Gradation automatic regulation and control system and method based on aggregate image recognition

Abstract

The invention discloses an automatic grading regulation and control system and method based on aggregate image recognition, relates to the technical field of equipment automation, and solves the technical problem that in the prior art, only aggregate pressure is considered, and actually transported aggregate is not recognized, so that deviation exists in grading parameter setting. The method comprises the following steps: acquiring an aggregate image through image acquisition equipment, and preprocessing the aggregate image to obtain a standard aggregate image; uniformly segmenting the standard aggregate image into a plurality of regional images and numbering the regional images; extracting regional features of the regional image, comparing the regional features with a regional feature range in a database, identifying to obtain aggregate categories of aggregates in the regional image and abnormal aggregates, and obtaining aggregate data; based on the aggregate data and the area numbers, calculating to obtain image coefficients of the area images, obtaining partition specifications of the image groups based on the image coefficients, regulating and controlling equipment grading parameters of the equipment according to the partition specifications, and adjusting the equipment in time according to the equipment grading parameters.

Description

Technical Field

[0001] The present invention belongs to the field of equipment automation, relates to image analysis technology, and specifically is a grading automatic regulation system and method based on aggregate image recognition. Background Art

[0002] Grading refers to the particle size distribution of aggregates in asphalt mixtures, that is, the proportion of aggregates with different particle sizes in asphalt mixtures. Asphalt mixtures with different gradings have obvious differences in mechanical properties. Aggregates that are too fine and too coarse will reduce the strength and stability of asphalt mixtures. Appropriate grading can improve performance indicators such as the shear strength, compressive strength, and durability of asphalt mixtures. Therefore, the grading of asphalt mixtures has a great influence on the performance and service life of asphalt mixtures.

[0003] The invention patent with the application number CN2023116226844 discloses an aggregate grading detection method, a silo discharging control method, a system and a device. This invention predicts the current grading and current discharging speed of aggregates in the silo through the pressure generated by the falling aggregates hitting the pressure sensor, and thus calculates the total grading of aggregates in all silos under the current discharging state. When obtaining the required grading parameters in this invention, different types of aggregates have different corresponding grading parameters set. If only the pressure of the falling aggregates is considered, it may be due to the distribution or superposition of aggregates resulting in the pressure not corresponding to the actual transported aggregates, leading to deviations in the set grading parameters.

[0004] The present invention provides a grading automatic regulation system and method based on aggregate image recognition to solve the above technical problems. Summary of the Invention

[0005] The present invention aims to solve at least one of the technical problems existing in the prior art; for this purpose, the present invention proposes a grading automatic regulation system and method based on aggregate image recognition to solve the technical problem that only the aggregate pressure is considered in the prior art, and the actual transported aggregates are not recognized, resulting in deviations in the set grading parameters.

[0006] To achieve the above object, the first aspect of the present invention provides a grading automatic regulation system based on aggregate image recognition, including: a database, an image collection module, a recognition and classification module, and a grading analysis module;

[0007] The image collection module: is used to collect the aggregate images collected by the image acquisition device;

[0008] The recognition and classification module: is used to preprocess the aggregate image to obtain a standard aggregate image; divide the standard aggregate image into several regional images and number the regions, extract the regional features in the regional images, and identify the aggregate data in the regional images based on the matching results between the regional features and the database;

[0009] Gradation analysis module: used to obtain the partition specifications according to the aggregate data and the area number, regulate the corresponding equipment gradation parameters according to the partition specifications; automatically regulate the equipment according to the equipment gradation parameters.

[0010] Preferably, the preprocessing of the aggregate image to obtain a standard aggregate image includes:

[0011] Extract the aggregate image;

[0012] Perform noise removal and grayscale processing on the aggregate image in sequence, and perform image processing on the processed aggregate image through edge calculation to separate the aggregate from the background to obtain a standard aggregate image.

[0013] Preferably, the segmentation of the standard aggregate image into several regional images and the area numbering include:

[0014] Extract the standard aggregate image;

[0015] Evenly divide the standard aggregate image into n regional images, and respectively obtain the image distance between the center point of the regional image and the feed inlet;

[0016] Number the regional images according to the size of the image distance and the order from left to right, numbered 1, 2, 3... n respectively; where n is a positive integer.

[0017] Preferably, the identification of the aggregate data in the regional image based on the matching result of the regional features and the database includes:

[0018] S1: Extract the regional features and the database; the regional features include the color features, shape features and texture features of the aggregates in the regional image; the database includes the color feature ranges, shape feature ranges, texture feature ranges of different categories of aggregates in the historical data and the corresponding aggregate categories;

[0019] S2: Judge whether the color feature of the aggregate in the regional image conforms to the color feature range; if so, jump to S3; if not, compare the color feature with the color features of other aggregate categories, and if there is no matching result, mark the corresponding aggregate as an abnormal aggregate;

[0020] S3: Judge whether the shape feature of the aggregate in the regional image conforms to the shape feature range; if so, jump to S4; if not, mark the corresponding aggregate as an abnormal aggregate;

[0021] S4: Judge whether the texture feature of the aggregate in the regional image conforms to the texture feature range; if so, mark the category of the aggregate as the corresponding aggregate category; if not, mark the corresponding aggregate as an abnormal aggregate;

[0022] S5: Mark the aggregate categories and abnormal aggregates within the regional image as aggregate data.

[0023] Preferably, the obtaining of the partition specifications according to the aggregate data and the regional numbers includes:

[0024] Extract the aggregate data and the regional numbers;

[0025] Distinguish the regional numbers according to the image distance between the center point of the regional image and the feeding port, and divide the regional numbers with the same image distance into the same group to obtain several groups of image groups;

[0026] Through the formula Calculate the image coefficient SVn of the regional image numbered n; where, SLn represents the quantity of the aggregate numbered q in the regional image numbered n, and αq represents the grading coefficient of the aggregate numbered q; q = 1, 2, 3... p, and p is a positive integer;

[0027] Mark the average value of the image coefficients of the regional images in the same group as the image group coefficient of the image group; mark the sizes of several groups of image groups and the corresponding image group coefficients as the partition specifications.

[0028] Preferably, the regulating of the corresponding equipment grading parameters according to the partition specifications includes:

[0029] Extract the partition specifications; extract the historical partition specifications and the corresponding historical equipment grading parameters from the database;

[0030] Use the historical partition specifications and the corresponding historical equipment grading parameters as training data and test data, train the neural network model with the training data, test the trained neural network model with the test data, and adjust the parameters of the neural network model according to the test results to obtain a grading model with the input data being the partition specifications and the output data being the equipment grading parameters;

[0031] Input the partition specifications into the grading model to obtain the equipment grading parameters corresponding to the image group.

[0032] The second aspect of the present invention provides a grading automatic regulation method based on aggregate image recognition, which is applied to the grading automatic regulation system based on aggregate image recognition as described above, and is characterized by including:

[0033] Step 1: Collect the aggregate images collected by the image acquisition device;

[0034] Step 2: Preprocess the aggregate images to obtain standard aggregate images; divide the standard aggregate images into several regional images and number the regions;

[0035] Step 3: Extract the regional features in the regional image, and identify the aggregate data in the regional image based on the matching result between the regional features and the database;

[0036] Step 4: Obtain the partition specifications according to the aggregate data and the regional numbers, and regulate the corresponding equipment grading parameters according to the partition specifications; automatically regulate the equipment according to the equipment grading parameters.

[0037] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0038] 1. The present invention collects the aggregate image through the image acquisition device, and preprocesses the aggregate image to obtain the standard aggregate image, making the image clearer, the details more obvious, and highlighting the edge information of the aggregate; evenly divides the standard aggregate image into several regional images and numbers them, which can realize the automatic analysis of the aggregate distribution, can more accurately analyze the aggregate distribution in each region, and helps to improve the accuracy of grading regulation; extracts the regional features of the regional image, compares them with the regional feature range in the database, identifies the aggregate categories and abnormal aggregates in the regional image to obtain the aggregate data. By comparing the feature extraction with the feature range in the database, the accuracy of aggregate category identification can be improved. By identifying abnormal aggregates, measures can be taken in time to avoid unqualified materials from entering the production process, thereby improving the product quality; based on the aggregate data and the regional numbers, calculate the image coefficient of the regional image, obtain the partition specifications of the image group based on the image coefficient, regulate the equipment grading parameters of the equipment according to the partition specifications, and adjust the equipment in time according to the equipment grading parameters. By automatically adjusting the grading parameters, the operation process is simplified, the work efficiency of the operator is improved, the number of manual interventions is reduced by automatically adjusting the grading parameters, the automation degree of the production line is improved, and the production quality is improved.

[0039] 2. The present invention divides the aggregate image into several regional images, and divides the regional images into several image groups according to the image distance. The image distances of different image groups from the feed inlet are different, and they are sorted from near to far. The equipment grading parameters corresponding to different image groups are also different. By adjusting the equipment grading parameters, the production quality can be improved, the parameter settings of the production equipment can be controlled more accurately, thereby improving the production efficiency, reasonably allocating the production resources, and avoiding the waste of resources caused by overusing some equipment. The grading parameters can be dynamically adjusted according to the real-time monitored aggregate distribution, improving the flexibility and adaptability of the system. Description of the Drawings

[0040] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for use in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0041] Figure 1 It is a flowchart of the operation of an embodiment of the present invention.

[0042] Figure 2 It is a system composition diagram of the present invention.

[0043] Figure 3 It is a complete flowchart for obtaining aggregate data in an embodiment of the present invention. Detailed implementation manners

[0044] The following will clearly and completely describe the technical solutions of the present invention in combination with the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0045] Please refer to Figures 1-3 , an embodiment of the first aspect of the present invention provides a grading automatic regulation system based on aggregate image recognition, including: an image collection module, an identification and classification module, and a grading analysis module;

[0046] The image collection module: used to collect the aggregate images collected by the image acquisition device.

[0047] Exemplarily, a high-definition camera is set above the aggregate conveyor belt, and the aggregate images on the conveyor belt are collected in real time through the high-definition camera. The aggregate is building materials such as sand and gravel; the coverage range of the camera needs to ensure that it can fully cover the transmission path of the aggregate or the area containing the aggregate on the conveyor belt.

[0048] The identification and classification module: used to preprocess the aggregate image to obtain a standard aggregate image; divide the standard aggregate image into several regional images and number the regions, extract the regional features in the regional images, and identify the aggregate data in the regional images based on the matching results of the regional features and the database.

[0049] Exemplarily, the collected aggregate images are subjected to noise removal and grayscale processing at one time, the aggregate images are converted into grayscale images, and the grayscale images are processed through edge computing to separate the aggregates from the image background to obtain standard aggregate images; through preprocessing the images, it is more convenient to extract aggregate features and can also improve the recognition accuracy of the target object.

[0050] In this embodiment, the aggregate image is evenly divided into 9 regional images, numbered as regional image one, regional image two, regional image three, regional image four, regional image five, regional image six, regional image seven, regional image eight, and regional image nine respectively; the image distance between the center points of regional image one, regional image two, and regional image three and the image of the feed inlet is 1 meter, the image distance between the center points of regional image four, regional image five, and regional image six and the image of the feed inlet is 2 meters, and the image distance between the center points of regional image seven, regional image eight, and regional image nine and the image of the feed inlet is 3 meters; the RGB color values of the aggregates in regional image one are extracted as color features, the shape features of the aggregates are obtained by edge calculation, and the texture features of the aggregates are calculated by calculating the gray-level co-occurrence matrix; the color feature ranges, shape feature ranges, texture feature ranges of different aggregates and the corresponding aggregate categories are extracted from the database.

[0051] Judge whether the color features of the aggregates in regional image one conform to the color feature range. It is judged that the color features of the aggregates in regional image one conform to the color feature range of sand and gravel; continue to judge whether the shape features of the aggregates in the regional image conform to the shape feature range. It is judged that the shape features of the aggregates in regional image one conform to the shape feature range of sand and gravel; continue to judge whether the texture features of the aggregates in the regional image conform to the texture feature range. It is judged that the texture features of the aggregates in regional image one conform to the texture feature range of sand and gravel, then the aggregates in regional image one are marked as sand and gravel.

[0052] In some other embodiments, if the corresponding color features, shape features, and texture features of the aggregates do not conform to the corresponding feature ranges, the corresponding aggregates are marked as abnormal aggregates, indicating that the aggregates do not belong to the aggregate categories recorded in the database and may be other construction wastes, etc. When abnormal aggregates are detected, screening can be arranged to ensure the production quality; and the judgment order can be freely selected, and it is preferred to select the features with relatively easy contrast for priority comparison, which can reduce the calculation amount and improve the work efficiency.

[0053] Gradation analysis module: used to obtain the partition specifications according to the aggregate data and the regional numbers, regulate the corresponding equipment gradation parameters according to the partition specifications; automatically regulate the equipment according to the equipment gradation parameters.

[0054] Exemplarily, regional image one, regional image two, and regional image three are divided into image group one, regional image four, regional image five, and regional image six are divided into image group two, and regional image seven, regional image eight, and regional image nine are divided into image group three; in this embodiment, there are three types of aggregates: crushed stone, sand and gravel, and slag. The number of crushed stone is set to 1, and the gradation coefficient α1 = 0.4; the number of sand and gravel is 2, and the gradation coefficient α2 = 0.2; the number of slag is 3, and the gradation coefficient α3 = 0.6.

[0055] In this embodiment, the number of gravels detected in Region Image 1 is 17, the number of sands and gravels is 35, and the number of slag is 22; through the formula it is calculated that the image coefficient SV1 of Region Image 1 is 27; it is calculated that the image coefficient SV2 of Region Image 2 is 32, and the image coefficient SV3 of Region Image 3 is 31; the average value is calculated to obtain the image group coefficient of Image Group 1 as 30. The transmission length of Image Group 1 on the conveyor belt is 1 meter, and the transmission length is the length of Image Group 1 mapped to the edge of the conveyor belt. The transmission length and image group coefficient of the image group are marked as the partition specifications.

[0056] Extract the partition specifications; extract the historical partition specifications and the corresponding historical equipment grading parameters from the database;

[0057] Use the historical partition specifications and the corresponding historical equipment grading parameters as training data and test data. Use the training data to train the neural network model, use the test data to test the trained neural network model, and adjust the parameters of the neural network model according to the test results to obtain a grading model with the input data being the partition specifications and the output data being the equipment grading parameters;

[0058] Input the partition specifications into the grading model to obtain the equipment grading parameters corresponding to the image group; according to the equipment grading parameters, the time for adjusting the equipment parameters can be determined in a timely manner. For the same batch of transported aggregates, the quantity and size of the aggregates will affect the parameters to be adjusted. Segment the aggregate images and adjust the real-time parameters for the transported aggregates corresponding to different image groups, which can adapt to different transportation situations.

[0059] The second aspect of the present invention provides a grading automatic regulation method based on aggregate image recognition, which is applied to the grading automatic regulation system based on aggregate image recognition described above, and is characterized by including:

[0060] Step 1: Collect the aggregate images collected by the image acquisition device;

[0061] Step 2: Preprocess the aggregate images to obtain standard aggregate images; segment the standard aggregate images into several region images and number the regions;

[0062] Step 3: Extract the region features in the region images, and identify the aggregate data in the region images based on the matching results between the region features and the database;

[0063] Step 4: Obtain the partition specifications according to the aggregate data and the region numbers, and regulate the corresponding equipment grading parameters according to the partition specifications; automatically regulate the equipment according to the equipment grading parameters.

[0064] Some of the data in the above formula is calculated by removing the dimension and taking its numerical value. The formula is the one that is closest to the actual situation obtained through software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0065] The working principle of the present invention:

[0066] The present invention collects aggregate images through an image acquisition device, preprocesses the aggregate images to obtain standard aggregate images, evenly divides the standard aggregate images into several regional images and numbers them respectively, extracts the regional features of the regional images, compares them with the regional feature ranges in the database to obtain aggregate data, calculates the image coefficients of the regional images based on the aggregate data and the regional numbers, obtains the partition specifications of the image group based on the image coefficients, regulates the equipment grading parameters of the equipment according to the partition specifications, and timely adjusts the equipment according to the equipment grading parameters.

[0067] The above embodiments are only used to illustrate the technical method of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An automatic grading control system based on aggregate image recognition, characterized in that: include: Database, image collection module, recognition and classification module and gradation analysis module; Image collection module: used to collect aggregate images collected by image acquisition equipment; Identification and classification module: used to pre-process the aggregate image to obtain a standard aggregate image; divide the standard aggregate image into several regional images and number the regions, extract regional features in the regional images, and identify the aggregate data in the regional images based on the matching results of the regional features and the database; Grading analysis module: used to obtain partition specifications based on aggregate data and area numbers, and adjust the corresponding equipment grading parameters according to the partition specifications; The equipment is automatically controlled according to the equipment grading parameters.

2. The automatic grading control system based on aggregate image recognition according to claim 1 is characterized in that: The step of preprocessing the aggregate image to obtain a standard aggregate image includes: Extracting aggregate images; The aggregate image is processed by removing noise and grayscale in turn, and the processed aggregate image is processed by edge computing to separate the aggregate from the background and obtain a standard aggregate image.

3. The automatic grading control system based on aggregate image recognition according to claim 1 is characterized in that: The step of dividing the standard aggregate image into a plurality of region images and numbering the regions comprises: Extract standard aggregate images; The standard aggregate image is evenly divided into n regional images, and the image distance between the center point of the regional image and the feed port is obtained respectively; The regional images are numbered according to the size of the image distance and the order from left to right, and are numbered 1, 2, 3...n respectively; where n is a positive integer.

4. The automatic grading control system based on aggregate image recognition according to claim 1 is characterized in that: The method of obtaining aggregate data in the regional image based on the matching result between the regional features and the database includes: S1: extracting regional features and a database; the regional features include color features, shape features and texture features of aggregates in the regional image; the database includes color feature ranges, shape feature ranges, texture feature ranges and corresponding aggregate categories of different categories of aggregates in historical data; S2: Determine whether the color feature of the aggregate in the regional image meets the color feature range; if yes, jump to S3; if no, compare the color feature with the color features of other aggregate categories, and if there is no match, mark the corresponding aggregate as abnormal aggregate; S3: Determine whether the shape feature of the aggregate in the regional image meets the shape feature range; if yes, jump to S4; if no, mark the corresponding aggregate as abnormal aggregate; S4: judging whether the texture feature of the aggregate in the regional image meets the texture feature range; if yes, marking the category of the aggregate as the corresponding aggregate category; if no, marking the corresponding aggregate as abnormal aggregate; S5: Mark the aggregate categories and abnormal aggregates in the regional image as aggregate data.

5. The automatic grading control system based on aggregate image recognition according to claim 1 is characterized in that: The obtaining of the zoning specifications according to the aggregate data and the area number includes: Extract aggregate data and zone numbers; The region numbers are distinguished according to the image distance between the center point of the region image and the feed port, and the region numbers with the same image distance are divided into the same group to obtain a plurality of image groups; By formula The image coefficient SVn of the image of the region numbered n is calculated; wherein SLn represents the number of aggregates numbered q in the image of the region numbered n, and αq represents the gradation coefficient of aggregates numbered q; q=1, 2, 3…p, and p is a positive integer; The average value of the image coefficients of the same group of regional images is marked as the image group coefficient of the image group; the sizes of several groups of image groups and the corresponding image group coefficients are marked as partition specifications.

6. The automatic grading control system based on aggregate image recognition according to claim 1 is characterized in that: The adjusting and controlling of the corresponding equipment grading parameters according to the partition specifications includes: Extract partition specifications; extract historical partition specifications and corresponding historical equipment grading parameters from the database; The partition specifications are input into the grading model to obtain the equipment grading parameters corresponding to the image group; wherein, the input grading model uses the historical partition specifications and the corresponding historical equipment grading parameters as training data and test data, uses the training data to train the neural network model, uses the test data to test the trained neural network model, and adjusts the parameters of the neural network model according to the test results to obtain a grading model whose input data is the partition specifications and output data is the equipment grading parameters.

7. A method for automatic grading control based on aggregate image recognition, applied to an automatic grading control system based on aggregate image recognition as claimed in any one of claims 1 to 6, characterized in that: include: Step 1: Collect aggregate images acquired by image acquisition equipment; Step 2: Preprocess the aggregate image to obtain a standard aggregate image; Segment the standard aggregate image into several regional images and number the regions; Step 3: extracting regional features from the regional image, and identifying the aggregate data in the regional image based on the matching results between the regional features and the database; Step 4: Obtain the partition specifications based on aggregate data and area numbers, and adjust the corresponding equipment grading parameters according to the partition specifications; The equipment is automatically controlled according to the equipment grading parameters.

Images