Sandstone particle size anomaly detection method and system, medium and product

Abstract

The invention discloses a gravel particle size anomaly detection method and system, a medium and a product, and relates to the field of computer vision. The method comprises the steps that the target feeding amount of the vibrating feeder is calculated; the vibration amplitude and the vibration frequency of the vibration feeder are determined; the vibration feeder is controlled to vibrate at the vibration amplitude and the vibration frequency; acquiring a first sandstone image of a target detection area of the conveyor belt; according to the first sandstone image, identifying first sandstone particle sizes of a plurality of sandstones in the target detection area; determining a target vibration amplitude and a target vibration frequency of a target vibration device under the condition that a target gravel particle size in the first gravel particle size exceeds a preset threshold value; sending a control instruction to the target vibration device; under the condition that the target vibration device completes the vibration operation, according to the second sandstone image, identifying a second sandstone particle size of the sandstone in the target detection area; and performing anomaly detection on each sandstone in the target detection area. By implementing the technical scheme, the accuracy of the detection result of the sandstone particle size anomaly is improved.

Description

Technical Field

[0001] This application relates to the field of computer vision, and in particular, to a method, system, medium and product for detecting abnormal sand and gravel particle sizes. Background Art

[0002] In modern construction projects, concrete, as a widely used building material, its quality directly affects the structural safety and durability of buildings. Sand and gravel, as the main aggregates of concrete, their particle size and distribution play a crucial role in the workability, mechanical properties and durability of concrete. To ensure that the quality of concrete meets high standards, it is usually necessary to perform abnormal detection on the purchased sand and gravel before making concrete.

[0003] Currently, abnormal detection of sand and gravel particle sizes is usually carried out by means of image recognition. This method mainly uses a high-definition camera to collect sand and gravel sample images, and then uses image processing algorithms to segment and extract features of the sand and gravel particles in the images, so as to calculate the particle size based on the extracted features and determine whether there are abnormalities.

[0004] However, in practical applications, the sand and gravel cannot be completely laid flat, and some sand and gravel particles will stack, block or adhere to each other, resulting in an increased difficulty in segmenting the sand and gravel particles during image recognition, and the stacked and blocked sand and gravel particles cannot be accurately segmented, causing deviations in feature extraction based on the segmentation results, affecting the accuracy of particle size calculation, and resulting in deviations in abnormal detection results. Summary of the Invention

[0005] This application provides a method, system, medium and product for detecting abnormal sand and gravel particle sizes, which can improve the accuracy of abnormal detection results of sand and gravel particle sizes.

[0006] In a first aspect, the present application provides a method for detecting abnormal sand and gravel particle sizes. The method includes: calculating a target feeding amount of a vibrating feeder based on the running speed of a conveyor belt, the preset width of the conveyor belt, and the target particle size range of preset normal sand and gravel; the vibrating feeder is located above the conveyor belt; determining the vibration amplitude and vibration frequency of the vibrating feeder according to the target feeding amount; controlling the vibrating feeder to vibrate with the vibration amplitude and the vibration frequency, so that the sand and gravel in the vibrating feeder are conveyed onto the conveyor belt according to the target feeding amount; acquiring a first sand and gravel image of a target detection area of the conveyor belt; the first sand and gravel image shows a plurality of sand and gravel; identifying the first sand and gravel particle sizes of the plurality of sand and gravel in the target detection area according to the first sand and gravel image; in the case where there is a target sand and gravel particle size exceeding a preset threshold among the first sand and gravel particle sizes, determining the target vibration amplitude and target vibration frequency of a target vibration device corresponding to the target detection area according to the maximum value of the target particle size range, the target sand and gravel particle size, and the type of sand and gravel; the target vibration device is located below the conveyor belt; sending a corresponding control instruction to the target vibration device according to the target vibration amplitude and the target vibration frequency; in the case where the target vibration device completes the vibration operation according to the control instruction, identifying the second sand and gravel particle sizes of the sand and gravel in the target detection area according to a second sand and gravel image of the target detection area re-acquired; performing abnormal detection on each sand and gravel in the target detection area according to the second sand and gravel particle sizes, to obtain an abnormal detection result of the target detection area; the abnormal detection is to determine whether the sand and gravel particle size of the sand and gravel exceeds the target particle size range of the preset normal sand and gravel.

[0007] By adopting the above technical solution, by adjusting the vibration frequency and vibration amplitude of the vibrating feeder, the sand and gravel can be evenly placed on the conveyor belt, and the situation that the sand and gravel in the same area are dense due to excessive feeding amount, and then a large amount of sand and gravel are stacked can be avoided, ensuring that the sand and gravel entering the detection area are evenly distributed. At the same time, after calculating the sand and gravel particle size through image recognition, it is judged whether there is a situation of sand and gravel stacking by judging whether the sand and gravel particle size exceeds a preset threshold. When there is a situation of sand and gravel stacking, the stacked sand and gravel are evenly distributed by the vibration device, and then the sand and gravel particle size is recalculated through image recognition, ensuring that the calculated sand and gravel particle size is the actual particle size of a single sand and gravel particle, rather than the combined particle size of multiple stacked sand and gravel particles. Performing abnormal detection according to the recalculated sand and gravel particle size reduces the false detection caused by sand and gravel stacking, and can effectively improve the accuracy of the abnormal detection result of the sand and gravel particle size.

[0008] In combination with some embodiments of the first aspect, in some embodiments, when there is a target sand and gravel particle size exceeding a preset threshold in the first sand and gravel particle size, the target vibration amplitude and target vibration frequency of the target vibration device corresponding to the target detection area are determined according to the maximum value of the target particle size range, the target sand and gravel particle size, and the sand and gravel type. Specifically, it includes: when there is a target sand and gravel particle size exceeding a preset threshold in the first sand and gravel particle size, the amplitude basic calculation parameters and frequency basic calculation parameters corresponding to the sand and gravel type and the target particle size range are obtained according to the sand and gravel type and the target particle size range; the amplitude basic calculation parameters include basic vibration amplitude, amplitude influence coefficient, amplitude correction coefficient, standard sand and gravel mass, sand and gravel density, sand and gravel elastic modulus, and sand and gravel Poisson's ratio; the frequency basic calculation parameters include basic vibration frequency, frequency influence coefficient, frequency correction coefficient, standard sand and gravel mass, sand and gravel elastic modulus, and sand and gravel Poisson's ratio; the sand and gravel mass is calculated according to the sand and gravel density and the target sand and gravel particle size; the amplitude basic calculation parameters, the target sand and gravel particle size, the maximum value of the target particle size range, and the sand and gravel mass are substituted into the vibration amplitude calculation formula to obtain the target vibration amplitude of the target vibration device corresponding to the target detection area, and the frequency basic calculation parameters, the target sand and gravel particle size, the maximum value of the target particle size range, and the sand and gravel mass are substituted into the vibration frequency calculation formula to obtain the target vibration frequency of the target vibration device corresponding to the target detection area; Among them, the vibration amplitude calculation formula is: The vibration frequency calculation formula is: Among them, A is the target vibration amplitude, f is the target vibration frequency, A0 is the basic vibration amplitude, f0 is the basic vibration frequency, d is the target sand and gravel particle size, d max is the maximum value of the target particle size range, k1 is the amplitude influence coefficient, k2 is the frequency influence coefficient, m is the sand and gravel mass, m0 is the standard sand and gravel mass, ρ is the sand and gravel density, E is the sand and gravel elastic modulus, v is the sand and gravel Poisson's ratio, k3 is the amplitude correction coefficient, k4 is the frequency correction coefficient.

[0009] By adopting the above technical solution, the vibration amplitude and vibration frequency of the vibration device are calculated through the vibration amplitude calculation formula and the vibration frequency calculation formula, comprehensively considering the influences of various factors such as sand and gravel characteristics, physical properties, and actual working conditions. Compared with the traditional method of setting by single factor or simple experience, this precise calculation method can more accurately adapt to sand and gravel in different states, can more effectively change the distribution and state of sand and gravel, make the stacked sand and gravel evenly disperse, and reduce the risk of ineffective treatment or over-treatment caused by unreasonable vibration parameters.

[0010] In some embodiments in combination with some embodiments of the first aspect, determining the vibration amplitude and vibration frequency of the vibrating feeder according to the target feeding amount specifically includes: calculating the maximum sand and gravel mass based on the sand and gravel density corresponding to the sand and gravel type and the maximum value of the target particle size range; obtaining the corresponding feeding amount table according to the sand and gravel type and the maximum sand and gravel mass; there is one or more feeding amount tables, each feeding amount table corresponding to a unique sand and gravel type and sand and gravel mass, and the table includes the feeding amount and the corresponding vibration amplitude and vibration frequency of the vibrating feeder; searching for the vibration amplitude and vibration frequency of the vibrating feeder corresponding to the target feeding amount in the feeding amount table.

[0011] Adopting the above technical solution, obtaining the corresponding feeding amount table according to the sand and gravel type and the maximum sand and gravel mass, and then determining the vibration amplitude and vibration frequency of the vibrating feeder according to the feeding amount table and the target feeding amount, so that the sand and gravel distributed in the same area on the conveyor belt are uniform, avoiding the phenomenon that the sand and gravel are too dense or too sparse in the same area on the conveyor belt, and reducing the situation of sand and gravel stacking caused by the sand and gravel being too dense.

[0012] In some embodiments in combination with some embodiments of the first aspect, after the step of performing abnormal detection on each sand and gravel in the target detection area according to the second sand and gravel particle size to obtain the abnormal detection result of the target detection area, the method further includes: determining the flatness of each sub-area in the target detection area based on the no-sand-and-gravel image collected in the target detection area; calculating the oversize ratio and the undersize ratio of each sub-area when there is one or more flatness values in the flatness that exceed the preset flatness threshold; the oversize ratio is the ratio of the total number of sand and gravel with a particle size greater than the maximum value of the target particle size range in the sub-area to the total number of abnormal sand and gravel; the undersize ratio is the ratio of the total number of sand and gravel with a particle size less than the minimum value of the target particle size range in the sub-area to the total number of abnormal sand and gravel; screening all the sand and gravel in each sub-area according to the flatness, the oversize ratio, and the undersize ratio of each sub-area; conveying the other sand and gravel in the target detection area after screening to the area of non-detected sand and gravel through the conveyor belt.

[0013] Adopting the above technical solution, the accuracy of the abnormal detection result is further determined by calculating the flatness of each sub-area in the detection area. When there is a sub-area with a flatness exceeding the preset threshold, it means that the sub-area is uneven, resulting in a deviation in the sand and gravel image collected in the sub-area, affecting the accuracy of the sand and gravel particle size calculated by image recognition, and further affecting the accuracy of the sand and gravel particle size abnormal detection result. At this time, according to the flatness, the oversize ratio, and the undersize ratio of each sub-area, the sand and gravel that can be determined as abnormal are screened out, and the other sand and gravel are re-detected for abnormalities, avoiding misjudgment caused by local unevenness and improving the accuracy of the sand and gravel particle size abnormal detection result.

[0014] In combination with some embodiments of the first aspect, in some embodiments, based on the gravel-free image of the target detection area collected, to determine the flatness of each sub-area within the target detection area, specifically including: based on the gravel-free image of the target detection area collected, identifying the actual coordinate information of each preset point in the image; according to the preset coordinate information of each preset point and the actual coordinate information of each preset point, calculating the offset of each preset point; according to the offsets of one or more preset points within each sub-area in the target detection area, calculating the flatness of each sub-area.

[0015] Adopting the above technical solution, taking the gravel-free image as a reference to identify the actual coordinates of the preset points, avoiding the interference of gravel, and ensuring the integrity of the image information of the preset points collected. By comparing the preset and actual coordinates to calculate the offset, and then calculating the flatness according to the offsets of the preset points within each sub-area, the flatness can objectively evaluate the surface conditions of each sub-area within the target detection area, reflecting the overall flatness of the target detection area from the local, providing a basis for judging the credibility of the abnormal detection result, and further providing a basis for judging the accuracy of the abnormal detection result, thus avoiding misjudgment caused by local unevenness.

[0016] In combination with some embodiments of the first aspect, in some embodiments, according to the flatness, the oversize ratio, and the undersize ratio of each sub-area, screening all the gravels within each sub-area, specifically including: in the case where the flatness of the target sub-area does not exceed the preset flatness threshold, screening out the gravels in the target sub-area whose particle size exceeds the target particle size range; the target sub-area is one of all the sub-areas; in the case where the flatness of the target sub-area exceeds the preset flatness threshold and the oversize ratio of the target sub-area is greater than the undersize ratio, screening out the gravels in the target sub-area whose particle size is less than the minimum value of the target particle size range; in the case where the flatness of the target sub-area exceeds the preset flatness threshold and the oversize ratio of the target sub-area is less than the undersize ratio, screening out the gravels in the target sub-area whose particle size is greater than the maximum value of the target particle size range.

[0017] By adopting the above technical scheme, when the flatness of a sub-region does not exceed the preset threshold, it indicates that the sub-region is relatively flat, and the credibility of the abnormal detection result of the sand and gravel in the sub-region is high. The sand and gravel whose particle size exceeds the target particle size range in the sub-region can be directly screened out to remove the abnormal sand and gravel in the sub-region; when the flatness of a sub-region exceeds the preset threshold, it indicates that the sub-region is uneven, and the credibility of the abnormal detection result of the sand and gravel in the sub-region is low. By comparing the larger proportion value and the smaller proportion value, it can be determined whether the overall calculation result of the sand and gravel particle size in the sub-region is larger or smaller than the actual result, and then the abnormal sand and gravel that can be determined in the sub-region can be screened out. When the oversized proportion is greater than the undersized proportion, it means that the overall calculation result of the sand and gravel particle size in the sub-region is larger than the actual result. At this time, the sand and gravel with a particle size smaller than the minimum value of the target particle size range can be determined as abnormal sand and gravel; on the contrary, if the undersized proportion is greater than the oversized proportion, it means that the overall calculation result of the sand and gravel particle size in the sub-region is smaller than the actual result. At this time, the sand and gravel with a particle size larger than the maximum value of the target particle size range can be determined as abnormal sand and gravel. Screening out abnormal sand and gravel that can be determined reduces the number of sand and gravel that need to be re-detected for abnormality, and can reduce the time for abnormal detection of sand and gravel while ensuring the accuracy of abnormal detection results.

[0018] In combination with some embodiments of the first aspect, in some embodiments, after the step of performing abnormal detection on each sand and gravel in the target detection area according to the second sand and gravel particle size to obtain the abnormal detection result of the target detection area, the method also includes: after detecting all the sand and gravel to be detected, obtaining the sand and gravel particle sizes of all abnormal sand and gravel to obtain an abnormal sand and gravel particle size set; according to the abnormal sand and gravel particle size set, determining the abnormal level of each abnormal sand and gravel; the abnormal level includes primary abnormality and secondary abnormality; the primary abnormality is that the sand and gravel particle size of the abnormal sand and gravel is in the preset commonly used sand and gravel particle size table; the secondary abnormality is that the sand and gravel particle size of the abnormal sand and gravel is not in the commonly used sand and gravel particle size table; obtaining the abnormal The first number of all abnormal sand and gravel with the first level abnormality and the second number of all abnormal sand and gravel with the abnormal level of the second level abnormality; according to the total number of the abnormal sand and gravel, the first number and the second number, calculate the first proportion value and the second proportion value; the first proportion value is the value of the first number divided by the total number of the abnormal sand and gravel; the second proportion value is the value of the second number divided by the total number of the abnormal sand and gravel; determine the quality of sand and gravel according to the qualified rate of sand and gravel, the first proportion value and the second proportion value; the qualified rate of sand and gravel is the value of the total number of all qualified sand and gravel divided by the total number of all sand and gravel to be tested; the qualified sand and gravel is the sand and gravel with a particle size within the target particle size range.

[0019] With the above technical solution, after all the sand and gravel are detected, the abnormal sand and gravel are classified according to whether the particle size is in the common table, and the quality of the sand and gravel is comprehensively and accurately evaluated based on the proportion of the abnormal sand and gravel corresponding to each abnormal level and the qualified rate of the sand and gravel, providing a basis for users to select high-quality sand and gravel suppliers, thereby reducing the sand and gravel cost of making concrete.

[0020] In a second aspect, an embodiment of the present application provides an abnormal detection system, including a conveyor belt, a vibrating feeder, a vibrating device, and a server. Among them, the server includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, and the computer program code includes computer instructions. The one or more processors call the computer instructions to cause the abnormal detection system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, including instructions, which, when running on the abnormal detection system, cause the abnormal detection system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0022] In a fourth aspect, the present application provides a computer program product, which, when running on the abnormal detection system, causes the abnormal detection system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0023] It can be understood that the abnormal detection system provided in the second aspect above, the storage medium provided in the third aspect, and the computer program product provided in the fourth aspect are all used to execute the method provided by the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, which will not be elaborated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. By adjusting the vibration frequency and vibration amplitude of the vibrating feeder, the sand and gravel can be evenly placed on the conveyor belt, and the situation of dense sand and gravel in the same area caused by excessive feeding amount, which leads to a large amount of stacking of sand and gravel, can be avoided, ensuring the uniform distribution of the sand and gravel entering the detection area. At the same time, when there is a situation of sand and gravel stacking in the detection area, the vibrating device is used to evenly distribute the stacked sand and gravel, and then the particle size of the sand and gravel is recalculated through image recognition to ensure that the calculated particle size of the sand and gravel is the actual particle size of a single sand and gravel particle, rather than the combined particle size of multiple stacked sand and gravel particles. Abnormal detection is performed based on the recalculated particle size of the sand and gravel, reducing false detection caused by sand and gravel stacking and effectively improving the accuracy of the abnormal detection result of the sand and gravel particle size.

[0025] 2. This application further determines the accuracy of the abnormal detection result by calculating the flatness of each sub-region within the detection area. When there is an uneven sub-region, based on the flatness, the oversize ratio, and the undersize ratio of each sub-region, the sands and gravels that can be determined as abnormal are screened out, and the other sands and gravels are re-detected for abnormalities, avoiding misjudgment caused by local unevenness and improving the accuracy of the abnormal detection result of the sands and gravels particle size. At the same time, the sands and gravels that can be determined as abnormal are screened out, reducing the number of sands and gravels that need to be re-detected for abnormalities. On the premise of ensuring the accuracy of the abnormal detection result, the time for detecting the abnormalities of the sands and gravels can be reduced.

[0026] 3. This application classifies the abnormal level of the sands and gravels according to whether the particle size of the abnormal sands and gravels is in the common table, and comprehensively and accurately evaluates the quality of the sands and gravels based on the proportion of abnormal sands and gravels corresponding to each abnormal level and the qualification rate of the sands and gravels, providing a basis for users to select high-quality sands and gravels suppliers, thereby reducing the cost of sands and gravels in the process of making concrete. Description of the Drawings

[0027] Figure 1 is a structural schematic diagram of a system architecture to which the abnormal detection method for the sands and gravels particle size in the embodiment of the present application can be applied; Figure 2 is a flowchart of the abnormal detection method for the sands and gravels particle size in the embodiment of the present application; Figure 3 is another flowchart of the abnormal detection method for the sands and gravels particle size in the embodiment of the present application; Figure 4 is an exemplary hardware structural schematic diagram of the abnormal detection system in the embodiment of the present application. Detailed Embodiments

[0028] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. As used in the specification and appended claims of the present application, the singular forms "a", "an", "the", "above", "said", "this" are also intended to include the plural forms, unless clearly indicated to the contrary in the context. It should also be understood that the term " / and" used in the present application refers to any or all possible combinations including one or more of the listed items.

[0029] Hereinafter, the terms "first" and "second" are only used for descriptive purposes, and cannot be construed as implying or indicating relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present application, unless otherwise specified, the meaning of "a plurality" is two or more.

[0030] Figure 1 It is a schematic structural diagram of a system architecture to which the gravel particle size anomaly detection method in the embodiments of the present application can be applied.

[0031] Please refer to Figure 1 , the anomaly detection system includes a server, a vibrating feeder, a conveyor belt, a vibration device, a camera, and a robotic arm.

[0032] The server, as the core component of the system, is used to process the image data collected by the camera and send control instructions to the vibrating feeder, the conveyor belt, the vibration device, and the robotic arm. The vibrating feeder is located above the conveyor belt and is used to evenly drop the gravel onto the conveyor belt. The conveyor belt is used to transport the gravel. The vibration device is located below the conveyor belt and is used to apply vibrations with a specific amplitude and frequency to the gravel on the conveyor belt, so that the gravel is evenly distributed on the conveyor belt. The robotic arm is located above or on both sides of the conveyor belt and is used to sort and grab the abnormal gravel on the conveyor belt according to the instructions sent by the server when anomalies such as abnormal gravel particle size are detected. The camera is used to collect the gravel image information on the conveyor belt and transmit the collected image information to the server.

[0033] Through the above system architecture, the anomaly detection system can, according to the image information with gravel collected by the camera, segment and extract the features of the gravel particles in the image through image processing algorithms, thereby calculate the particle size of the gravel based on the extracted features and determine whether there is an anomaly in the gravel particle size, and when abnormal gravel is detected, sort out the abnormal gravel through the robotic arm.

[0034] In the related art, since the gravel cannot be completely laid flat, some gravel particles will stack and block each other, resulting in an increased difficulty in segmenting the gravel particles during image recognition, and the stacked and blocked gravel particles cannot be accurately segmented, causing deviations in feature extraction based on the segmentation results, affecting the accuracy of particle size calculation, and leading to deviations in the anomaly detection results.

[0035] By using the gravel particle size anomaly detection method in the embodiments of the present application, after calculating the gravel particle size of the gravel through image recognition, it is determined whether there is a situation of gravel stacking by judging whether the calculated gravel particle size exceeds a preset threshold. Gravel stacking will cause the calculated gravel particle size to be the combined particle size of multiple stacked gravel particles. When there is a situation of gravel stacking, the stacked gravel is evenly distributed by the vibration device, and then the gravel particle size of the gravel is recalculated through image recognition to ensure that the calculated gravel particle size is the actual particle size of a single gravel particle, rather than the combined particle size of multiple stacked gravel particles. Anomaly detection is carried out according to the recalculated gravel particle size, reducing the false detection caused by gravel stacking and effectively improving the accuracy of the gravel particle size anomaly detection results.

[0036] The following is combined with Figure 2 to illustrate the method of the embodiment of the present application.

[0037] Please refer to Figure 2 , which is a schematic flow diagram of the gravel particle size abnormal detection method in the embodiment of the present application.

[0038] S201. Calculate the target feeding amount of the vibrating feeder based on the running speed of the conveyor belt, the preset width of the conveyor belt, and the target particle size range of the preset normal gravel.

[0039] Specifically, first calculate the moving area of the conveyor belt per unit time according to the running speed of the conveyor belt and the preset width of the conveyor belt. Then, calculate the volume of the gravel on the conveyor belt per unit time according to the moving area and the average thickness of the gravel in a flat state. Then, calculate the volume of a single gravel according to the preset gravel volume calculation formula and the maximum value of the target particle size range of the preset normal gravel. If the target feeding amount is the number of gravel particles put on the conveyor belt by the vibrating feeder per unit time, divide the volume of the gravel on the conveyor belt per unit time by the volume of a single gravel to obtain the target feeding amount. If the target feeding amount is the mass of the gravel put on the conveyor belt by the vibrating feeder per unit time, multiply the bulk density of the gravel by the volume of the gravel on the conveyor belt per unit time to obtain the target feeding amount.

[0040] Among them, the average thickness of the gravel in a flat state is determined according to the empirical data in previous similar projects or production practices, pre-stored in the database, associated with the gravel type and the target particle size range, and the corresponding average thickness can be directly obtained according to the gravel type and the target particle size range during the calculation process. The bulk density of the gravel is the mass per unit volume of the gravel in the natural stacking state, obtained by experimental measurement and pre-stored in the database.

[0041] S202. Determine the vibration amplitude and vibration frequency of the vibrating feeder according to the target feeding amount.

[0042] Based on the preset vibration parameter table, find the vibration amplitude and vibration frequency of the vibrating feeder corresponding to the target feeding amount in the table.

[0043] Among them, the vibration parameter table is pre-determined through experiments or historical data, and stores the target feeding amount and the corresponding vibration amplitude and vibration frequency of the vibrating feeder.

[0044] In some embodiments, a mathematical model can be established, input the feeding amount into the mathematical model, and obtain the vibration amplitude and vibration frequency of the vibrating feeder through the output result of the model.

[0045] If the feed rate has a linear relationship with the vibration amplitude and vibration frequency of the vibrating feeder, regression analysis is carried out using a large amount of existing target feed rate and its corresponding vibration amplitude and vibration frequency data to establish a mathematical model between the target feed rate, vibration amplitude, and vibration frequency. For example, methods such as linear regression and polynomial regression can be used. By analyzing the linear relationship between the target feed rate, vibration amplitude, and vibration frequency, a regression equation is established, and then the target feed rate is substituted into the equation to obtain the vibration amplitude and vibration frequency.

[0046] If the feed rate has a non-linear relationship with the vibration amplitude and vibration frequency of the vibrating feeder, a non-linear regression algorithm (such as polynomial regression, neural network regression, support vector regression, etc.) is used to construct a mathematical model. Taking the neural network regression algorithm as an example, a large amount of target feed rate and its corresponding vibration amplitude and vibration frequency data are used as training samples to train the neural network, enabling the neural network to learn the complex non-linear mapping relationship between the feed rate, vibration amplitude, and vibration frequency. After training, the target feed rate is input into the trained neural network model, and the model can output the corresponding vibration amplitude and vibration frequency.

[0047] S203. Control the vibrating feeder to vibrate with a vibration amplitude and vibration frequency, so that the sand and gravel in the vibrating feeder are conveyed to the conveyor belt according to the target feed rate.

[0048] According to the vibration amplitude and vibration frequency of the vibrating feeder, send corresponding control instructions to the vibrating feeder. After receiving the control instructions, the vibrating feeder vibrates with the vibration amplitude and vibration frequency, and evenly discharges the sand and gravel onto the conveyor belt according to the target feed rate.

[0049] S204. Obtain the first sand and gravel image of the target detection area of the conveyor belt.

[0050] Collect the first sand and gravel image of the target detection area of the conveyor belt through a camera. Among them, the first sand and gravel image shows multiple sand and gravel. The target detection area is one of multiple detection areas on the conveyor belt, and the detection area is obtained by dividing the area on the conveyor belt according to a certain length.

[0051] S205. Identify the first sand and gravel particle size of multiple sand and gravel in the target detection area according to the first sand and gravel image.

[0052] Specifically, first preprocess the first sand and gravel image. The preprocessing includes operations such as grayscale processing, filtering and noise reduction, and image enhancement. Then, use an edge detection algorithm (such as the Canny algorithm, etc.) to extract the edge contours of multiple sand and gravels in the image, and then use a contour recognition algorithm to convert the edge contour information obtained by edge detection into specific contours one by one. These contours are represented in the form of point sets, precisely outlining the shape of each sand and gravel. Finally, based on the obtained sand and gravel contours and combined with the image calibration parameters, calculate the particle size of each sand and gravel through a preset calculation method, so as to obtain the first sand and gravel particle size of multiple sand and gravels in the target detection area.

[0053] Among them, the image calibration parameters are a set of parameters used to establish the correspondence between image pixels and actual physical dimensions, mainly including the internal parameters of the camera (such as focal length, principal point coordinates, etc.) and external parameters (such as rotation matrix, translation vector), which are the key data for realizing the accurate conversion from image measurement to actual physical quantity measurement.

[0054] S206. When there is a target sand and gravel particle size in the first sand and gravel particle size that exceeds the preset threshold, determine the target vibration amplitude and target vibration frequency of the target vibration device corresponding to the target detection area according to the maximum value of the target particle size range, the target sand and gravel particle size, and the sand and gravel type.

[0055] Among them, since the stacking of sand and gravel will make the calculated sand and gravel particle size the combined particle size of multiple stacked sand and gravel particles, it can be judged that there is a stacking situation of sand and gravel when the sand and gravel particle size exceeds the preset threshold. The preset threshold is greater than the maximum value of the target particle size range, which can be a critical value obtained by artificially simulating the stacking situation of sand and gravel to clearly distinguish the normal particle size and the abnormally increased particle size caused by stacking, or a value obtained by statistical analysis based on a large amount of historical data that exceeds the normal particle size mean plus several times the standard deviation. There is no limitation here.

[0056] Specifically, when there is a target sand and gravel particle size in the first sand and gravel particle size that exceeds the preset threshold, use a preset vibration adjustment model to calculate the target vibration amplitude and target vibration frequency of the target vibration device. This vibration adjustment model is constructed based on a machine learning model (such as neural network, decision tree regression model, support vector regression model, etc.).

[0057] Taking the neural network as an example, construct a deep learning model based on the neural network as the vibration adjustment model.

[0058] Before training the model, collect a large number of data samples of different sand and gravel types, different target particle size ranges, and different degrees of particle size abnormality (that is, the degree to which the target sand and gravel particle size exceeds the preset threshold). These samples contain the actual values of vibration amplitude and vibration frequency that can evenly separate the stacked sand and gravels in various situations.

[0059] When training the model, the neural network model is trained with these data samples. During the training process, the model continuously adjusts its own weights and biases to minimize the error between the predicted vibration amplitude and vibration frequency and the actual labels, so that the trained model has good generalization ability and can output the vibration parameters required to evenly separate the stacked sand and gravel under different particle size anomaly degrees and sand and gravel characteristics according to parameters such as the maximum value of the input target particle size range, the target sand and gravel particle size, and the sand and gravel type.

[0060] When there is a target sand and gravel particle size exceeding the preset threshold in the first sand and gravel particle size, the maximum value of the target particle size range, the target sand and gravel particle size, and the sand and gravel type are used as input parameters and input into the trained vibration adjustment model. The model performs computational processing such as weighted summation and non-linear transformation on the input parameters through internal multi-layer neurons, and outputs the target vibration amplitude and target vibration frequency of the target vibration device corresponding to the target detection area.

[0061] In some embodiments, in order to more accurately determine whether there is a situation of sand and gravel stacking and avoid misjudgment and missed judgment, techniques such as morphological analysis and contour detection in image recognition can be combined for further judgment. The sand and gravel image is preprocessed using operations such as morphological opening and closing operations to highlight the sand and gravel contour, and then the contour detection algorithm is used to accurately outline the shape of each sand and gravel. If it is detected that there is a situation where the contours intersect or partially overlap, it is determined that there is a sand and gravel stacking situation.

[0062] In some embodiments, an instance segmentation algorithm based on deep learning (such as Mask R-CNN, YOLACT, BlendMask, SOLO, etc.) can also be combined to accurately segment and identify each sand and gravel in the image, and further determine whether there is a sand and gravel stacking situation by analyzing the boundary and position relationship of the sand and gravel in the segmentation result. Among them, the instance segmentation algorithm is a computer vision algorithm that combines object detection and semantic segmentation techniques, aiming to identify each independent object instance in the image and accurately segment the corresponding pixel region for it, and can accurately distinguish different individuals of the same category.

[0063] S207. Send corresponding control instructions to the target vibration device according to the target vibration amplitude and target vibration frequency.

[0064] Generate corresponding control instructions according to the target vibration amplitude and target vibration frequency, and send the control instructions to the target vibration device. After receiving the control instructions, the target vibration device vibrates according to the target vibration amplitude and target vibration frequency.

[0065] S208. In the case where the target vibration device completes the vibration operation according to the control instructions, identify the second sand and gravel particle size of the sand and gravel in the target detection area according to the second sand and gravel image of the re-acquired target detection area.

[0066] Step S208 is similar to the above-mentioned steps S204 - S205. For details, please refer to the descriptions in steps S204 - S205 and will not be elaborated here.

[0067] S209. Perform anomaly detection on each piece of sand and gravel in the target detection area according to the second sand and gravel particle size to obtain the anomaly detection result of the target detection area.

[0068] Among them, the anomaly detection is to determine whether the particle size of the sand and gravel exceeds the target particle size range of the preset normal sand and gravel. The anomaly detection result includes the total number of all sand and gravel in the target detection area, the total number of abnormal sand and gravel, the particle size of each piece of sand and gravel, the position information of each piece of sand and gravel in the target detection area, and the detection result. The detection result is divided into qualified and abnormal. The abnormal sand and gravel are those whose particle size exceeds the target particle size range.

[0069] Specifically, determine whether the second particle size of each piece of sand and gravel in the target detection area exceeds the target particle size range of the preset normal sand and gravel. If the second particle size of the sand and gravel does not exceed the target particle size range, then determine that the sand and gravel is qualified; if the second particle size of the sand and gravel exceeds the target particle size range, then determine that the sand and gravel is abnormal. During the anomaly detection process, for each piece of sand and gravel subjected to anomaly detection, record its relevant information, including its particle size, position information in the target detection area, and detection result, etc. When the detection of all sand and gravel in the target detection area is completed, count the qualified sand and gravel and the abnormal sand and gravel respectively, and count the total number of all sand and gravel in the target detection area and the total number of abnormal sand and gravel. Finally, perform information aggregation to generate the anomaly detection result of the target detection area.

[0070] Among them, the position information of the sand and gravel is obtained through an image recognition algorithm. First, preprocess the image through the image recognition algorithm, then use the edge detection algorithm to outline the contour of the sand and gravel, and then use the contour analysis algorithm to determine the geometric center or centroid of each sand and gravel contour as the image position coordinates of the sand and gravel in the image. Finally, convert the image position coordinates into the actual physical position in the target detection area through the coordinate transformation formula to obtain the position information of the sand and gravel in the target detection area.

[0071] In some embodiments, when abnormal sand and gravel are detected, the robotic arm can be controlled to remove the abnormal sand and gravel to a designated area. The position information of the abnormal sand and gravel is sent to the robotic arm. After receiving the position information, the robotic arm automatically plans a movement path and transports the abnormal sand and gravel to the designated collection area according to the movement path. Among them, there can be one or more collection areas, and each area is divided according to the particle size of the sand and gravel, so that the particle size of the sand and gravel in each collection area is within the same particle size range, which is convenient for subsequent classification processing of the abnormal sand and gravel. For example, for sand and gravel in different particle size ranges, different processing techniques can be used to reuse them, improving the utilization rate of sand and gravel resources.

[0072] In the embodiments of the present application, after calculating the particle size of the sand and gravel through image recognition, it is determined whether there is a situation of sand and gravel stacking by judging whether the particle size of the sand and gravel exceeds a preset threshold. When there is a situation of sand and gravel stacking, the stacked sand and gravel are evenly distributed by a vibration device, and then the particle size of the sand and gravel is recalculated through image recognition to ensure that the calculated particle size of the sand and gravel is the actual particle size of a single sand and gravel particle, rather than the combined particle size of multiple stacked sand and gravel particles, improving the accuracy of the particle size calculation of the sand and gravel. Abnormal detection is performed based on the recalculated particle size of the sand and gravel, reducing false detection caused by sand and gravel stacking, and effectively improving the accuracy of the abnormal detection result of the particle size of the sand and gravel.

[0073] The following combines Figure 3 to further illustrate the method of the embodiments of the present application.

[0074] Please refer to Figure 3 , which is another process schematic diagram of the method for detecting abnormal particle size of sand and gravel in the embodiments of the present application.

[0075] S301. Calculate the target feeding amount of the vibrating feeder.

[0076] Step S301 is similar to step S201 in the Figure 2 illustrated embodiment, and reference can be made to the description in step S201, which will not be elaborated here.

[0077] S302. Calculate the maximum sand and gravel mass according to the sand and gravel density corresponding to the sand and gravel type and the maximum value of the target particle size range.

[0078] According to the maximum value of the target particle size range and the sand and gravel density corresponding to the sand and gravel type, the maximum sand and gravel mass is calculated through the mass calculation formula.

[0079] S303. Obtain the corresponding feeding amount table according to the sand and gravel type and the maximum sand and gravel mass.

[0080] Search the database for the feeding amount table corresponding to the sand and gravel type and the maximum sand and gravel mass.

[0081] Among them, there is one or more feed quantity tables, and each feed quantity table corresponds to a unique type of sand and gravel and the quality of sand and gravel. The table includes the feed quantity and the vibration amplitude and vibration frequency of the corresponding vibrating feeder.

[0082] S304. Search for the vibration amplitude and vibration frequency of the vibrating feeder corresponding to the target feed quantity in the feed quantity table.

[0083] Based on the target feed quantity, search for the vibration amplitude and vibration frequency of the vibrating feeder corresponding to the feed quantity identical to the target feed quantity in the feed quantity table.

[0084] S305. Control the vibrating feeder to vibrate with the vibration amplitude and vibration frequency.

[0085] S306. Obtain the first sand and gravel image of the target detection area of the conveyor belt.

[0086] S307. Identify the first sand and gravel particle sizes of multiple sand and gravel in the target detection area according to the first sand and gravel image.

[0087] Steps S305 - S307 are similar to Figure 2 Steps S203 - S205 in the illustrated embodiment. Refer to the descriptions in steps S203 - S205, and details are not described herein again.

[0088] S308. In the case where there is a target sand and gravel particle size exceeding the preset threshold among the first sand and gravel particle sizes, obtain the amplitude basic calculation parameters and vibration frequency basic calculation parameters corresponding to the sand and gravel type and the target particle size range.

[0089] Obtain the amplitude basic calculation parameters and vibration frequency basic calculation parameters corresponding to the sand and gravel type and the target particle size range pre - stored in the database.

[0090] Among them, the amplitude basic calculation parameters include the basic vibration amplitude, amplitude influence coefficient, amplitude correction coefficient, standard sand and gravel quality, sand and gravel density, sand and gravel elastic modulus, and sand and gravel Poisson's ratio. The vibration frequency basic calculation parameters include the basic vibration frequency, vibration frequency influence coefficient, vibration frequency correction coefficient, standard sand and gravel quality, sand and gravel elastic modulus, and sand and gravel Poisson's ratio.

[0091] S309. Calculate the sand and gravel quality according to the sand and gravel density and the target particle size range.

[0092] Obtain the intermediate value of the target particle size range as the average sand and gravel particle size, and calculate the sand and gravel quality through the mass calculation formula according to the average sand and gravel particle size and the sand and gravel density corresponding to the sand and gravel type.

[0093] S310. Calculate the target vibration amplitude of the target vibration device corresponding to the target detection area according to the vibration amplitude calculation formula.

[0094] Substitute the amplitude-based calculation parameters, the target sand and gravel particle size, the maximum value of the target particle size range, and the sand and gravel mass into the vibration amplitude calculation formula to obtain the target vibration amplitude of the target vibration device corresponding to the target detection area.

[0095] Among them, the vibration amplitude calculation formula is: Among them, A is the target vibration amplitude, A0 is the basic vibration amplitude, d is the target sand and gravel particle size, d max is the maximum value of the target particle size range, k1 is the amplitude influence coefficient, m is the sand and gravel mass, m0 is the standard sand and gravel mass, ρ is the sand and gravel density, E is the sand and gravel elastic modulus, v is the sand and gravel Poisson's ratio, and k3 is the amplitude correction coefficient.

[0096] In the vibration amplitude calculation formula, A0 is an initial vibration amplitude value determined based on the design parameters of the vibration device and tests under standard working conditions. It can be understood as the initial vibration amplitude benchmark set to enable the vibration device to effectively act on the sand and gravel under the most ideal and standard sand and gravel treatment conditions (such as standard sand and gravel particle size, fixed environmental temperature and humidity, specific initial equipment state, etc.).

[0097] In the formula part, is the square of the difference between the target sand and gravel particle size d (representing the total particle size of the stacked sand and gravel, reflecting the stacking degree of the sand and gravel, and d is greater than d max ) and the maximum value d of the target particle size range max , which reflects the influence of the sand and gravel stacking degree on the vibration amplitude. When the sand and gravel stacking degree is higher, d is relatively larger than d max , the interaction and friction between particles are greater, the dispersion difficulty increases, and a larger vibration amplitude is required to achieve uniform dispersion of the sand and gravel. k1 is the amplitude influence coefficient, which is an empirical parameter obtained through a large number of experiments and the accumulation of actual working condition data. Different types of sand and gravel, such as quartz sand, river sand, etc., have different characteristics such as particle shape, hardness, and surface roughness, and their motion characteristics during vibration are also different. It can reasonably adjust the influence weight of the difference in sand and gravel stacking degree on the vibration amplitude according to these differences in sand and gravel characteristics and factors such as vibration equipment type and working environment in the actual working conditions. Since the target vibration amplitude A is increased and adjusted on the basis of the basic vibration amplitude A0, the value is the amplitude that needs to be adjusted on the basis of A0. Therefore, when calculating, it needs to be calculated by adding 1 to .

[0098] In the formula part, It reflects the comprehensive influence of physical properties such as the difference between the sand and gravel quality m and the standard sand and gravel quality m0 (the sand and gravel quality m is greater than or equal to the standard sand and gravel quality m0), the sand and gravel density ρ, the elastic modulus E, and the Poisson's ratio v on the vibration amplitude. The greater the sand and gravel quality, the greater the inertia, and a greater vibration amplitude is required to drive it; while the density, elastic modulus, and Poisson's ratio of the sand and gravel reflect its internal structure and mechanical response characteristics, and these characteristics will affect the force and deformation of the sand and gravel during vibration. The form of the logarithmic function makes the adjustment of the vibration amplitude show a reasonable gradualness with the changes of these physical property factors. When the sand and gravel quality is the same as the standard quality and other physical properties are in a certain reference state, this part has basically no influence on the vibration amplitude; as the difference in sand and gravel quality increases or the physical properties change, the value of this part will change accordingly, so as to adjust the basic vibration amplitude to meet the requirements of different quality and physical property sand and gravel during vibration. Generally, the physical properties such as the sand and gravel density, elastic modulus, and Poisson's ratio of the sand and gravel are fixed.

[0099] k3 is the amplitude correction coefficient, which takes into account factors not reflected in other parts of the formula in the actual working conditions, such as the humidity of the sand and gravel, the wear of the vibration device, the environmental temperature, etc. The actual sand and gravel processing environment is complex and changeable, and these factors will affect the vibration effect of the sand and gravel and the required vibration amplitude. By finely adjusting the vibration amplitude through the amplitude correction coefficient, the calculated target vibration amplitude can be made more suitable for the actual working conditions.

[0100] Generally speaking, the entire vibration amplitude calculation formula comprehensively considers various factors such as sand and gravel characteristics, physical properties, and actual working conditions, and can scientifically and accurately calculate the target vibration amplitude that meets the requirement of evenly dispersing the stacked sand and gravel.

[0101] S311. Calculate the target vibration frequency of the target vibration device corresponding to the target detection area according to the vibration frequency calculation formula.

[0102] Substitute the vibration frequency basic calculation parameters, the target sand and gravel particle size, the maximum value of the target particle size range, and the sand and gravel quality into the vibration frequency calculation formula to obtain the target vibration frequency of the target vibration device corresponding to the target detection area.

[0103] The vibration frequency calculation formula is: Among them, f is the target vibration frequency, f0 is the basic vibration frequency, d is the target sand and gravel particle size, d max is the maximum value of the target particle size range, k2 is the vibration frequency influence coefficient, m is the sand and gravel quality, m0 is the standard sand and gravel quality, E is the sand and gravel elastic modulus, v is the sand and gravel Poisson's ratio, and k4 is the vibration frequency correction coefficient.

[0104] In the vibration frequency calculation formula, f0 is the starting vibration frequency value determined based on the design of the vibration device and tested under standard working conditions.

[0105] In the formula part, reflects the influence of the stacking degree of sand and gravel on the vibration frequency. When the stacking degree of sand and gravel is higher, the interaction and friction between particles are greater, and a higher vibration frequency is required to achieve uniform dispersion of sand and gravel. k2 is the vibration frequency influence coefficient, which is an empirical parameter obtained through a large number of experiments and the accumulation of actual working condition data. It can reasonably adjust the influence weight of the difference in the stacking degree of sand and gravel on the vibration frequency according to the differences in the characteristics of sand and gravel and the actual working conditions. Since the target vibration frequency f is increased and adjusted based on the basic vibration frequency f0, the value is the adjustment amplitude required on the basis of f0. Therefore, when calculating, it is necessary to add 1 on the basis of for calculation.

[0106] In the formula part, reflects the comprehensive influence of the differences between the quality of sand and gravel and the quality of standard sand and gravel, physical properties such as elastic modulus and Poisson's ratio on the vibration frequency. The form of the arctangent function makes the adjustment of the vibration frequency show a reasonable range and gradualness with the changes of these factors. When the quality and physical properties of sand and gravel are in the reference state, this part has little influence on the vibration frequency; as the quality difference and physical properties change, it will adjust the basic vibration frequency to adapt to the frequency requirements of different sand and gravel during vibration.

[0107] k4 is the amplitude correction coefficient, which takes into account factors not reflected in other parts of the formula in the actual working conditions, such as the humidity of sand and gravel, wear of the vibration device, environmental temperature, etc. The actual sand and gravel processing environment is complex and changeable, and these factors will affect the vibration effect of sand and gravel and the required vibration frequency. By finely adjusting the vibration frequency through the amplitude correction coefficient, the calculated target vibration frequency can be made more in line with the actual working conditions.

[0108] Generally speaking, the entire vibration frequency calculation formula comprehensively considers various factors such as the characteristics of sand and gravel, physical properties, and actual working conditions, and can scientifically and accurately calculate the target vibration frequency that meets the requirement of uniformly dispersing the stacked sand and gravel.

[0109] S312. According to the target vibration amplitude and target vibration frequency, send corresponding control commands to the target vibration device.

[0110] S313. In the case where the target vibration device completes the vibration operation according to the control command, identify the second sand and gravel particle size in the target detection area based on the second sand and gravel image of the re-acquired target detection area.

[0111] S314. Perform anomaly detection on each gravel in the target detection area according to the second gravel particle size to obtain the anomaly detection result of the target detection area.

[0112] Steps S312 - S314 are similar to Figure 2 Steps S207 - S209 in the embodiment shown. Refer to the descriptions in Steps S207 - S209, and details are not repeated here.

[0113] S315. Based on the gravel - free image of the target detection area collected, identify the actual coordinate information of each preset point in the image.

[0114] Specifically, collect the gravel - free image of the target detection area through a camera, then identify the coordinate information of each preset point in the gravel - free image through an image recognition algorithm, and then based on the mapping relationship established in advance between the image coordinate system and the actual physical coordinate system, through corresponding mathematical transformations and calibration parameters, convert the image coordinates into actual coordinates to obtain the actual coordinate information of each preset point. Among them, the coordinates are two - dimensional coordinates.

[0115] S316. Calculate the offset of each preset point according to the preset coordinate information of each preset point and the actual coordinate information of each preset point.

[0116] According to the preset coordinate information of each preset point and the actual coordinate information of each preset point, calculate the coordinate distance between the preset coordinate and the actual coordinate of each preset point according to the distance formula between two points as the offset. Among them, the coordinate systems of the preset coordinate and the actual coordinate are the same.

[0117] S317. Calculate the flatness of each sub - area in the target detection area according to the offset of one or more preset points in each sub - area of the target detection area.

[0118] Specifically, divide the target detection area into multiple sub - areas according to the preset area division rules, and each sub - area has a corresponding position range in the target detection area. Traverse the preset coordinate information of each preset point, and judge the sub - area where the preset point is located according to the position range of the sub - area. After traversing all the preset coordinate information of the preset points, sequentially obtain the offsets of the preset points in each sub - area. If there is only one preset point in the sub - area, use the offset of this preset point as the flatness of this sub - area. If there are two or more preset points in the sub - area, calculate the average value of all offsets, then calculate the sum of the squared deviations of the offset of each preset point from the average value, and finally calculate the standard deviation of the offset as the flatness of this sub - area according to the sum of the squared deviations and the number of preset points.

[0119] Among them, the value of flatness is greater than or equal to 0. The closer the flatness is to 0, the flatter the surface of the area is, and the more accurate the calculated sand and gravel particle size in the area is, and the closer it is to the actual sand and gravel particle size.

[0120] S318: When one or more flatnesses among all the flatnesses are greater than a preset flatness threshold, calculate the percentage of the smaller sum of each sub-region.

[0121] Specifically, when there is one or more flatnesses greater than a preset flatness threshold among all flatnesses, based on the abnormal detection results of the target detection area, the position information of each gravel in the abnormal detection results in the target detection area is traversed, and the sub-area where the gravel is located is determined according to the position range of the sub-area. After traversing the position information of all gravel, the gravel information of each sub-area is obtained in sequence. According to the gravel information and the target gravel particle size range, the total number of gravel with a gravel particle size greater than the maximum value of the target particle size range, the total number of gravel with a gravel particle size less than the minimum value of the target particle size range, and the total number of abnormal gravel are counted. The ratio of the total number of gravel with a gravel particle size greater than the maximum value of the target particle size range to the total number of abnormal gravel is calculated in turn in each sub-area as the larger proportion value. The ratio of the total number of gravel with a gravel particle size less than the minimum value of the target particle size range to the total number of abnormal gravel is calculated in turn in each sub-area as the smaller proportion value.

[0122] The sand and gravel information includes the sand and gravel particle size of one or more sand and gravel and its position information in the target detection area.

[0123] S319: When the flatness of the target sub-region is less than a preset flatness threshold, screen out the sand and gravel in the target sub-region whose particle size exceeds the target particle size range.

[0124] Specifically, when the flatness of the target sub-area is less than the preset flatness threshold, the position information set of the sand and gravel whose particle size exceeds the target particle size range is obtained according to the sand and gravel information of the target sub-area. The position information set and the sand and gravel particle size corresponding to each position are sent to the robotic arm. After receiving the position information set, the robotic arm takes its own initial position as the starting position, the position in the position information set as the first arrival position, the collection area corresponding to the position as the second arrival position, and the initial position of the robotic arm as the terminal position, and automatically plans multiple motion paths from the starting position to the first arrival position, then to the second arrival position, and finally to the terminal position. According to the motion path, the sand and gravel are transported to the designated collection area in sequence. Among them, the target sub-area is one of all sub-areas. Each position in the position information set corresponds to a collection area, and the collection area is determined by the sand and gravel particle size corresponding to each position.

[0125] S320. When the flatness of the target sub-region is greater than the preset flatness threshold and the oversize ratio of the target sub-region is greater than the undersize ratio, screen out the gravel in the target sub-region with a particle size smaller than the minimum value of the target particle size range.

[0126] Specifically, when the flatness of the target sub-region is greater than the preset flatness threshold and the oversize ratio of the target sub-region is greater than the undersize ratio, according to the gravel information of the target sub-region, obtain the set of position information of the gravel with a particle size smaller than the minimum value of the target particle size range. Send the set of position information and the gravel particle size corresponding to each position to the robotic arm. After receiving the set of position information, the robotic arm takes its own initial position as the starting position, the positions in the set of position information as the first arrival positions, the collection area corresponding to this position as the second arrival position, and its own initial position as the end position, and automatically plans multiple movement paths from the starting position to the first arrival position, then to the second arrival position, and finally to the end position. According to the movement paths, sequentially transport the gravel to the designated collection area.

[0127] S321. When the flatness of the target sub-region is greater than the preset flatness threshold and the oversize ratio of the target sub-region is less than the undersize ratio, screen out the gravel in the target sub-region with a particle size greater than the maximum value of the target particle size range.

[0128] Specifically, when the flatness of the target sub-region is greater than the preset flatness threshold and the oversize ratio of the target sub-region is less than the undersize ratio, according to the gravel information of the target sub-region, obtain the set of position information of the gravel with a particle size greater than the maximum value of the target particle size range. Send the set of position information and the gravel particle size corresponding to each position to the robotic arm. After receiving the set of position information, the robotic arm takes its own initial position as the starting position, the positions in the set of position information as the first arrival positions, the collection area corresponding to this position as the second arrival position, and its own initial position as the end position, and automatically plans multiple movement paths from the starting position to the first arrival position, then to the second arrival position, and finally to the end position. According to the movement paths, sequentially transport the gravel to the designated collection area.

[0129] S322. Convey the other gravel in the screened target detection area to the area of unscreened gravel through a conveyor belt.

[0130] After the robotic arm has transported all the gravel corresponding to the set of position information to the designated collection area, convey the remaining gravel in the detection area to the area of unscreened gravel through a conveyor belt. Among them, the area of unscreened gravel stores the unscreened gravel that has not been subjected to abnormal detection, and the unscreened gravel will be transported to the vibrating feeder and undergo abnormal detection according to the process.

[0131] S323. After detecting all the sand and gravel to be detected, obtain the particle sizes of all the abnormal sand and gravel to obtain a set of abnormal sand and gravel particle sizes.

[0132] Specifically, after detecting all the sand and gravel to be detected, obtain all the abnormal detection results and summarize them into a sand and gravel information set. Search for the particle sizes of the sand and gravel with abnormal detection results in the sand and gravel information set to generate a set of abnormal sand and gravel particle sizes for all the abnormal sand and gravel.

[0133] Among them, the sand and gravel information set includes the particle sizes of all the sand and gravel to be detected and their detection results. The set of abnormal sand and gravel particle sizes includes the particle sizes of all the abnormal sand and gravel and the abnormal levels.

[0134] S324. Determine the abnormal levels of each abnormal sand and gravel according to the set of abnormal sand and gravel particle sizes.

[0135] Specifically, obtain the common sand and gravel particle size table corresponding to the sand and gravel type, traverse the particle sizes of each abnormal sand and gravel in the set of abnormal sand and gravel particle sizes, and determine whether the particle size is in the common sand and gravel particle size table. If the particle size is in the common sand and gravel particle size table, it is determined that the abnormal level of the abnormal sand and gravel corresponding to the particle size is level one, and the abnormal level of the abnormal sand and gravel is modified to level one; if the particle size is not in the common sand and gravel particle size table, it is determined that the abnormal level of the abnormal sand and gravel corresponding to the particle size is level two, and the abnormal level of the abnormal sand and gravel is modified to level two.

[0136] Among them, the abnormal levels include level-one abnormality and level-two abnormality. Level-one abnormality means that the particle size of the abnormal sand and gravel is in the common sand and gravel particle size table, indicating that the abnormal sand and gravel can be directly used for the production of other concretes. Level-two abnormality means that the particle size of the abnormal sand and gravel is not in the common sand and gravel particle size table, indicating that the abnormal sand and gravel needs to be processed before it can be used for the production of other concretes or cannot be used for concrete production.

[0137] S325. Obtain the first quantity of all the abnormal sand and gravel with the abnormal level of level-one abnormality and the second quantity of all the abnormal sand and gravel with the abnormal level of level-two abnormality.

[0138] According to the set of abnormal sand and gravel particle sizes, count the first quantity of the abnormal sand and gravel with the abnormal level of level one and the second quantity of the abnormal sand and gravel with the abnormal level of level two.

[0139] S326. Calculate the first proportion value and the second proportion value according to the total quantity, the first quantity, and the second quantity of the abnormal sand and gravel.

[0140] Specifically, according to the set of abnormal sand and gravel particle sizes, count the total quantity of the abnormal sand and gravel. Calculate the value of the first quantity divided by the total quantity of the abnormal sand and gravel as the first proportion value, and calculate the value of the second quantity divided by the total quantity of the abnormal sand and gravel as the second proportion value.

[0141] S327. Determine the quality of the sand and gravel according to the qualified rate of the sand and gravel, the first proportion value and the second proportion value.

[0142] Specifically, according to the sand and gravel information set, the total number of all sand and gravel to be detected is counted. According to the total number of abnormal sand and gravel and the total number of all sand and gravel to be detected, the total number of all qualified sand and gravel is calculated. The total number of all qualified sand and gravel is divided by the total number of all sand and gravel to be detected to obtain the sand and gravel qualified rate.

[0143] If the qualified rate of sand and gravel is greater than the preset qualified rate threshold, and the first proportion value is greater than the second proportion value, it is determined whether the first proportion value is greater than the preset threshold. If so, the quality of the sand and gravel is determined to be high quality; if not, the quality of the sand and gravel is determined to be medium quality.

[0144] If the qualified rate of sand and gravel is greater than the preset qualified rate threshold, and the first proportion value is less than the second proportion value, it is determined whether the second proportion value is less than the preset threshold value. If so, the quality of the sand and gravel is determined to be low quality; if not, the quality of the sand and gravel is determined to be medium quality.

[0145] If the sand and gravel pass rate is less than the preset pass rate threshold, the sand and gravel quality is judged to be low quality.

[0146] Among them, qualified sand and gravel are those whose particle size is within the target particle size range. Sand and gravel quality is divided into high quality, medium quality and low quality.

[0147] In the embodiment of the present application, it is determined whether there is sand and gravel stacking by judging whether the sand and gravel particle size exceeds a preset threshold. When there is sand and gravel stacking, the stacked sand and gravel are evenly distributed through a vibration device, and then the sand and gravel particle size of the sand and gravel is recalculated through image recognition to ensure that the calculated sand and gravel particle size is the actual particle size of a single sand and gravel particle, thereby improving the accuracy of sand and gravel particle size calculation. Abnormal detection is performed based on the recalculated sand and gravel particle size, reducing false detections caused by sand and gravel stacking, and can effectively improve the accuracy of sand and gravel particle size abnormality detection results. At the same time, when the abnormal detection results of sand and gravel particle size are obtained, the accuracy of the abnormal detection results is further determined by calculating the flatness of each sub-area in the detection area. When the flatness of a sub-area is greater than the preset threshold, it means that the collected sand and gravel image is uneven due to the sub-area, which affects the accuracy of the sand and gravel particle size obtained by image recognition calculation, and then affects the accuracy of the abnormal detection results of sand and gravel particle size. At this time, according to the flatness, large proportion value and small proportion value of each sub-area, the sand and gravel that can be determined as abnormal are screened out, and other sand and gravel are re-detected for abnormalities to avoid misjudgment caused by local unevenness and improve the accuracy of the abnormal detection results of sand and gravel particle size. After all sand and gravel are tested, the quality of the sand and gravel is determined according to the abnormal detection results, which provides a basis for users to choose high-quality sand and gravel suppliers.

[0148] The above describes the method for detecting abnormal sand and gravel particle sizes in the embodiments of the present application. Below, in combination with the above method for detecting abnormal sand and gravel particle sizes, the abnormal detection system in the embodiments of the present application will be described in detail.

[0149] Please refer to Figure 4 , which is an exemplary hardware structure diagram of the abnormal detection system in the embodiments of the present application.

[0150] In some embodiments, the abnormal detection system 400 includes a computer device, which may be a terminal device. The computer device includes a processor 401, a memory 402, a communication module 403, an input device 404, and an output device 405 connected through a system bus. Among them, the processor 401 of the computer device is used to provide computing and control capabilities. The memory 402 of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database is used to store data. The communication module 403 of the computer device is used to send control instructions to the vibrating feeder, the vibrating device, and the robotic arm. The input device 404 of the computer device is used to receive image information transmitted by the camera, etc. The output device 405 of the computer device is used to display the abnormal detection results, the quality of sand and gravel, etc. When the computer program is executed by the processor 401, it realizes the method for detecting abnormal sand and gravel particle sizes in the embodiments of the present application.

[0151] Those skilled in the art can understand that Figure 4 the structure shown in

[0152] is only a block diagram of some structures related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0153] In some embodiments of the present application, a computer-readable storage medium is provided, including instructions that, when run on the abnormal detection system 400, can cause the abnormal detection system 400 to execute the method for detecting abnormal sand and gravel particle sizes in the embodiments of the present application.

[0154] As described above, 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 described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the various embodiments of the present application.

[0155] In the above embodiments, depending on the context, the term "when..." can be interpreted to mean "if...", or "after...", or "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if detecting (the stated condition or event)" can be interpreted to mean "if determining...", or "in response to determining...", or "when detecting (the stated condition or event)", or "in response to detecting (the stated condition or event)".

[0156] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another, for example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line) or wireless (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that includes one or more integrated available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid-state drive), etc.

[0157] Those of ordinary skill in the art can understand all or part of the processes in the methods of the above embodiments. These processes can be completed by relevant hardware instructed by a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above method embodiments. The foregoing storage media include: ROM or random access memory RAM, magnetic disks, or optical disks and other media that can store program codes.

Claims

1. A method for detecting abnormal sand and gravel particle size, characterized in that: include: Calculate the target feeding amount of the vibrating feeder based on the running speed of the conveyor belt, the preset width of the conveyor belt, and the preset target particle size range of normal sand and gravel; The vibrating feeder is located above the conveyor belt; Determining the vibration amplitude and vibration frequency of the vibrating feeder according to the target feeding amount; Controlling the vibrating feeder to vibrate at the vibration amplitude and the vibration frequency so that the sand and gravel in the vibrating feeder are transported to the conveyor belt according to the target feeding amount; Acquire a first sand and gravel image of a target detection area of ​​the conveyor belt; The first sandstone image shows a plurality of sandstones; According to the first gravel image, identifying a first gravel particle size of a plurality of gravels in the target detection area; In the case where a target gravel particle size exceeds a preset threshold value in the first gravel particle size, a target vibration amplitude and a target vibration frequency of a target vibration device corresponding to the target detection area are determined according to the maximum value of the target particle size range, the target gravel particle size, and the gravel type; the target vibration device is located below the conveyor belt; sending corresponding control instructions to the target vibration device according to the target vibration amplitude and the target vibration frequency; When the target vibration device completes the vibration operation according to the control instruction, identifying a second sand and gravel particle size of the sand and gravel in the target detection area according to the re-collected second sand and gravel image of the target detection area; According to the second gravel particle size, performing abnormality detection on each gravel in the target detection area to obtain an abnormality detection result of the target detection area; The abnormality detection is to determine whether the particle size of the sand and gravel exceeds the target particle size range of the preset normal sand and gravel.

2. The method according to claim 1, characterized in that In the case where there is a target gravel particle size exceeding a preset threshold value in the first gravel particle size, determining the target vibration amplitude and target vibration frequency of the target vibration device corresponding to the target detection area according to the maximum value of the target particle size range, the target gravel particle size, and the gravel type, specifically includes: In the case that there is a target gravel particle size exceeding a preset threshold value in the first gravel particle size, according to the gravel type and the target particle size range, the amplitude basic calculation parameters and the frequency basic calculation parameters corresponding to the gravel type and the target particle size range are obtained; the amplitude basic calculation parameters include basic vibration amplitude, amplitude influence coefficient, amplitude correction coefficient, standard gravel mass, gravel density, gravel elastic modulus and gravel Poisson's ratio; the frequency basic calculation parameters include basic vibration frequency, frequency influence coefficient, frequency correction coefficient, standard gravel mass, gravel elastic modulus and gravel Poisson's ratio; Calculating the mass of the sand and gravel according to the sand and gravel density and the target sand and gravel particle size; Substitute the amplitude basic calculation parameters, the target gravel particle size, the maximum value of the target particle size range, and the gravel mass into the vibration amplitude calculation formula to obtain the target vibration amplitude of the target vibration device corresponding to the target detection area, and substitute the vibration frequency basic calculation parameters, the target gravel particle size, the maximum value of the target particle size range, and the gravel mass into the vibration frequency calculation formula to obtain the target vibration frequency of the target vibration device corresponding to the target detection area; wherein, the vibration amplitude calculation formula is: The vibration frequency calculation formula is: Wherein, A is the target vibration amplitude, f is the target vibration frequency, A0 is the basic vibration amplitude, f0 is the basic vibration frequency, d is the target gravel particle size, d max is the maximum value of the target particle size range, k1 is the amplitude influence coefficient, k2 is the frequency influence coefficient, m is the sand mass, m0 is the standard sand mass, ρ is the sand density, W is the sand elastic modulus, v is the sand Poisson's ratio, k3 is the amplitude correction coefficient, and k4 is the frequency correction coefficient.

3. The method according to claim 1, characterized in that Determining the vibration amplitude and vibration frequency of the vibrating feeder according to the target feeding amount specifically includes: The maximum gravel mass is calculated according to the gravel density corresponding to the gravel type and the maximum value of the target particle size range; According to the sand and gravel type and the maximum sand and gravel mass, a corresponding feeding amount table is obtained; there are one or more feeding amount tables, each of which corresponds to a unique sand and gravel type and sand and gravel mass, and the table includes the feeding amount and the corresponding vibration amplitude and vibration frequency of the vibrating feeder; The vibration amplitude and vibration frequency of the vibrating feeder corresponding to the target feeding amount are searched in the feeding amount table.

4. The method according to claim 1, characterized in that: After the step of performing abnormality detection on each gravel in the target detection area according to the second gravel particle size to obtain an abnormality detection result of the target detection area, the method further includes: Determine the flatness of each sub-area within the target detection area based on the acquired sand and stone-free image of the target detection area; When one or more of the flatnesses exceeds the preset flatness threshold, the larger proportion value and the smaller proportion value of each sub-area are calculated; the larger proportion value is the ratio of the total number of sand and gravel with a grain size larger than the maximum value of the target grain size range to the total number of abnormal sand and gravel in the sub-area; the smaller proportion value is the ratio of the total number of sand and gravel with a grain size smaller than the minimum value of the target grain size range to the total number of abnormal sand and gravel in the sub-area; Screening all the sand and gravel in each sub-area according to the flatness, the larger proportion value and the smaller proportion value of each sub-area; The other sand and gravel in the target detection area after screening are transported to the undetected sand and gravel area through the conveyor belt.

5. The method according to claim 4, characterized in that Determining the flatness of each sub-area in the target detection area based on the acquired sand and stone-free image of the target detection area specifically includes: Based on the collected sand-free image of the target detection area, the actual coordinate information of each preset point in the image is identified; Calculating the offset of each preset point according to the preset coordinate information of each preset point and the actual coordinate information of each preset point; The flatness of each sub-region is calculated according to the offset of one or more preset points in each sub-region of the target detection region.

6. The method according to claim 4, characterized in that The screening of all sand and gravel in each sub-area according to the flatness, the larger proportion value and the smaller proportion value of each sub-area specifically includes: When the flatness of the target sub-region does not exceed the preset flatness threshold, screening out the sand and gravel in the target sub-region whose particle size exceeds the target particle size range; the target sub-region is one of all the sub-regions; When the flatness of the target sub-area exceeds the preset flatness threshold, and the larger proportion of the target sub-area is greater than the smaller proportion, screening out the sand and gravel in the target sub-area whose particle size is smaller than the minimum value of the target particle size range; When the flatness of the target sub-area exceeds the preset flatness threshold and the larger proportion of the target sub-area is less than the smaller proportion, the sand and gravel in the target sub-area whose particle size is larger than the maximum value of the target particle size range is screened out.

7. The method according to claim 1, characterized in that After the step of performing abnormality detection on each gravel in the target detection area according to the second gravel particle size to obtain an abnormality detection result of the target detection area, the method further includes: After all the sand and gravel to be tested are tested, the sand and gravel particle sizes of all abnormal sand and gravel are obtained to obtain an abnormal sand and gravel particle size set; According to the abnormal gravel particle size set, the abnormal level of each abnormal gravel is determined; the abnormal level includes primary abnormality and secondary abnormality; the primary abnormality is that the gravel particle size of the abnormal gravel is in the preset common gravel particle size table; the secondary abnormality is that the gravel particle size of the abnormal gravel is not in the common gravel particle size table; Acquire a first quantity of all abnormal sands and gravels whose abnormal level is the first-level abnormality and a second quantity of all abnormal sands and gravels whose abnormal level is the second-level abnormality; According to the total number of abnormal sand and gravel, the first number and the second number, a first proportion value and a second proportion value are calculated; the first proportion value is the value of the first number divided by the total number of abnormal sand and gravel; the second proportion value is the value of the second number divided by the total number of abnormal sand and gravel; The quality of the sand and gravel is determined according to the qualified rate of sand and gravel, the first proportion value and the second proportion value; the qualified rate of sand and gravel is the value of the total number of all qualified sand and gravel divided by the total number of all sand and gravel to be tested; the qualified sand and gravel is the sand and gravel whose particle size is within the target particle size range.

8. An anomaly detection system, characterized in that: It includes a conveyor belt, a vibrating feeder, a vibrating device and a server, wherein the server includes: one or more processors and memories; The memory is coupled to the one or more processors, and the memory is used to store computer program codes, wherein the computer program codes include computer instructions, and the one or more processors call the computer instructions to enable the anomaly detection system to perform the method according to any one of claims 1 to 7.

9. A computer-readable storage medium storing computer instructions, characterized in that: When the computer instructions are executed on an anomaly detection system, the anomaly detection system is caused to perform the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product runs on an anomaly detection system, the anomaly detection system is caused to perform the method according to any one of claims 1 to 7.

Images