Method for detecting the degree of shaft seizure of a mixer, processor and mixing station

Abstract

The present application relates to the field of engineering machinery, and in particular to a method for detecting the degree of shaft seizure of a mixer, a processor and a mixing station. The method comprises: acquiring a target image inside a mixer main machine; processing the target image to obtain a region image of a region of interest; inputting the region image into a target detection model to determine a clumping region of the region image and image coordinates of the clumping region; determining a clumping thickness of the mixer according to the clumping region and the image coordinates; and determining the degree of shaft seizure of the mixer according to the clumping thickness. Through the above technical solution, an image acquisition device acquires the internal situation of the mixer main machine, and the degree of shaft seizure of the mixer is determined according to the acquired image, without the need for manual inspection of the mixer. The problem of shaft seizure of the mixer can be discovered and processed in a timely manner, ensuring the normal operation of the mixer and improving production quality and production efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of construction machinery, in particular, to a method for detecting the degree of shaft seizure of a mixer, a processor and a mixing plant. BACKGROUND

[0002] In the production process of concrete, the mixer is in a closed state, and the mixer may have a shaft seizure failure during operation, that is, concrete clumps appear on the mixing shaft or mixing blades. The hazards of shaft seizure mainly include: 1. affecting the quality of concrete. The clumps affect the mixing of concrete and prolong the time required for uniform mixing of concrete. Secondly, in the mixing process, collision and friction will cause part of the clumps to fall into the mixed concrete, which will also affect the quality of concrete. 2. Damage to the mixer and increase energy consumption. The clumps are randomly distributed on the mixing shaft and mixing blades, causing the mixing shaft to move eccentrically, aggravating the wear of the shaft end. In addition, the clumps increase the resistance of the mixing shaft in operation, increasing energy consumption.

[0003] At present, since the mixer is closed for work, it is difficult to directly understand the internal situation of the mixer main machine. For the shaft seizure of the concrete mixer, the mixer is mainly tracked for a long time, the time period of shaft seizure is counted, and the manual cleaning period is set through the counting of the shaft seizure period of the mixing shaft. The mixer main machine is regularly checked and cleaned by manual. The regular processing method is easy to cause the mixer to be not timely treated, and the random falling of the clumps is difficult to ensure the high quality of the concrete. At the same time, when the shaft seizure is investigated in this way, the target is blindly searched, the main machine without shaft seizure is checked, the manpower is wasted, and the production efficiency is affected. SUMMARY

[0004] The purpose of the present application is to provide a method for detecting the degree of shaft seizure of a mixer, a processor and a mixing plant which can detect the degree of shaft seizure of a mixer.

[0005] In order to achieve the above-mentioned purpose, the present application provides a method for detecting the degree of shaft seizure of a mixer, applied to a mixing plant, the method comprising:

[0006] obtaining a target image inside the mixer main machine;

[0007] processing the target image to obtain a region image of a region of interest;

[0008] inputting the region image into a target detection model to determine a clump region of the region image and an image coordinate of the clump region;

[0009] determining a clump thickness of the mixer according to the clump region and the image coordinate;

[0010] determining the degree of shaft seizure of the mixer according to the clump thickness.

[0011] In the embodiment of the present application, the mixer comprises a stirring shaft and stirring blades, and processing the target image to obtain the region image of the region of interest comprises: determining a region in the target image where the stirring shaft and the stirring blades are located as the region of interest; and cropping the target image according to the region of interest to obtain the region image.

[0012] In the embodiment of the present application, determining the agglomerate thickness of the mixer according to the agglomerate region and the image coordinates comprises determining the thickness of the concrete agglomerate by formula (1):

[0013]

[0014] wherein m d is the thickness of the concrete agglomerate, y max and y min are image coordinates, wherein the coordinate parameters comprise (x min , y min , x max , y max ), wherein (x min , y min ) is the top-left coordinate of the region image, (x max , y max ) is the bottom-right coordinate of the region image, d is the image diameter of the stirring shaft in the region image, and D is the actual diameter of the stirring shaft.

[0015] In the embodiment of the present application, determining the shaft-holding degree of the mixer according to the agglomerate thickness comprises: in the case that the thickness is a first preset value, determining the shaft-holding degree as no shaft-holding output; in the case that the thickness is greater than the first preset value and less than or equal to a second preset value, determining the shaft-holding degree as mild shaft-holding; in the case that the thickness is greater than the second preset value and less than or equal to a third preset value, determining the shaft-holding degree as moderate shaft-holding; and in the case that the thickness is greater than the third preset value, determining the shaft-holding degree as severe shaft-holding. In the embodiment of the present application, determining the agglomerate thickness of the mixer according to the agglomerate region and the image coordinates comprises: in the case that the agglomerate region contained in the region image is multiple, determining first image coordinates of the agglomerate region in the discharge port region of the mixer in the region image; and determining the agglomerate thickness of the mixer according to the first image coordinates.

[0016] In the embodiment of the present application, determining the shaft-holding degree of the mixer according to the agglomerate thickness further comprises: in the case that the agglomerate region contained in the region image is multiple, determining image coordinates of each agglomerate region and determining the agglomerate thickness corresponding to each agglomerate region according to the image coordinates; comparing the agglomerate thickness corresponding to each agglomerate region; and determining the shaft-holding degree of the mixer according to the agglomerate thickness with the largest value.

[0017] In the embodiment of the present application, the target image inside the mixer host is obtained by: collecting an image video inside the mixer host through an image collection device; uniformly extracting a single-frame video frame from the image video according to a preset frame rate; and taking the extracted single-frame video frame as the target image.

[0018] In the embodiment of the present application, the image video inside the mixer host is collected by: cleaning the host of the mixer; and collecting a video inside the mixer to obtain the image video after the cleaning is completed.

[0019] In the embodiment of the present application, the method further comprises: obtaining a sample image, and processing the sample image to obtain a sample region image of a region of interest; calibrating a region containing a concrete lump in the sample region image to obtain a standard region, and determining a standard image coordinate corresponding to the standard region; performing image enhancement processing on the sample region image; for each processed sample region image, inputting the processed sample region image into the target detection model to be trained to obtain a sample lump region output by the target detection model to be trained; determining an intersection-over-union ratio of each sample lump region relative to the standard region; determining an evaluation index value of the target detection model to be trained according to the intersection-over-union ratio of each sample lump region; and determining that the target detection model to be trained is trained in a case where the evaluation index value reaches a preset threshold.

[0020] In the embodiment of the present application, the evaluation index value of the target detection model to be trained is determined according to the intersection-over-union ratio of each sample lump region, comprising: determining an intersection-over-union ratio of each sample lump region relative to the standard region; determining that the sample lump region is qualified in a case where the intersection-over-union ratio reaches a preset intersection-over-union ratio threshold; determining an evaluation index of the target detection model according to a proportion of the number of qualified sample lump regions to the number of all sample lump regions output by the target detection model; and determining that the target detection model to be trained is trained in a case where the evaluation index reaches a preset evaluation index threshold.

[0021] The second aspect of the present application provides a processor configured to execute the method for detecting the degree of shaft seizure of the mixer.

[0022] The third aspect of the present application provides a mixing station, comprising:

[0023] a mixer, the mixer comprising a stirring shaft and a stirring blade, the stirring shaft being used for stirring concrete, and the stirring blade being used for stirring the concrete;

[0024] an image collection device configured to collect an image video inside the host of the mixer; and the above processor.

[0025] In the technical solution, the image inside the mixer host is collected, the target image is processed, the region of interest in the target image is divided to determine the region image, the region image is input into the target detection model to determine the agglomeration region in the region image and the corresponding image coordinates, the agglomeration thickness of the mixer is determined according to the label of the agglomeration region determined by the target detection model and the corresponding image coordinates, and the degree of shaft holding of the mixer is determined according to the agglomeration thickness. In the technical solution of the application, the internal situation of the mixer host is collected by the image collection device, and the degree of shaft holding of the mixer is determined according to the collected image. The mixer does not need to be checked by artificial judgment, and the shaft holding problem of the mixer can be found in time and handled in time to ensure the normal operation of the mixer and improve the production quality and production efficiency.

[0026] Other features and advantages of the present application will be described in detail in the following detailed description. BRIEF DESCRIPTION OF DRAWINGS

[0027] The accompanying drawings are included to provide a further understanding of the application, and constitute a part of the specification, and are used to explain the application together with the following detailed description, but do not constitute a limitation on the application. In the drawings:

[0028] Figure 1 The flowchart of the method for detecting the degree of shaft holding of the mixer according to an embodiment of the application is schematically shown;

[0029] Figure 2 The structural schematic diagram of the mixing station according to an embodiment of the application is schematically shown. DETAILED DESCRIPTION

[0030] The specific embodiments of the application will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to illustrate and explain the application, and are not used to limit the application.

[0031] It should be noted that if the embodiments of the application involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, motion condition, etc. between the components in a certain posture (as shown in the drawings), if the certain posture changes, the directional indications also change accordingly.

[0032] In addition, if the description of "first", "second" and the like is involved in the embodiments of the present application, the description of "first", "second" and the like is only for the purpose of description, and cannot be understood as indicating or implying the relative importance of the indicated technical features or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first", "second" can be explicitly or implicitly included at least one of the features. In addition, the technical solutions of various embodiments can be combined with each other, but it must be based on the fact that the technical solutions can be realized by those skilled in the art. When the combination of technical solutions appears contradictory or unachievable, it should be considered that the combination of technical solutions does not exist and is not within the protection scope required by the present application.

[0033] As shown in Figure 1 , a flow chart of a method for detecting the degree of shaft seizure of a mixer according to an embodiment of the present application is schematically shown. As shown in Figure 1 , a method for detecting the degree of shaft seizure of a mixer is provided, including some steps:

[0034] Step 101, obtaining a target image inside the main machine of the mixer.

[0035] In one embodiment, collecting the image video inside the main machine of the mixer includes: cleaning the main machine of the mixer; after the cleaning is completed, collecting the video inside the mixer to obtain the image video.

[0036] The processor can obtain the target image inside the main machine of the mixer through the image acquisition device. After the mixer completes the stirring work, the processor can control the cleaning of the inside of the main machine of the mixer, and after the cleaning is completed, the image acquisition device can be used to collect the video inside the main machine of the mixer to obtain the image video inside the main machine of the mixer.

[0037] In one embodiment, obtaining a target image inside the main machine of the mixer includes: collecting an image video inside the main machine of the mixer through an image acquisition device; uniformly extracting a single frame video frame from the image video according to a preset frame rate; and taking the extracted single frame video frame as a target image.

[0038] The processor can obtain the image video inside the main machine of the mixer through the image acquisition device. After obtaining the image video, the image video can be processed by frame. Single frame video frames are extracted from the video data, and the frame extraction method adopts uniform frame extraction. The operator can set the frame rate of uniform frame extraction in advance according to the hardware device computing power in the production environment combined with the on-site program debugging result. After the processor obtains the preset frame rate, the processor can extract the obtained image video according to the preset frame rate, so as to obtain a single frame video frame. The processor can take the single frame video frame extracted according to the preset frame rate as a target image.

[0039] Step 102, processing the target image to obtain a region image of a region of interest.

[0040] After the processor obtains the target image, the target image can be processed, the region of interest in the target image is divided, and the target image is cropped according to the divided region of interest, so as to obtain the region image.

[0041] In one embodiment, the mixer includes a stirring shaft and a stirring blade, and processing the target image to obtain the region image of the region of interest includes: determining the region where the stirring shaft and the stirring blade are located in the target image as the region of interest; and cropping the target image according to the region of interest to obtain the region image.

[0042] The mixer includes a stirring shaft and a stirring blade for stirring concrete. After the processor obtains the target image, the processor can divide the region where the stirring shaft and the stirring blade are located in the target image into the region of interest. And according to the divided region of interest, the target image is cropped to obtain the region image.

[0043] Step 103, input the region image into the target detection model to determine the agglomerate region of the region image and the image coordinates of the agglomerate region.

[0044] After the processor obtains the region image by cropping the region of interest on the target image, the processor can input the region image into the target detection model. The target detection model determines the agglomerate region of the region image and the image coordinates corresponding to the agglomerate region.

[0045] Step 104, determining the agglomerate thickness of the mixer according to the agglomerate region and the image coordinates.

[0046] After the processor determines the agglomerate region and the corresponding image coordinates through the target detection model, the processor can determine the agglomerate thickness of the material in the mixer according to the obtained agglomerate region and the corresponding image coordinates.

[0047] In one embodiment, determining the agglomerate thickness of the mixer according to the agglomerate region and the image coordinates includes determining the thickness of the concrete agglomerate through formula (1):

[0048]

[0049] Wherein, m d is the thickness of the concrete agglomerate, y max and y min are the agglomerate region image coordinate parameters output by the target detection model, wherein the coordinate parameters include (x min , y min , x max , y max ), wherein (x min , y min) is the image coordinate of the top-left corner of the region image, (x max ,y max ) is the image coordinate of the bottom-right corner of the region image, d is the image diameter of the stirring shaft in the region image, and D is the actual diameter of the stirring shaft.

[0050] The processor can determine the agglomerate thickness of the material in the mixer according to formula (1) y max y min is the image coordinate of the region image determined by the target detection model, wherein y min is the coordinate of the top-left corner of the region image, and y max is the coordinate of the bottom-right corner of the region image. D is the image diameter of the stirring shaft displayed in the cropped region image, that is, the image diameter of the stirring shaft wrapped by the agglomerated material; and D is the actual diameter of the stirring shaft, that is, the net diameter of the stirring shaft itself. The processor can determine the agglomerate thickness of the material in the mixer according to formula (1) according to the image coordinates, the image diameter of the stirring shaft, and the net diameter.

[0051] Step 105, determining the shaft holding degree of the mixer according to the agglomerate thickness.

[0052] After the processor determines the agglomerate thickness of the material in the mixer according to formula (1), the processor can determine the shaft holding degree of the mixer according to the obtained agglomerate thickness.

[0053] In one embodiment, determining the shaft holding degree of the mixer according to the agglomerate thickness comprises: determining the shaft holding degree as no shaft holding output when the thickness is a first preset value; determining the shaft holding degree as mild shaft holding when the thickness is greater than the first preset value and less than or equal to a second preset value; determining the shaft holding degree as moderate shaft holding when the thickness is greater than the second preset value and less than or equal to a third preset value; and determining the shaft holding degree as severe shaft holding when the thickness is greater than the third preset value.

[0054] The processor can set the first preset value as 0, and when the agglomerate thickness is 0, the processor can determine that the mixer has no shaft holding phenomenon at this time. The processor can perform no shaft holding phenomenon output. When the processor determines that the agglomerate thickness is greater than 0 and less than or equal to a second preset value set by the processor, the processor can determine that the mixer has mild shaft holding at this time, and the processor can perform mild shaft holding phenomenon output. When the processor determines that the agglomerate thickness is greater than the second preset value and less than or equal to a third preset value set by the processor, the processor can determine that the mixer has moderate shaft holding at this time, and the processor can perform moderate shaft holding phenomenon output. When the processor determines that the agglomerate thickness is greater than the third preset value, the processor can determine that the mixer has severe shaft holding at this time, and the processor can perform severe shaft holding phenomenon output.

[0055] In one embodiment, determining the agglomerate thickness of the mixer according to the agglomerate region and the image coordinates comprises: in the case that there are multiple agglomerate regions contained in the region image, determining first image coordinates corresponding to the agglomerate region in the discharge port region of the mixer in the region image; and determining the agglomerate thickness of the mixer according to the first image coordinates.

[0056] After the processor inputs the region image after cropping into the target detection model, the target detection model can output the label and the corresponding image coordinates of the agglomerate region in the region image. When the region image output by the target detection model contains multiple agglomerate regions, the processor can determine the discharge port region of the mixer in the region image and determine the first image coordinates corresponding to the agglomerate region in the discharge port region of the mixer. And according to the first image coordinates, the agglomerate thickness of the mixer is determined by formula (1), and the degree of shaft sticking of the mixer is determined according to the obtained agglomerate thickness. Since the discharge port region of the mixer will be stuck frequently, and the agglomerate of the material adhered to the stirring shaft in the discharge port region is generally the largest, when multiple agglomerate regions appear in the region image, the processor can determine the agglomerate condition of the mixer according to the agglomerate condition of the discharge port region.

[0057] In one embodiment, determining the degree of shaft sticking of the mixer according to the agglomerate thickness further comprises: in the case that there are multiple agglomerate regions contained in the region image, determining the image coordinates of each agglomerate region, and determining the agglomerate thickness corresponding to each agglomerate region according to the image coordinates; comparing the agglomerate thickness corresponding to each agglomerate region; and determining the degree of shaft sticking of the mixer according to the agglomerate thickness with the largest value.

[0058] After the processor inputs the region image after cropping into the target detection model, the target detection model can output the label and the corresponding image coordinates of the agglomerate region in the region image. When the region image output by the target detection model contains multiple agglomerate regions, the target detection model can output the image coordinates of each agglomerate region. The processor can determine the agglomerate thickness corresponding to each agglomerate region according to the image coordinates of each agglomerate region according to formula (1). After obtaining the agglomerate thickness corresponding to each agglomerate region, the agglomerate thickness corresponding to each agglomerate region can be compared. The degree of shaft sticking of the mixer is determined according to the agglomerate with the largest thickness. For example, assuming that the target detection model receives the input region image and labels three agglomerate regions and the image coordinates corresponding to each region in the region image. The processor can determine the thickness of each agglomerate according to each agglomerate region and the corresponding image coordinates, and determine the degree of shaft sticking of the mixer according to the agglomerate with the largest thickness among the three agglomerates.

[0059] In one embodiment, a sample image is acquired, and the sample image is processed to obtain a sample region image of a region of interest; a region containing a concrete block in the sample region image is demarcated to obtain a standard region, and a standard image coordinate corresponding to the standard region is determined; the sample region image is subjected to image enhancement processing; for each processed sample region image, the processed sample region image is input into a target detection model to be trained to obtain a sample block region output by the target detection model to be trained; an intersection-over-union ratio of each sample block region relative to the standard region is determined; an evaluation index value of the target detection model to be trained is determined according to the intersection-over-union ratio of each sample block region; and in a case where the evaluation index value reaches a preset threshold, it is determined that the target detection model to be trained is trained.

[0060] In one embodiment, determining the evaluation index value of the target detection model to be trained according to the intersection-over-union ratio of each sample block region includes: determining the intersection-over-union ratio of each sample block region relative to the standard region; in a case where the intersection-over-union ratio reaches a preset intersection-over-union ratio threshold, determining that the sample block region is qualified; determining an evaluation index of the target detection model according to a proportion of the number of qualified sample block regions to the number of all sample block regions output by the target detection model; and in a case where the evaluation index reaches a preset evaluation index threshold, determining that the target detection model to be trained is trained.

[0061] Determining the evaluation index value of the target detection model to be trained according to the sample region label of the sample block region and the sample image coordinate, and the standard region label of the standard region and the standard image coordinate includes: determining an intersection-over-union ratio between the sample block region and the standard region; and in a case where the intersection-over-union ratio reaches a preset intersection-over-union ratio, determining that the sample block region is qualified. An evaluation index value of the target detection model to be trained is determined according to a proportion of sample block regions reaching the preset intersection-over-union ratio among all sample block regions output by the target detection model to be trained by using a confusion matrix, and in a case where the evaluation index value reaches a preset evaluation index threshold, it is determined that the target detection model to be trained is trained.

[0062] Before detecting the degree of shaft seizure of the mixer, the target detection model needs to be trained. The processor can acquire a sample image of the inside of the mixer main machine after the mixer is cleaned. The processor can divide a region where the stirring shaft and the stirring blade are located in the sample image into a region of interest, and crop the region of interest to obtain a sample region image. The processor can output the obtained sample region image. A region containing a concrete block in the sample region image can be demarcated by a person to obtain a standard region and set a standard region label, and a standard image coordinate corresponding to the standard region is determined. By demarcating the sample region image by a person, a reference standard can be obtained, which is used to determine whether the target detection model is trained.

[0063] The processor performs image enhancement processing on the sample region image, for example, randomly changing the brightness of the sample region image or the contrast of the sample region image, and the like. And the image is scaled to the model input size. After the sample region image is processed, it is input to the target detection model to be trained. The target detection model to be trained can calibrate the clumping region in the sample region image to obtain the label of the sample clumping region and the corresponding sample image coordinates, and output the obtained label of the sample clumping region and the corresponding sample image coordinates.

[0064] After the processor obtains the sample clumping region and the corresponding sample image coordinates, it can compare each sample clumping region and the corresponding sample image coordinates with the standard clumping region and the corresponding standard image coordinates, so as to obtain the intersection over union between each sample clumping region and the standard clumping region. When the intersection over union between the sample clumping region and the standard clumping region reaches the preset intersection over union set by the processor, the processor can determine that the sample clumping region is qualified. The processor can determine the evaluation index value of the target detection model according to the proportion of the number of qualified sample clumping regions to the total number of sample clumping regions. The processor can determine the evaluation index value of the target detection model to be trained according to the confusion matrix. In the case that the evaluation index value of the target detection model reaches the preset evaluation index threshold set by the processor, the processor can determine that the target detection model to be trained is completed.

[0065] The processor can also determine that the target detection model to be trained is predicted accurately in the case that the error value between the sample image coordinates and the standard image coordinates is lower than the preset error value set by the processor. The processor can determine the prediction accuracy of the target detection model to be trained by determining the number of times of prediction accuracy of the target detection model to be trained. When the number of times of prediction accuracy of the target detection model to be trained reaches the preset number of times set by the processor, the processor can determine that the prediction accuracy reaches the preset accuracy, at which time the processor can determine that the target detection model to be trained is completed.

[0066] In one embodiment, a processor configured to perform any of the above methods for detecting the degree of shaft seizure of a blender is provided.

[0067] After the blender completes the stirring work, the processor can control the cleaning of the inside of the main machine of the blender. After cleaning, the image video of the inside of the main machine of the blender can be obtained by video acquisition of the image acquisition device. The processor can perform frame processing on the image video, select images in a preset time period in the image video as candidate images, and select single-frame images that meet the preset conditions from the candidate images as target images. For example, the clearest image is selected from multiple images as the target image.

[0068] After the processor obtains the target image, the processor can process the target image, divide a region where the stirring shaft and the stirring blade are located in the target image into a region of interest, and crop the target image according to the region of interest, so as to obtain a region image. The region image is input into the target detection model. The target detection model is used to determine a caking region of the region image and image coordinates corresponding to the caking region. According to formula (1) to determine the caking thickness of the material in the mixer. In formula (1), y max and y min are image coordinates of the region image determined by the target detection model, in which y min is a coordinate of the upper left corner of the region image, y max is a coordinate of the lower right corner of the region image. D is an image diameter of the stirring shaft displayed in the cropped region image, that is, an image diameter of the stirring shaft wrapped by the caked material; and D is an actual diameter of the stirring shaft, that is, a net diameter of the stirring shaft itself.

[0069] After the processor determines the caking thickness of the image region, the processor can determine the shaft holding degree of the mixer according to the caking thickness. The processor can set a first preset value as 0, and in the case that the caking thickness is 0, the processor can determine that the mixer has no shaft holding phenomenon at this time. The processor can perform no shaft holding phenomenon output. When the processor determines that the caking thickness is greater than 0 and less than or equal to a second preset value set by the processor, the processor can determine that the mixer has a mild shaft holding at this time, and the processor can perform mild shaft holding phenomenon output. When the processor determines that the caking thickness is greater than the second preset value and less than or equal to a third preset value set by the processor, the processor can determine that the mixer has a moderate shaft holding at this time, and the processor can perform moderate shaft holding phenomenon output. When the processor determines that the caking thickness is greater than the third preset value, the processor can determine that the mixer has a severe shaft holding at this time, and the processor can perform severe shaft holding phenomenon output. An operator can determine whether to clean the mixer according to the shaft holding degree.

[0070] After the target detection model receives the regional image, it can calibrate and output the agglomeration area and the corresponding image coordinates in the regional image. However, there may be multiple agglomeration areas in the regional image at the same time. Since the feed port area of ​​the mixer often sticks to the shaft, and the material agglomerates adhering to the mixer shaft in the feed port area are generally the largest, when multiple agglomeration areas appear in the regional image, the processor can determine the agglomeration condition of the mixer based on the agglomeration condition of the feed port area. After the target detection model outputs all the agglomeration areas and the corresponding image coordinates in the regional image, the processor can determine the agglomeration area and the corresponding image coordinates of the feed port area of ​​the mixer in the regional image as the first agglomeration area and the corresponding first image coordinates. The processor can also determine the agglomeration thickness of the feed port area of ​​the mixer based on the first agglomeration area and the corresponding first image coordinates and determine the degree of agglomeration of the mixer based on the obtained agglomeration thickness.

[0071] In the case where there may be multiple agglomerated areas in the regional image, the processor can also determine the agglomeration thickness of each agglomerated area based on the output of each agglomerated area and the corresponding image coordinates of the target detection model, and determine the agglomeration condition of the mixer based on the area with the most serious agglomeration. The processor can obtain each agglomerated area and the corresponding image coordinates through the target detection model, and determine the agglomeration thickness corresponding to each agglomerated area using formula (1); compare the agglomeration thickness corresponding to each agglomerated area; and determine the degree of shaft holding of the mixer based on the agglomeration thickness with the largest value.

[0072] With this technical solution, the processor uses an image acquisition device to capture the internal conditions of the mixer main unit and determines the degree of shaft seizure based on the captured images, eliminating the need for manual inspection of the mixer. This allows for timely detection and resolution of shaft seizure issues, ensuring proper operation and improving production quality and efficiency.

[0073] In one embodiment, Figure 2 As shown, a structural diagram of a mixing station 200 is schematically shown, including: a mixer 201, the mixer 201 includes a mixing shaft 201-1 and a mixing blade 201-2, the mixing shaft 201-1 is used to mix concrete, and the mixing blade 201-2 is used to mix concrete; an image acquisition device 202, configured to capture images and videos inside the main unit of the mixer 201; and a processor 203.

[0074] An embodiment of the present application provides a device, which includes a processor, a memory, and a program stored in the memory and runnable on the processor. When the processor executes the program, the above-mentioned steps for detecting the degree of shaft seizure of the mixer are implemented.

[0075] Those skilled in the art will appreciate that embodiments of the application can be readily used as software, hardware, or a combination of software and hardware. In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0076] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 means for functionally implementing the steps listed in the flowchart block or blocks.

[0077] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture including instructions which implement the function specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 means for functionally implementing the steps listed in the flowchart block or blocks.

[0078] The computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart block or blocks. Figure 1 one or more flow or blocks Figure 1 means for functionally implementing the steps listed in the flowchart block or blocks.

[0079] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0080] The memory can include non-persistent memory and / or persistent memory, such as flash memory, readonly memory (ROM), or similar storage elements, in a computer readable medium. Memory is an example of computer readable media.

[0081] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0082] It should also be noted that the terms "comprising", "containing", or any other variant thereof are intended to cover non-exclusive inclusions, so that a process, method, article or apparatus that includes a list of elements does not only include those elements, but also includes other elements not explicitly listed, or further includes elements inherent in such a process, method, article or apparatus. Without more limitations, the element defined by the statement "comprising a" does not exclude the presence of additional identical elements in the process, method, article or apparatus that includes the element.

[0083] The above only is an embodiment of the present application, and is not used to limit the present application. For those skilled in the art, the present application can have various changes and variations. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the scope of claims of the present application.

Claims

1. A method for detecting the degree of shaft seizure of a mixer, characterized by, The method is applied to a mixing station, and the method comprises: obtaining a target image inside a host of the mixer; processing the target image to obtain a region image of a region of interest; inputting the region image into a target detection model to determine a lump region of the region image and an image coordinate of the lump region; determining a lump thickness of the mixer according to the lump region and the image coordinate; determining a shaft holding degree of the mixer according to the lump thickness; the determination of the lump thickness of the mixer according to the lump region and the image coordinate comprises: in the case where the lump region contained in the region image is multiple, determining a first image coordinate corresponding to the lump region in a discharge port region of the mixer in the region image; determining the lump thickness of the mixer according to the first image coordinate.

2. The method of claim 1, wherein, The mixer comprises a stirring shaft and stirring blades, and the processing of the target image to obtain the region image of the region of interest comprises: determining a region in which the stirring shaft and the stirring blades are located in the target image as the region of interest; cropping the target image according to the region of interest to obtain the region image.

3. The method of claim 1, wherein, The determination of the lump thickness of the mixer according to the lump region and the image coordinate comprises determining the thickness of the concrete lump by formula (1): Among them, m d is the thickness of the concrete block, y max with y min The coordinate parameters of the agglomeration area image output by the target detection model, wherein the coordinate parameters include (x min ,y min ,x max ,y max ), where (x min ,y min ) is the upper left coordinate of the region image, (x max ,y max ) is the lower right coordinate of the area image, d is the image diameter of the stirring shaft in the area image, and D is the actual diameter of the stirring shaft.

4. The method of claim 1, wherein, The determination of the shaft holding degree of the mixer according to the lump thickness comprises: in the case where the thickness is a first preset value, determining that the shaft holding degree is no shaft holding output; in the case where the thickness is greater than the first preset value and less than or equal to a second preset value, determining that the shaft holding degree is slight shaft holding; in the case where the thickness is greater than the second preset value and less than or equal to a third preset value, determining that the shaft holding degree is moderate shaft holding; in the case where the thickness is greater than the third preset value, determining that the shaft holding degree is severe shaft holding.

5. The method of claim 1, wherein, The determination of the shaft holding degree of the mixer according to the lump thickness further comprises: in the case where the lump region contained in the region image is multiple, determining the image coordinate of each lump region and determining the lump thickness corresponding to each lump region according to the image coordinate; comparing the lump thickness corresponding to each lump region; determining the shaft holding degree of the mixer according to the lump thickness with the largest value.

6. The method of claim 1, wherein, The obtaining of the target image inside the host of the mixer comprises: collecting an image video inside the host of the mixer by an image collection device; uniformly extracting a single video frame from the image video according to a preset frame rate; taking the extracted single video frame as the target image.

7. The method of claim 6, wherein, The collection of the image video inside the host of the mixer comprises: cleaning the host of the mixer; after the cleaning is completed, collecting a video inside the mixer to obtain the image video.

8. The method of claim 1, wherein, The method further comprises: obtaining a sample image and processing the sample image to obtain a sample region image of a region of interest; labeling a region containing a concrete lump in the sample region image to obtain a standard region and determining a standard image coordinate corresponding to the standard region; perform image enhancement processing on the sample area image; for each processed sample area image, input the processed sample area image into the target detection model to be trained to obtain a sample clumping area output by the target detection model to be trained; determine an intersection-over-union of each sample clumping area relative to the standard area; determine an evaluation index value of the target detection model to be trained according to the intersection-over-union of each sample clumping area; if the evaluation index value reaches a preset threshold, determine that the target detection model to be trained is trained.

9. The method of claim 8, wherein, The method comprises the following steps: determine an intersection-over-union of each sample clumping area relative to the standard area; if the intersection-over-union reaches a preset intersection-over-union threshold, determine that the sample clumping area is qualified; determine an evaluation index of the target detection model according to a proportion of the number of qualified sample clumping areas to the number of all sample clumping areas output by the target detection model; if the evaluation index reaches a preset evaluation index threshold, determine that the target detection model to be trained is trained.

10. A processor, comprising: The method is configured to detect the degree of shaft seizure of a mixer according to any one of claims 1 to 9.

11. A mixing plant, characterized in that The mixing station comprises: a mixer comprising a mixing shaft for mixing concrete and mixing blades for mixing the concrete; an image acquisition device configured to acquire an image video inside the mixer host; and a processor according to claim 10.

Images