KR20230055479A - Ready-mixed concrete quality image judgment system using artificial intelligence - Google Patents

Abstract

The present invention relates to a system for determining a ready-mixed concrete quality image in real time using artificial intelligence. For the purpose, provided is a ready-mixed concrete quality image determining system using artificial intelligence, comprising: a camera (100) for photographing a hopper (11) when ready-mixed concrete (40) is input into a pump car (10) from a ready-mixed concrete vehicle (20); an artificial intelligence inference model (150) for inferring to divide the surface of the ready-mixed concrete (40) into a normal area (62) and a defective area (64) based on the ratio of aggregate exposed to the surface of the ready-mixed concrete (40) in the image of the camera (100); and a control unit (180) for calculating the area ratio of the normal area (62) and the defective area (64) and determining the quality of the ready-mixed concrete based on the calculated ratio. Therefore, the objectivity of determination can be improved.

Description

Ready-mixed concrete quality image judgment system using artificial intelligence}

The present invention relates to a technology for determining the quality of concrete ready-mixed concrete brought into the field, and more particularly, to a system for image-determining the quality of ready-mixed concrete in real time using artificial intelligence.

At civil engineering and construction sites, a ready-mixed concrete vehicle transports ready-mixed concrete from the outside, and a pump car receives it and pours it into a formwork. However, since the ready-mixed concrete that was transported and poured was mixed by an external ready-mixed concrete company, thorough quality control was difficult. For example, some ready-mixed concrete manufacturers used to produce defective ready-mixed concrete through expedient methods such as manipulating the mixing ratio and adding water above the standard value.

In the case of pouring such defective ready-mixed concrete as it is, side effects such as quality defects, non-uniformity of ready-mixed concrete, demolition and re-construction, and civil complaints occur. In particular, when a plurality of ready-mixed concrete vehicles are utilized, there is a practical difficulty in that it is difficult to detect partially defective ready-mixed concrete. This is because the current quality regulation stipulates only one quality test per 150 m ³ of ready-mixed concrete.

In addition, since the ready-mixed concrete is transported closedly through the pipe, and the moment the ready-mixed concrete is discharged into the hopper is all that is exposed to the outside before pouring, it is very difficult to visually check the quality in real time. And even if it is judged to be a defective ray cone, most of it has already been cast.

1. Korean Patent Publication No. 10-2013-0040296 (Concrete Quality Management System and Management Method), 2. Republic of Korea Patent Publication No. 10-2016-0123691 (concrete quality management prediction system using a mobile terminal).

Therefore, the present invention has been made to solve the above problems, and the problem to be solved by the present invention is to provide an image determination system of ready-mixed concrete quality using artificial intelligence that can determine the quality of discharged ready-mixed concrete in real time will be.

However, the technical problems to be achieved in the present invention are not limited to the above-mentioned technical problems, and other technical problems not mentioned will be clear to those skilled in the art from the description below. You will be able to understand.

In order to achieve the above technical problem, when the ready-mixed concrete 40 is put into the pump car 10 from the ready-mixed concrete vehicle 20, the camera 100 for photographing the hopper 11; An artificial intelligence inference model 150 that infers to divide the surface of the ready-mixed concrete 40 into a normal area 62 and a defective area 64 based on the ratio of the aggregate exposed to the surface of the ready-mixed concrete 40 in the image of the camera 100 ); And a control unit 180 that calculates the area ratio of the normal area 62 and the defective area 64 and determines the quality of the ready-mixed concrete based on the calculated ratio; An image determination device is provided.

In addition, the camera 100 may be mounted on the ready-mixed concrete vehicle 20 or a camera of a mobile phone 120 carried by a worker 125.

In addition, the image of the camera 100 is composed of an exclusion region 50 not used for inference and a region of interest 60 used for inference among the planar images of the hopper 11 .

In addition, the exclusion area 50 includes an area 45 into which the ready-mixed concrete 40 is injected from the ready-mixed concrete vehicle 20 .

In addition, an image processing unit 130 for extracting an image of each frame from the image of the camera 100 and extracting a region of interest 60 from the extracted image is further included.

In addition, the artificial intelligence inference model 150 may infer based on the ratio of the shading of the aggregate exposed to the surface of the ready-mixed concrete 40 among the images of the camera 100.

In addition, the artificial intelligence inference model 150 infers a normal area 62 when the shade of the aggregate indicates a density equal to or higher than a threshold value, and infers a defective area 64 when the shade does not reach the threshold value.

In addition, another object of the present invention as described above, as another embodiment, a history database 210 in which a plurality of history information about the quality determination of ready-mixed concrete is stored; A model learning unit 220 for generating an artificial intelligence inference model by machine learning using a plurality of history information; a transmission/update unit 240 that transmits the inference model to the application program 140 in the workplace or updates it with the latest inference model; Database means (260, 270, 280) for storing the worker information input by the worker 125 of the workshop through the application 140, the workshop information and the ready-mixed concrete vehicle 20 information; And a server control unit 250 for storing the information of the database means and the image used by the reasoning model 150 for the decision on the workplace and the decision result in the history database 210; It can also be achieved by the learning server device of the artificial intelligence inference model for.

In addition, the model learning unit 220 is a normal region 62 when the shade of the aggregate in the image of the surface of the ready-mixed concrete indicates a density equal to or higher than the threshold value, and a defective region 64 when the threshold value is not reached. generate

In addition, the workplace is at least one workplace located in a remote location, and the application program 140 is executed on at least one of a personal computer, a tablet PC, and a mobile phone 120 located in the workplace.

In addition, the object of the present invention as described above is another category, the step of photographing the hopper 11 by the camera 100 when the ready-mixed concrete 40 is introduced from the ready-mixed concrete vehicle 20 to the pump car 10 (S100 ); The image processor 130 extracts an image from the image of the camera 100 and extracts a region of interest 60 from the image (S110); The inference model 150 reasoned to divide the surface of the ready-mixed concrete 40 into a normal region 62 and a defective region 64 based on the shade ratio of the aggregate exposed on the surface of the ready-mixed concrete 40 for the region of interest 60 Step (S120); and calculating, by the controller 180, an area ratio between the normal area 62 and the defective area 64 (S130); And the control unit 180 determining and outputting the quality of the ready-mixed concrete based on the calculated ratio (S140); can also be achieved by a method for determining the image of the ready-mixed concrete quality using artificial intelligence, characterized in that it includes.

In addition, in the inference step (S120), if the shadow of the aggregate indicates a density equal to or higher than a threshold value, it is inferred as a normal region 62, and if it does not reach the threshold value, it is inferred as a defective region 64.

In addition, in the determination step (S140), when the ratio of the normal region 62 is 30% or more, it is determined as normal ready-mixed concrete, and when it is less than 30%, it is determined as defective ready-mixed concrete.

In addition, when it is determined that the ready-mixed concrete is defective, the control unit 180 further includes generating a warning (S150).

In addition, the communication unit 190 further includes transmitting the image used for inference and the result of the determination to the server 200 .

According to one embodiment of the present invention, it is possible to determine whether the ready-mixed concrete is defective regardless of the skill level of the operator, thereby increasing the objectivity of the determination.

In addition, accurate real-time judgment is possible by using a machine-learned artificial intelligence inference model.

If it is determined that it is defective ready-mixed concrete, it is possible to immediately warn and stop pouring at the same time to prevent the defective ready-mixed concrete from being poured.

In addition, the central server can share information from each workplace in real time by communicating data with multiple workplaces, and since various history information received from each workplace is accumulated in the server, the latest artificial intelligence inference model can be maintained. there is

However, the effects obtainable in the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below. You will be able to.

The following drawings attached to this specification illustrate preferred embodiments of the present invention, and together with the detailed description of the present invention serve to further understand the technical idea of the present invention, the present invention is the details described in such drawings should not be construed as limited to
1 is a configuration diagram in which a ready-mixed concrete vehicle 20 and a pump car 10 are arranged for concrete pouring in a workshop;
2 is a plan view dividing a region of interest 60 and an exclusion region 50 so that an embodiment of the present invention is applied to the hopper 11 shown in FIG. 1;
3 is a schematic block diagram of an image determination system of ready-mixed concrete quality using artificial intelligence according to an embodiment of the present invention;
4 is a schematic internal block diagram of the server 200 in FIG. 3;
5a is a photograph of a region of interest 60 determined to be defective ready-mixed concrete according to the present invention;
Figure 5b is a photograph determined to be normal ready-mixed concrete for the region of interest 60 according to the present invention;
6 is a flowchart schematically illustrating a method for determining an image of ready-mixed concrete quality using artificial intelligence according to an embodiment of the present invention.

Hereinafter, with reference to the accompanying drawings, embodiments of the present invention will be described in detail so that those skilled in the art can easily carry out the present invention. However, since the description of the present invention is only an embodiment for structural or functional description, the scope of the present invention should not be construed as being limited by the embodiments described in the text. That is, since the embodiment can be changed in various ways and can have various forms, it should be understood that the scope of the present invention includes equivalents capable of realizing the technical idea. In addition, since the object or effect presented in the present invention does not mean that a specific embodiment should include all of them or only such effects, the scope of the present invention should not be construed as being limited thereto.

The meaning of terms described in the present invention should be understood as follows.

Terms such as "first" and "second" are used to distinguish one component from another, and the scope of rights should not be limited by these terms. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element. It should be understood that when an element is referred to as “connected” to another element, it may be directly connected to the other element, but other elements may exist in the middle. On the other hand, when an element is referred to as being “directly connected” to another element, it should be understood that no intervening elements exist. Meanwhile, other expressions describing the relationship between components, such as “between” and “immediately between” or “adjacent to” and “directly adjacent to” should be interpreted similarly.

Singular expressions should be understood to include plural expressions unless the context clearly dictates otherwise, and terms such as “comprise” or “having” refer to a described feature, number, step, operation, component, part, or It should be understood that it is intended to indicate that a combination exists, and does not preclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

All terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs, unless defined otherwise. Terms defined in commonly used dictionaries should be interpreted as consistent with meanings in the context of related art, and cannot be interpreted as having ideal or excessively formal meanings unless explicitly defined in the present invention.

Configuration of the embodiment

Hereinafter, the configuration of a preferred embodiment will be described in detail with reference to the accompanying drawings. 1 is a configuration diagram in which a ready-mixed concrete vehicle 20 and a pump car 10 are arranged for concrete pouring in a workshop. As shown in FIG. 1, the ready-mixed concrete vehicle 20 discharges the externally mixed ready-mixed concrete to the hopper 11 of the pump car 10. The pump car 10 pours the ready-mixed concrete into the formwork 1 through the boom 12 and the end hose 13.

FIG. 2 is a plan view illustrating a region of interest 60 and an exclusion region 50 so that an embodiment of the present invention is applied to the hopper 11 shown in FIG. 1 . As shown in FIG. 2 , the camera 10 takes a video of the hopper 11 from the top of the hopper 11 . The camera 10 may be a stationary CCD camera fixed to the pump car 10 . In addition, the camera 10 may be a camera mounted on the mobile phone 120 or tablet PC of the worker 125 .

The hopper 11 is composed of a frame 30 and a plurality of ribs 32 for reinforcement. The ready-mixed concrete 45 being injected is introduced into the pump car 10 through the gap between the ribs 32.

The exclusion area 50 is an area that may interfere with judgment and may be set as a rectangle. The exclusion area 50 is not entered into the reasoning model 150 or is ignored. In this embodiment, the exclusion area 50 is set as an area where the ready-mixed concrete 45 discharged from the ready-mixed concrete vehicle 20 and being put in is located.

The Region of Interest (ROI) 60 is a region used for inference by the inference model 150 . The region of interest 60 is defined as a rectangle, and may be defined as the largest rectangle of the plane of the hopper 11 excluding the exclusion region 50 .

3 is a schematic block diagram of an image determination system of ready-mixed concrete quality using artificial intelligence according to an embodiment of the present invention. As shown in FIG. 3 , the decision system may include a server 200 and a client installed in at least one workplace.

The server 200 may be placed in a central computer room or field office, may be a server computer, and is connected to a wired or wireless network through a LAN, the Internet, or an intranet.

Field data 110 may be a video generated by the camera 100 or the mobile phone 120 taken. The field data 110 includes login data entered by the operator 125, operator information (eg ID, password, mobile phone number, email, name, position, etc.), workplace information (eg, workplace location or address, tools, etc.), Includes ready-mixed concrete vehicle information (e.g. company name, license plate number, driver), date and time information, etc.

The operating program 140 is a program that is installed and executed in the personal computer or mobile phone 120 connected to the camera 100 .

The image processing unit 130 is located between the field data 110 and the inference model 150. The image processing unit 130 extracts individual images (eg, JPG files) from 24 to 30 frames of video, and extracts (crops) the region of interest 60 from the extracted images. The image processing unit 130 converts the extracted image of the region of interest 60 into a unified standard by performing rotation, up/down/left/right inversion, enlargement/reduction, etc., and brightness, saturation, hue, etc. to emphasize the shade of the aggregate. is adjusted according to predetermined parameters. Image processing of the image processing unit 130 may be performed using a known algorithm.

The artificial intelligence inference model 150 divides the surface of the ready-mixed concrete 40 into normal areas 62 and defective areas 64 based on the shadow density of the aggregate exposed on the surface of the ready-mixed concrete 40 among the images provided by the image processing unit 130. ) is inferred to divide by That is, the artificial intelligence inference model 150 infers a normal area 62 when the shade of the aggregate indicates a density equal to or higher than a threshold value, and infers a defective area 64 when the shade does not reach the threshold value.

If the mixing ratio of ready-mixed concrete is abnormal or excessive water is added (aqueduct), material separation (separation of aggregate) occurs in concrete. In concrete where material separation has occurred, the constituent materials are not evenly distributed, and the coarse aggregate and water move separately, and the water floats to the surface, enabling image analysis. In view of this point, the artificial intelligence inference model 150 infers the defective area 64 when the surface of the ready-mixed concrete 40 is smooth or the shading density of the aggregate is low, and the normal area 62 when the shading density of the aggregate is high. infer as

The controller 180 calculates the areas of the inferred normal region 62 and the defective region 64 and calculates a relative ratio. If the normal area 62 of the calculated ratio is 30% or more, it is determined as normal ready-mixed concrete, and if it is less than 30%, it is determined as defective ready-mixed concrete. The control unit 180 may be a CPU, an AP (application processor), MICOM, or the like. The control unit 180 transmits the determination result to the output unit 160 and to the server 200 through the communication unit 190 .

The output unit 160 may be a monitor, a touch screen display 170 of the mobile phone 120, or a printer.

The communication unit 190 transmits the judgment result of the control unit 180, the field data 110, the video or image used in the judgment to the server 200, and the latest learned inference model 230 from the server 200 is sent The communication unit 190 is wired and wirelessly connected through a LAN, Internet, intranet, etc., and may be a transmission control protocol/internet protocol (TCP/IP), Wi-Fi, Bluetooth, 4G, or 5G communication module.

FIG. 4 is a schematic internal block diagram of the server 200 in FIG. 3 . As shown in Figure 4, the history database 210 is stored a plurality of history information on the ready-mixed concrete quality determination. The history information includes 730,000 learning images for machine learning, and images received from multiple workplaces and judgment results are stored. The history database 210 is continuously updated and provides the latest information to the model learning unit 220.

The model learning unit 220 generates an artificial intelligence inference model by machine learning using the history information from the history database 210. In addition, the model learning unit 220 is a normal region 62 when the shade of the aggregate in the image of the surface of the ready-mixed concrete indicates a density equal to or higher than the threshold value, and a defective region 64 when the threshold value is not reached. generate The generated inference model is a Convolutional Neural Network (CNN) model applied with various technologies such as a neural network model, shortcut connection or skip connection, and cross-layer connections. It can be.

The model learning unit 220 performs a first convolution step using history data, a first convolution step using the first activation function, and a second convolution step using the calculated data to generate an inference model of the neural network. , calculating using the second activation function (S134); pooling the calculated data; And a step of dropping out the neural network model is sequentially performed.

In the embodiment of the present invention, it was composed of six neural network layers, and included pooling and dropout for each layer. The activation function is a function that converts the sum of input signals entering individual neurons of the neural network into an output signal, and ReLU (rectified linear unit) was used for the first and second activation functions. Various optimization techniques can be used, but in this example, the neural network optimization process was performed using Adam Optimizer. Learning and verification are repeated through such a machine learning process, and when an accuracy of 99.99% or more is obtained, learning is terminated and the learned inference model 230 is confirmed.

The transmission/update unit 240 transmits the inference model to the application program 140 in the workplace or updates the latest inference model at regular intervals. In addition, the transmission/update unit 240 receives field data 110 from the communication unit 190 of the workplace and transmits it to the server control unit 250 . Field data 110 received by the transmission / update unit 240 is stored in each database. For example, the worker database 260 stores log-in data and worker information (eg, ID, password, mobile phone number, e-mail, name, position, etc.) input by the worker 125 . The workshop database 270 stores workshop information (eg, workshop location or address, tools, etc.). The vehicle database 280 includes ready-mixed concrete vehicle information (eg, company name, license plate number, driver) and operation date and time information.

Operation of the embodiment

Hereinafter, the operation of the preferred embodiment will be described in detail with reference to the accompanying drawings. 6 is a flowchart schematically illustrating a method for determining an image of ready-mixed concrete quality using artificial intelligence according to an embodiment of the present invention. As shown in Figure 6, first, when the ready-mixed concrete 40 is introduced from the ready-mixed concrete vehicle 20 to the pump car 10, the camera 100 takes a video of the hopper 11 (S100). In the video, the surface of the ready-mixed concrete 40 in flow through input and stirring is photographed.

Next, the image processing unit 130 extracts an image from the image of the camera 100 and extracts a region of interest 60 from the image (S110).

Then, the inference model 150 divides the surface of the ready-mixed concrete 40 into a normal region 62 and a defective region 64 based on the shade ratio of the aggregate exposed on the surface of the ready-mixed concrete 40 with respect to the region of interest 60. Inference to divide (S120).

Figure 5a is a picture of the area of interest 60 according to the present invention determined to be defective ready-mixed concrete, Figure 5b is a picture of the area of interest 60 according to the present invention determined to be normal ready-mixed concrete. In FIGS. 5A and 5B, S135, S170, S190, S240, F350, F630, and F800 indicate grades of concrete [strength (S, F)-slump value], and among them, S135, S170, S190, F350, and F630 are “ OK”.

As shown in FIGS. 5A and 5B, it can be seen that there are a number of aggregate shadows on the surface of the ready-mixed concrete 40 in the normal region 62, and material separation occurs in the defective region 64, so the shadow of the aggregate is small. It can be confirmed that it is a ready-mixed concrete with a poor or almost smooth surface. On the other hand, while a stirrer (not shown) rotates inside the hopper 11, the ready-mixed concrete is continuously stirred up, down, left and right.

Next, the controller 180 calculates the area ratio of the normal region 62 and the defective region 64 (S130). “OK 5%” in FIG. 5A indicates that the area ratio of the normal region 62 is 5%, and “NOT OK 95.0%” indicates that the area ratio of the defective region 64 is 95.0%. Also, “OK 30%” shown in FIG. 5B indicates that the area ratio of the normal region 62 is 30.0%, and “NOT OK 70.0%” indicates that the area ratio of the defective region 64 is 70.0%.

Then, the controller 180 determines and outputs the quality of ready-mixed concrete based on the calculated ratio (30:70) (S140). At this time, if the ratio of the normal area 62 is 30% or more, it is determined as normal ready-mixed concrete, and if it is less than 30%, it is determined as defective ready-mixed concrete. Therefore, FIG. 5a is determined as defective ready-mixed concrete, and FIG. 5b is determined as normal ready-mixed concrete. The standard value of 30% may be appropriately selected in the range of 10 to 40% depending on the type of ready-mixed concrete, dough consistency, season, and construction specifications.

If the controller 180 determines that the ready-mixed concrete is defective, a warning is generated (S150). The warning may be an alarm sound and/or vibration of the mobile phone 120, a red light indicator on the display 170, and the like.

The communication unit 190 transmits the image used in the inference and the result of the judgment to the server 200 to be stored in the history database 210, the worker database 260, the workshop database 270, and the ready-mixed concrete vehicle database 280, respectively. do.

In one embodiment of the present invention, the artificial intelligence inference model divides and infers the normal area “OK” and the defective area “NOT OK”, but it can be modified to learn and infer more precisely by subdividing the area where material separation has partially occurred. can

Detailed descriptions of the preferred embodiments of the present invention disclosed as described above are provided to enable those skilled in the art to implement and practice the present invention. Although the above has been described with reference to preferred embodiments of the present invention, those skilled in the art will understand that the present invention can be variously modified and changed without departing from the scope of the present invention. For example, those skilled in the art can use each configuration described in the above-described embodiments in a manner of combining with each other. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

The present invention may be embodied in other specific forms without departing from the spirit and essential characteristics of the present invention. Accordingly, the above detailed description should not be construed as limiting in all respects and should be considered as illustrative. The scope of the present invention should be determined by reasonable interpretation of the appended claims, and all changes within the equivalent scope of the present invention are included in the scope of the present invention. The invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein. In addition, claims that do not have an explicit citation relationship in the claims may be combined to form an embodiment or may be included as new claims by amendment after filing.

1: formwork,
10: pump car,
11: Hopper,
12: Boom,
13: end hose,
20: ready-mixed concrete vehicle,
30: frame,
32: rib,
40: ready-mixed concrete,
45: area where ready-mixed concrete is being put in,
50: exclusion area,
60: region of interest (ROI),
62: normal area,
64: defective area,
100: camera,
110: field data,
120: mobile phone,
125: worker,
130: image processing unit,
140: application,
150: artificial intelligence reasoning model,
160: output unit,
170: display,
180: control unit,
190: Ministry of Communications,
200: server,
210: history database,
220: model learning unit,
230: learned inference model,
240: transmission/update unit,
250: server control unit,
260: worker database,
270: workshop database,
280: vehicle database.

Claims (15)

When the ready-mixed concrete 40 is injected from the ready-mixed concrete vehicle 20 to the pump car 10, the camera 100 for photographing the hopper 11;
An artificial intelligence inference model that infers that the surface of the ready-mixed concrete 40 is divided into a normal region 62 and a defective region 64 based on the ratio of aggregate exposed to the surface of the ready-mixed concrete 40 in the image of the camera 100. (150); and
A control unit 180 that calculates the area ratio of the normal area 62 and the defective area 64 and determines the quality of the ready-mixed concrete based on the calculated ratio; Quality image determination device. According to claim 1,
The camera 100 is an image determination device of ready-mixed concrete quality using artificial intelligence, characterized in that the camera of the mobile phone 120 mounted on the ready-mixed concrete vehicle 20 or carried by the worker 125. According to claim 1,
The image of the camera 100, among the flat images of the hopper 11, an exclusion region 50 not used for inference and a region of interest 60 used for inference; Remicon quality video judgment device. According to claim 3,
The exclusion area 50 comprises an area 45 into which the remicon 40 is injected from the remicon vehicle 20. Remicon quality image determination device using artificial intelligence. According to claim 3,
Extracting an image of each frame from the image of the camera 100,
An image determination device of ready-mixed concrete quality using artificial intelligence, characterized in that it further comprises an image processing unit 130 for extracting the region of interest 60 from the extracted image. According to claim 1,
The artificial intelligence inference model 150 is an image determination device of ready-mixed concrete quality using artificial intelligence, characterized in that inferring based on the shading ratio of the aggregate exposed to the surface of the ready-mixed concrete 40 among the images of the camera 100. According to claim 6,
The artificial intelligence inference model 150 infers as the normal area 62 when the shade of the aggregate indicates a density equal to or higher than a threshold value, and infers it as a defective area 64 when the shade does not reach the threshold value. Image determination device of ready-mixed concrete quality using. A history database 210 in which a plurality of history information about the quality of ready-mixed concrete is stored;
A model learning unit 220 generating an artificial intelligence inference model by machine learning using the plurality of history information;
a transmission/update unit 240 that transmits the inference model to the application program 140 in the workplace or updates it with the latest inference model;
Database means (260, 270, 280) for storing the worker information, the workshop information and the ready-mixed concrete vehicle 20 information input by the worker 125 of the workshop through the application 140; and
A server control unit 250 for storing the information of the database means and the image used by the inference model 150 for determination of the workplace and the determination result in the history database 210; Learning server device of artificial intelligence inference model for image judgment. According to claim 8,
The model learning unit 220 is a normal region 62 when the shade of the aggregate in the image of the surface of the ready-mixed concrete indicates a density equal to or higher than a threshold value, and the artificial intelligence inference model machine-learned as a defective region 64 when it does not reach the threshold value A learning server device of an artificial intelligence inference model for image determination of ready-mixed concrete quality, characterized in that for generating. According to claim 8,
The workplace is at least one workplace located in a remote location,
The application program 140 is a learning server device of an artificial intelligence inference model for image determination of ready-mixed concrete quality, characterized in that running on at least one of a personal computer, a tablet PC, and a mobile phone 120 located in the workplace. When the ready-mixed concrete 40 is introduced from the ready-mixed concrete vehicle 20 to the pump car 10, the camera 100 photographs the hopper 11 (S100);
The image processing unit 130 extracts an image from the image of the camera 100 and extracts a region of interest 60 from the image (S110);
The inference model 150 divides the surface of the ready-mixed concrete 40 into a normal region 62 and a defective region 64 based on the shade ratio of the aggregate exposed on the surface of the ready-mixed concrete 40 with respect to the region of interest 60 inferring (S120); and
calculating, by the controller 180, an area ratio between the normal area 62 and the defective area 64 (S130); and
The control unit 180 determines and outputs the quality of the ready-mixed concrete based on the calculated ratio (S140); method for determining the image of the ready-mixed concrete quality using artificial intelligence, characterized in that it comprises a. According to claim 11,
In the inference step (S120), when the shade of the aggregate indicates a density equal to or higher than a threshold value, it is inferred as the normal region (62), and when it is less than the threshold value, it is inferred as a defective region (64). A method for determining the image quality of ready-mixed concrete. According to claim 11,
In the determination step (S140), if the ratio of the normal area 62 is 30% or more, it is determined as normal remicon, and if it is less than 30%, it is determined as defective remicon. According to claim 13,
When it is determined that the defective ready-mixed concrete is determined, the control unit 180 generates a warning (S150). According to claim 11,
The communication unit 190 further comprises the step of transmitting the image used for inference and the result of the determination to the server 200.

Images