] As Figure 8 shown, the computer device 800 includes a central processing unit (CPU) 801 and a graphics processing unit (GPU) 802, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 803 or the program loaded from the storage section 809 into the random access memory (RAM) 804. In the RAM 804, various programs and data required for the operation of the device 800 are also stored. The CPU 801, GPU 802, ROM 803, and RAM 804 are connected to each other through a bus 805. The input / output (I / O) interface 806 is also connected to the bus 805.

[0096] The following components are connected to the I / O interface 806: an input part 807 including a keyboard, a mouse, etc.; an output part 808 including, for example, a liquid crystal display (LCD), etc. and a speaker, etc.; a storage part 809 including a hard disk, etc.; and a communication part 810 including a network interface card such as a LAN card, a modem, etc. The communication part 810 performs communication processing via a network such as the Internet. A drive 811 may also be connected to the I / O interface 806 as needed. A removable medium 812, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 811 as needed so that a computer program read from it is installed into the storage part 809 as needed.

[0097] Specifically, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, an embodiment of the present disclosure includes a computer program product that includes a computer program carried on a computer-readable medium, and the computer program includes program codes for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network via the communication part 810, and / or installed from the removable medium 812. When the computer program is executed by a central processing unit (CPU) 801 and a graphics processing unit (GPU) 802, the above functions defined in the method of the present application are executed.

[0098] It should be noted that the computer-readable medium described in this application can be a computer-readable signal medium, a computer-readable medium, or any combination of the two. The computer-readable medium can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor devices, apparatuses, or components, or any combination of the above. More specific examples of the computer-readable medium can include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this application, the computer-readable medium can be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution device, apparatus, or component. And in this application, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution device, apparatus, or component. The program code contained on the computer-readable medium can be transmitted by any suitable medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0099] The computer program code for performing the operations of this application can be written in one or more programming languages or combinations thereof. The programming languages include object-oriented programming languages - such as Java, Smalltalk, C++, and also include conventional procedural programming languages - such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network - including a local area network (LAN) or a wide area network (WAN) - or, it can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0100] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of apparatuses, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that, in some alternative implementations, the functions noted in the blocks may occur in a different order than that noted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based device that performs the specified function or operation, or can be implemented by a combination of dedicated hardware and computer instructions.

[0101] The modules described in the embodiments of the present application can be implemented in software or in hardware. The described modules can also be provided in a processor.

[0102] As another aspect, the present application also provides a computer-readable medium, which may be included in the electronic device described in the above embodiments; or may exist separately without being assembled into the electronic device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by the electronic device, the electronic device is caused to: acquire a mixing image during the concrete mixing process, extract the concrete area image in the mixing image through a target detection model, and preprocess the concrete area image to obtain a processed concrete area image; calculate an image gray histogram based on the processed concrete area image, and calculate a gray mean value according to the image gray histogram to obtain a gray mean value change curve; establish a first relationship between the gray mean value change curve and the working performance of the concrete, and determine a predicted value of the working performance of the concrete in the mixing image according to the gray mean value change curve and the first relationship; acquire an aggregate image before the concrete mixing, segment the aggregate image by using an instance segmentation model to obtain a segmentation result, and determine the gradation of the aggregate during the concrete mixing process based on the segmentation result; establish a second relationship between the change value of the working performance of the concrete corresponding to the gradation of different aggregates and the change value of the water consumption and / or the change value of the aggregate dosage, determine whether it is necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process according to the predicted value of the working performance of the concrete in the mixing image, and adjust the water consumption and / or the aggregate dosage during the concrete mixing process based on the predicted value of the working performance of the concrete, the gradation of the aggregate during the concrete mixing process, and the second relationship, and repeat the above steps to make the working performance of the concrete meet the requirements.

[0103] The above description is only a preferred embodiment of the present application and an explanation of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solutions formed by the specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, the technical solutions formed by the mutual replacement of the above features with the technical features (but not limited to) disclosed in the present application that have similar functions.

Claims

1. A method for adjusting the concrete formula based on deep learning, characterized in that, It includes the following steps: S1. Obtain the mixing image during the concrete mixing process, extract the concrete area image in the mixing image through a target detection model, and preprocess the concrete area image to obtain a processed concrete area image; S2. Calculate the image gray histogram based on the processed concrete area image, and calculate the gray mean value according to the image gray histogram to obtain a gray mean value change curve; S3. Establish a first relationship between the gray mean value change curve and the working performance of the concrete. The working performance of the concrete is the slump of the concrete. Determine the predicted value of the working performance of the concrete in the mixing image according to the gray mean value change curve and the first relationship; S4. Obtain the aggregate image before concrete mixing, segment the aggregate image by using an instance segmentation model to obtain a segmentation result, and determine the gradation of the aggregate during the concrete mixing process based on the segmentation result; S5. Establish a second relationship between the change value of the working performance of the concrete corresponding to the gradation of different aggregates and the change value of the water consumption and / or the change value of the aggregate dosage. Judge whether it is necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process according to the predicted value of the working performance of the concrete in the mixing image, and adjust the water consumption and / or the aggregate dosage during the concrete mixing process based on the predicted value of the working performance of the concrete, the gradation of the aggregate during the concrete mixing process and the second relationship. Repeat steps S1 - S5 to make the working performance of the concrete meet the requirements.

2. The method for adjusting the concrete formula based on deep learning according to claim 1, characterized in that In step S5, judging whether it is necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process according to the predicted value of the working performance of the concrete in the mixing image specifically includes: Judge whether the predicted value of the working performance of the concrete in the mixing image exceeds the preset threshold range. If so, it is necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process; otherwise, it is not necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process.

3. The method for adjusting the concrete formula based on deep learning according to claim 2, characterized in that, In the first relationship, the slump corresponds one-to-one with the fluctuation range of the gray mean value in the gray mean value change curve. In the second relationship, the change value of the slump corresponds one-to-one with the change value of the water consumption and / or the change value of the aggregate dosage. In step S5, adjusting the water consumption and / or the aggregate dosage during the concrete mixing process based on the predicted value of the working performance of the concrete, the gradation of the aggregate during the concrete mixing process and the second relationship specifically includes: If the predicted value of the slump of the concrete is lower than the preset threshold range, increase the water consumption during the concrete mixing process according to the relationship between the increase value of the slump and the increase value of the water consumption in the second relationship; If the predicted value of the slump of the concrete is higher than the preset threshold range, adjust the increase amount of the aggregate dosage during the concrete mixing process according to the relationship between the decrease value of the slump and the increase amount of the aggregate dosage in the second relationship corresponding to the gradation of the aggregate during the concrete mixing process.

4. The method for adjusting the concrete formula based on deep learning according to claim 1, characterized in that The target detection model is a trained first Mask-Rcnn neural network, the instance segmentation model is a trained second Mask-Rcnn neural network, and the backbone network of the first Mask-Rcnn neural network and the second Mask-Rcnn neural network is Resnet50.

5. The method for adjusting the concrete formula based on deep learning according to claim 1, characterized in that In step S1, preprocessing the concrete area image specifically includes: Performing binarization processing on the concrete area image to obtain a binarized concrete area image; Filtering out the background part in the binarized concrete area image to obtain the processed concrete area image.

6. The method for adjusting the concrete formula based on deep learning according to claim 1, wherein In step S2, the calcHist function in OpenCV is used to calculate the image gray histogram. The image gray histogram records the number of pixels corresponding to different gray values in the processed concrete area image, and the gray mean is the ratio between the sum of the gray values of all pixel points in the processed concrete area image and the number of pixels.

7. The method for adjusting the concrete formula based on deep learning according to claim 1, characterized in that, In step S4, determining the aggregate gradation during the concrete mixing process based on the segmentation result specifically includes: Obtaining the contour of each particle in the aggregate image according to the segmentation result; Based on the contour of each particle, using the fitEllipse function in OpenCV to calculate the fitted ellipse of the contour of each particle and the corresponding minor axis; Judging the particle size range to which each particle belongs according to the size of the minor axis; Counting the particle size ranges to which each particle belongs to obtain the aggregate gradation.

8. A device for adjusting the concrete formula based on deep learning, characterized in that, Including: A mixing image acquisition module configured to acquire a mixing image during the concrete mixing process, extract the concrete area image in the mixing image through a target detection model, and preprocess the concrete area image to obtain a processed concrete area image; A gray mean calculation module configured to calculate an image gray histogram based on the processed concrete area image, and calculate the gray mean according to the image gray histogram to obtain a gray mean change curve; A working performance prediction module configured to establish a first relationship between the gray mean change curve and the working performance of the concrete. The working performance of the concrete is the slump of the concrete, and determine the predicted value of the working performance of the concrete in the mixing image according to the gray mean change curve and the first relationship; An aggregate gradation calculation module configured to acquire an aggregate image before concrete mixing, segment the aggregate image using an instance segmentation model to obtain a segmentation result, and determine the aggregate gradation during the concrete mixing process based on the segmentation result; An adjustment module configured to establish a second relationship between a change value of the workability of concrete corresponding to the gradation of different aggregates and a change value of the water consumption and / or a change value of the aggregate dosage, determine whether it is necessary to adjust the water consumption and / or the aggregate dosage during the concrete mixing process according to the predicted value of the workability of the concrete in the mixing image, and adjust the water consumption and / or the aggregate dosage during the concrete mixing process based on the predicted value of the workability of the concrete, the gradation of the aggregates during the concrete mixing process, and the second relationship, and repeatedly execute the mixing image acquisition module to the adjustment module to make the workability of the concrete meet the requirements.

9. An electronic device, comprising: One or more processors; A storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, the method according to any one of claims 1-7 is implemented.

Images