Workshop conveying management method, system and equipment based on industrial internet of things, and medium

Abstract

The invention discloses a workshop conveying management method, system and equipment based on industrial Internet of Things and a medium, and relates to the technical field of image data processing, and the method comprises the following steps: obtaining a top surface image of target angle steel in a set conveying state, so as to recognize whether the target angle steel has a first deformation feature; if yes, the target angle steel is controlled to be conveyed to a first reworking area; if not, acquiring an end face image of the target angle steel in the set conveying state to identify whether the target angle steel has a second deformation feature; if yes, the target angle steel is controlled to be conveyed to a second reworking area; if not, acquiring a depth image of the target angle steel in a set conveying state based on a preset visual angle to identify whether the target angle steel has a third deformation feature; if yes, the target angle steel is controlled to be conveyed to a third reworking area; the method has the advantages that the quality inspection accuracy is improved, and classified conveying can be carried out according to defect types.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and particularly to a workshop conveying management method, system, device and medium based on industrial Internet of Things. Background Art

[0002] Angle steel is divided into equal-angle angle steel and unequal-angle angle steel. Angle steel can form various load-bearing members according to different structural requirements, and can also be used as connectors between members. It is widely used in various building structures and engineering structures, such as roof beams, bridges, transmission towers, hoisting and transportation machinery, ships, industrial furnaces, reaction towers, container racks, and warehouse shelves, etc.

[0003] In the workshop conveying pipeline, angle steel is generally inspected manually by visual inspection to screen out defective angle steel. Then, this part of the defective angle steel is marked and conveyed to the rework area for subsequent centralized rework and correction, while the qualified angle steel can be directly conveyed to the qualified area for subsequent factory packaging. However, the method of manual visual inspection not only consumes labor, but also has the problem of low inspection accuracy, which affects the quality of angle steel leaving the factory. Moreover, it is difficult to classify the types of angle steel defects in a timely manner during the conveying pipeline, and the type of defect needs to be searched again during subsequent rework and correction, which affects the rework efficiency. Summary of the Invention

[0004] The main purpose of the present invention is to provide a workshop conveying management method, system, device and medium based on industrial Internet of Things, aiming to solve the technical problem of low inspection accuracy in the existing angle steel conveying pipeline by manual visual inspection.

[0005] To achieve the above object, the present invention provides a workshop conveying management method based on industrial Internet of Things, including the following steps:

[0006] Obtain the top surface image of the target angle steel in a set conveying state; wherein, the set conveying state is the state where the target angle steel is placed in an inverted V shape on the conveyor;

[0007] According to the top surface image, identify whether the target angle steel has a first deformation feature; wherein, the first deformation feature is the feature that the target angle steel has a bending deformation in the length direction;

[0008] If there is a first deformation feature, mark the target angle steel as a first-class defective product and control the target angle steel to be conveyed to the first rework area; if there is no first deformation feature, obtain the end surface image of the target angle steel in the set conveying state;

[0009] According to the end surface image, identify whether the target angle steel has a second deformation feature; wherein, the second deformation feature is the feature that the two side plates of the target angle steel are deformed;

[0010] If there is a second deformation feature, mark the target angle steel as a second-class defective product and control the target angle steel to be transported to the second rework area; if there is no second deformation feature, obtain the depth image of the target angle steel in the set transportation state based on a preset perspective; where the preset perspective is a perspective above the target angle steel and perpendicular to the side plate of the target angle steel;

[0011] According to the depth image, identify whether the target angle steel has a third deformation feature; where the third deformation feature is that the side plate of the target angle steel has pits or protrusions;

[0012] If there is a third deformation feature, mark the target angle steel as a third-class defective product and control the target angle steel to be transported to the third rework area; if there is no third deformation feature, mark the target angle steel as a qualified product and control the target angle steel to be transported to the qualified area.

[0013] Optionally, according to the top surface image, identify whether the target angle steel has a first deformation feature, including:

[0014] According to the top surface image, extract the reference contour line of the vertex angle corresponding to the target angle steel and the edge contour line of any side plate corresponding to the target angle steel; where the reference contour line is the vertex angle contour line formed by the non-deformed part of the target angle steel;

[0015] Select multiple measurement points in the edge contour line and obtain the first distance value from each measurement point to the reference contour line; where the measurement points at least include the two end points of the edge contour line, and the multiple measurement points are evenly spaced;

[0016] Judge whether multiple first distance values are all equal to a preset first standard value. If so, identify that the target angle steel has no first deformation feature; if not, identify that the target angle steel has a first deformation feature.

[0017] Optionally, let the number of measurement points be n, then the expression of n is:

[0018] n = K * L;

[0019] In the formula, K is a proportionality coefficient, and L is the length of the target angle steel.

[0020] Optionally, according to the end face image, identify whether the target angle steel has a second deformation feature, including:

[0021] According to the end face image, identify the end face contour of the target angle steel;

[0022] Identify the second distance value from the vertex angle point to the transportation horizontal line in the end face contour; where the vertex angle point is the common point of the included angle positions of the two side plates of the target angle steel, and the transportation horizontal line is the horizontal contour line formed by the plane where the conveyor contacts the target angle steel;

[0023] Determine whether the second spacing value is equal to a preset second standard value. If so, it is recognized that the target angle steel does not have a second deformation feature. If not, it is recognized that the target angle steel has a second deformation feature.

[0024] Optionally, based on the depth image, determine whether the target angle steel has a third deformation feature, including:

[0025] Based on the depth image, obtain the measured depth values of multiple pixel points on the side plate of the target angle steel;

[0026] Filter out the standard depth value among the multiple measured depth values; wherein, the standard depth value is the pixel point depth value corresponding to the part of the side plate of the target angle steel without pits or protrusions;

[0027] Filter out the target depth value among the multiple measured depth values; wherein, the target depth value is the measured depth value that is greater than and / or less than the standard depth value;

[0028] Based on the depth image, obtain the pit area value and / or protrusion area value of the side plate of the target angle steel;

[0029] Based on the target depth value, pit area value and / or protrusion area value, obtain the concavo-convex deformation degree;

[0030] Determine whether the concavo-convex deformation degree is less than a preset third standard value. If so, it is recognized that the target angle steel does not have a third deformation feature. If not, it is recognized that the target angle steel has a third deformation feature.

[0031] Optionally, let the concavo-convex deformation degree be E, then the expression of E is:

[0032] E = λ1 * |H - h| + λ2 * S;

[0033] In the formula, H is the standard depth value, h is the target depth value, S is the pit area value and / or protrusion area value, λ1 is the first adjustment coefficient, and λ2 is the second adjustment coefficient.

[0034] Optionally, filtering out the standard depth value among the multiple measured depth values includes:

[0035] Respectively divide the equal measured depth values into the same set to obtain multiple groups of depth data sets;

[0036] Obtain the number of monomers of the measured depth value in each group of depth data sets;

[0037] Output the measured depth value with the largest number of monomers as the standard depth value.

[0038] To achieve the above object, the present invention also provides a workshop conveying management system based on the industrial Internet of Things, including a management platform, a sensing network platform, and an object platform that are sequentially communicatively connected. The management platform includes:

[0039] An image acquisition module for acquiring a top surface image of a target angle steel in a set conveying state; wherein, the set conveying state is a state in which the target angle steel is placed in an inverted V shape on a conveyor.

[0040] A first feature recognition module for recognizing whether the target angle steel has a first deformation feature according to the top surface image; wherein, the first deformation feature is a feature that the target angle steel has a bending deformation in the length direction.

[0041] A first data processing module, if there is a first deformation feature, marking the target angle steel as a first type of defective product and controlling the target angle steel to be conveyed to a first rework area; if there is no first deformation feature, acquiring an end face image of the target angle steel in a set conveying state.

[0042] A second feature recognition module for recognizing whether the target angle steel has a second deformation feature according to the end face image; wherein, the second deformation feature is a feature that the two side plates of the target angle steel are deformed.

[0043] A second data processing module, if there is a second deformation feature, marking the target angle steel as a second type of defective product and controlling the target angle steel to be conveyed to a second rework area; if there is no second deformation feature, acquiring a depth image of the target angle steel in a set conveying state based on a preset perspective; wherein, the preset perspective is a perspective above the target angle steel and perpendicular to the side plates of the target angle steel.

[0044] A third feature recognition module for recognizing whether the target angle steel has a third deformation feature according to the depth image; wherein, the third deformation feature is that the side plates of the target angle steel have pits or protrusions.

[0045] A third data processing module, if there is a third deformation feature, marking the target angle steel as a third type of defective product and controlling the target angle steel to be conveyed to a third rework area; if there is no third deformation feature, marking the target angle steel as a qualified product and controlling the target angle steel to be conveyed to a qualified area.

[0046] To achieve the above object, the present invention also provides a computer device, which includes a memory and a processor. A computer program is stored in the memory, and the processor executes the computer program to implement the above method.

[0047] To achieve the above object, the present invention also provides a computer-readable storage medium, on which a computer program is stored, and the processor executes the computer program to implement the above method.

[0048] The beneficial effects that the present invention can achieve are as follows:

[0049] In view of various possible quality defects of angle steel, in order to facilitate the acquisition of image features of angle steel from multiple perspectives, the target angle steel to be detected is placed on a conveyor in an inverted V shape for transportation. During the transportation process, the top surface image corresponding to the target angle steel is obtained from directly above the target angle steel first, so as to facilitate the detection of whether there are defect features of bending deformation in the length direction of the target angle steel. If so, the target angle steel is marked as the first type of defective product, and the transportation of the target angle steel to the first rework area is controlled, so as to directly correct the defect feature subsequently. If not, the end face image corresponding to the target angle steel is obtained from the end perspective of the target angle steel. The end face image can represent the contour features of the two side plates of the target angle steel, so as to facilitate the detection of whether there are defect features of deformation on the two side plates. If so, the target angle steel is marked as the second type of defective product, and the transportation of the target angle steel to the second rework area is controlled, so as to perform targeted correction subsequently. If not, the depth image corresponding to the target angle steel is obtained from the perspective perpendicular to the side plate of the target angle steel. Based on the depth image, it can be characterized whether there are pits or protrusions on the side plate of the target angle steel. If so, the target angle steel is marked as the third type of defective product, and the transportation of the target angle steel to the third rework area is controlled. If not, it means that the target angle steel does not have the above three defects, and then the target angle steel can be marked as a qualified product, and the transportation of the target angle steel to the qualified area is controlled. To sum up, the present invention sequentially identifies three common defects of angle steel based on machine vision recognition technology, improves the recognition accuracy, avoids human errors, and transports the angle steel to different areas based on the recognition results, so as to perform targeted defect correction subsequently, saving the time for re-finding the defect type and improving the quality management efficiency. Description of the Drawings

[0050] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts do not necessarily draw according to the actual scale.

[0051] Figure 1 It is a schematic flow chart of the workshop transportation management method based on industrial Internet of Things in the embodiment of the present invention;

[0052] Figure 2 It is a schematic diagram of the target angle steel in the set transportation state in the embodiment of the present invention;

[0053] Figure 3 It is a schematic diagram of identifying whether there is a first deformation feature of the target angle steel based on the top surface image in the embodiment of the present invention;

[0054] Figure 4 It is a schematic diagram of identifying whether there is a second deformation feature of the target angle steel based on the end face image in the embodiment of the present invention;

[0055] Figure 5 Schematic diagram of another form of the second deformation feature recognized based on the end face image in the embodiment of the present invention;

[0056] Figure 6 Schematic diagram when the depth camera in the embodiment of the present invention collects the depth image of the side plate of the target angle steel (when the side plate has a pit);

[0057] Figure 7 Schematic diagram when the depth camera in the embodiment of the present invention collects the depth image of the side plate of the target angle steel (when the side plate has a protrusion);

[0058] Figure 8 Schematic diagram of the framework of the Internet of Things system involved in the embodiment of the present invention.

[0059] Reference numerals:

[0060] 110 - Target angle steel, 120 - Conveyor, 130 - Industrial camera, 140 - Reference contour line, 150 - Edge contour line, 160 - Measurement point, 170 - Vertex point, 180 - Conveyor horizontal line, 190 - Depth camera.

[0061] The realization, functional features and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. Detailed implementation manners

[0062] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0063] If there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include at least one of such features. In addition, the meaning of "and / or" appearing throughout the text includes three parallel solutions. Taking "A and / or B" as an example, it includes solution A, or solution B, or a solution where A and B are satisfied simultaneously. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the fact that those of ordinary skill in the art can implement it. When the combination of technical solutions conflicts with each other or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the protection scope required by the present invention.

[0064] Embodiment 1

[0065] Referring to Figures 1 - 7 , this embodiment provides a workshop conveying management method based on the industrial Internet of Things, including the following steps:

[0066] Obtain the top surface image of the target angle steel 110 in a set conveying state; wherein, the set conveying state is the state where the target angle steel 110 is placed in an inverted V shape on the conveyor 120;

[0067] According to the top surface image, identify whether the target angle steel 110 has a first deformation feature; wherein, the first deformation feature is the feature that the target angle steel 110 has a bending deformation in the length direction;

[0068] If the first deformation feature exists, mark the target angle steel 110 as the first type of defective product and control the target angle steel 110 to be conveyed to the first rework area; if the first deformation feature does not exist, obtain the end face image of the target angle steel 110 in the set conveying state;

[0069] According to the end face image, identify whether the target angle steel 110 has a second deformation feature; wherein, the second deformation feature is the feature that the two side plates of the target angle steel 110 are deformed;

[0070] If the second deformation feature exists, mark the target angle steel 110 as the second type of defective product and control the target angle steel 110 to be conveyed to the second rework area; if the second deformation feature does not exist, obtain the depth image of the target angle steel 110 in the set conveying state based on a preset perspective; wherein, the preset perspective is the perspective above the target angle steel 110 and perpendicular to the side plate of the target angle steel 110;

[0071] According to the depth image, identify whether the target angle steel 110 has a third deformation feature; wherein, the third deformation feature is that the side plate of the target angle steel 110 has a pit or a protrusion;

[0072] If the third deformation feature exists, mark the target angle steel 110 as the third type of defective product and control the target angle steel 110 to be conveyed to the third rework area; if the third deformation feature does not exist, mark the target angle steel 110 as a qualified product and control the target angle steel 110 to be conveyed to the qualified area.

[0073] Due to various factors (such as collision, process error, etc.) affecting the angle steel during transportation or production, it is easy to have common quality defects such as overall bending deformation of a certain section of the angle steel, overall deformation of the side plates, and local unevenness of the side plates. In the scenario of conveying and moving, it is difficult for manual visual quality inspection to meet the standards of complete inspection and accurate inspection, thus resulting in low quality of the angle steel leaving the factory.

[0074] In this embodiment, considering various possible quality defects of the angle steel, in order to facilitate the acquisition of the image features of the angle steel from multiple perspectives respectively, the target angle steel 110 to be detected is placed on the conveyor 120 in an inverted V shape for conveying. During the conveying process, the corresponding top surface image is obtained from the perspective directly above the target angle steel 110, so as to facilitate the detection of whether there are defect features of bending deformation in the length direction of the target angle steel 110. If so, the target angle steel 110 is marked as a first-class defective product, and the conveyance of the target angle steel 110 is controlled to the first rework area for subsequent direct correction of this defect feature. If not, the corresponding end surface image is obtained from the perspective of the end of the target angle steel 110. The end surface image can represent the contour features of the two side plates of the target angle steel 110, so as to facilitate the detection of whether there are defect features of deformation on the two side plates. If so, the target angle steel 110 is marked as a second-class defective product, and the conveyance of the target angle steel 110 is controlled to the second rework area for subsequent targeted correction. If not, the corresponding depth image is obtained from the perspective perpendicular to the side plate of the target angle steel 110. Based on the depth image, it can be characterized whether there are pits or protrusions on the side plate of the target angle steel 110. If so, the target angle steel 110 is marked as a third-class defective product, and the conveyance of the target angle steel 110 is controlled to the third rework area. If not, it indicates that the target angle steel 110 does not have the above three defects, and then the target angle steel 110 can be marked as a qualified product, and the conveyance of the target angle steel 110 is controlled to the qualified area. To sum up, this embodiment uses machine vision recognition technology to identify three common defects of the angle steel in sequence, improving the recognition accuracy, avoiding human errors, and conveying the angle steel to different areas based on the recognition results for subsequent targeted defect correction, saving the time for re-determining the defect type and improving the quality management efficiency.

[0075] It should be noted that the reason for detecting in the above-mentioned detection order of the first deformation feature, the second deformation feature, and the third deformation feature in this embodiment is that when the target angle steel 110 has a bending deformation in the length direction, it will affect the detection accuracy of the second deformation feature and the third deformation feature. Therefore, only after the first deformation feature is detected to be qualified, can the next defect detection be carried out. Similarly, when the side plate of the target angle steel 110 is deformed, it will also affect the detection accuracy of the third deformation feature. Therefore, the third deformation feature can only be detected after the second deformation feature is detected to be qualified. Therefore, in this embodiment, according to the characteristics of each quality defect of the angle steel, the detection steps of the defect features are optimized, ensuring the detection accuracy and being more reasonable. When obtaining the top surface image and the end surface image, they can both be collected by the industrial camera 130. Just arrange the industrial camera 130 at the corresponding position of the conveyor 120. When obtaining the depth image, it can be collected by the depth camera 190. Since the two side plates of the angle steel are generally vertical structures, two depth cameras 190 are used here and are respectively arranged on the two side frames of the conveyor 120 at an angle of 45° to collect the depth images of the two side plates of the angle steel from a vertical perspective.

[0076] As an optional implementation manner, to identify whether the target angle steel 110 has the first deformation feature according to the top surface image, it includes:

[0077] According to the top surface image, extract the reference contour line 140 of the vertex angle corresponding to the target angle steel 110 and the edge contour line 150 of any one side plate corresponding to the target angle steel 110; among them, the reference contour line 140 is the vertex angle contour line formed by the non-deformed part of the target angle steel 110.

[0078] Select multiple measurement points 160 on the edge contour line 150, and obtain the first distance value from each measurement point 160 to the reference contour line 140; among them, the measurement points 160 at least include the two end points of the edge contour line 150, and the multiple measurement points 160 are evenly spaced.

[0079] Judge whether multiple first distance values are all equal to the preset first standard value. If so, it is identified that the target angle steel 110 does not have the first deformation feature. If not, it is identified that the target angle steel 110 has the first deformation feature.

[0080] In this embodiment, the top surface image here can be a black-and-white contour image after image processing, so as to visually display the lengthwise contour features formed by the target angle steel 110 from the top view. A contour line can be formed at its vertex position. The vertex contour line formed by selecting the non-deformed part among them, that is, the straight part is the reference contour line 140. Then, multiple measurement points 160 are selected from the edge contour line 150, and the first distance value from each measurement point 160 to the reference contour line 140 is obtained. If bending deformation occurs, the first distance value between the measurement point 160 on the edge contour line 150 of the deformed section and the reference contour line 140 will be larger or smaller. Thus, the first deformation feature can be characterized by numerical features. Compared with the image comparison method, the calculation accuracy is higher, and the detection accuracy is improved.

[0081] It should be noted that since the deformed sections mostly occur at both ends of the angle steel, the measurement points 160 should at least include the two end points of the edge contour line 150.

[0082] As an alternative embodiment, let the number of measurement points 160 be n, then the expression of n is:

[0083] n = K * L;

[0084] In the formula, K is the proportionality coefficient, and L is the length of the target angle steel 110.

[0085] In this embodiment, the number n of measurement points 160 should be proportional to the length L of the target angle steel 110. The more the number n, the higher the detection accuracy, but the more the number n, the lower the data processing efficiency. Therefore, to balance the detection accuracy and data processing efficiency, a proportionality coefficient K (which can be adjusted and selected according to the actual situation) can be set here, and a reasonable number of measurement points 160 can be selected according to the length of the target angle steel 110 to meet the actual needs.

[0086] As an alternative embodiment, according to the end face image, it is identified whether the target angle steel 110 has a second deformation feature, including:

[0087] According to the end face image, the end face contour of the target angle steel 110 is identified;

[0088] The second distance value from the vertex point 170 to the conveying horizontal line 180 in the end face contour is identified; among them, the vertex point 170 is the common point at the included angle position of the two side plates of the target angle steel 110, and the conveying horizontal line 180 is the horizontal contour line formed by the plane where the conveyor 120 contacts the target angle steel 110;

[0089] It is judged whether the second distance value is equal to a preset second standard value. If so, it is identified that the target angle steel 110 does not have a second deformation feature. If not, it is identified that the target angle steel 110 has a second deformation feature.

[0090] In this embodiment, the end face contour can also be a black-and-white contour image after image processing, which is convenient for extracting the end face contour features of the target angle steel 110. Generally, there are two cases of side plate deformation. The first is that the included angle between the two side plates is too large or too small, and the other is the bending deformation of the side plate (as shown in Figure 5 ). However, no matter which kind of deformation occurs, it will cause the distance between the apex point 170 and the conveying horizontal line 180 (equivalent to the reference line) in the end face contour to change. Therefore, by identifying the second distance value from the apex point 170 to the conveying horizontal line 180 and judging whether the second distance value is equal to the second standard value, the above two deformation cases can be characterized simultaneously. Not only is the detection accuracy high, but there is no need to detect the two deformation cases separately, which improves the detection efficiency.

[0091] As an alternative embodiment, according to the depth image, it is identified whether the target angle steel 110 has a third deformation feature, including:

[0092] According to the depth image, obtain the measured depth values of multiple pixel points of the side plate of the target angle steel 110;

[0093] Screen out the standard depth value among the multiple measured depth values; wherein, the standard depth value is the depth value of the pixel point corresponding to the part of the side plate of the target angle steel 110 without pits or protrusions;

[0094] Screen out the target depth value among the multiple measured depth values; wherein, the target depth value is the measured depth value that is greater than and / or less than the standard depth value the most;

[0095] According to the depth image, obtain the pit area value and / or protrusion area value of the side plate of the target angle steel 110;

[0096] According to the target depth value, the pit area value and / or the protrusion area value, obtain the concavo-convex deformation degree;

[0097] Judge whether the concavo-convex deformation degree is less than a preset third standard value. If so, it is identified that the target angle steel 110 does not have a third deformation feature. If not, it is identified that the target angle steel 110 has a third deformation feature.

[0098] In this embodiment, since the depth image is a grayscale image containing the distance information of each pixel point from the depth camera 190, the measured depth values of multiple pixel points on the side plate of the target angle steel 110 can be obtained according to the depth image. First, the standard depth value corresponding to the part of the target angle steel 110 without pit or protrusion defect features is determined, and then the target depth values corresponding to the parts with pit or protrusion defect features are screened out. Considering that the side plate may have pits or protrusions separately, or may have both pits and protrusions at the same time, and the depth values corresponding to the pits and protrusions are different, the measured depth value that is greater than and / or less than the standard depth value at most is used as a typical value to output as the target depth value. Then, considering the influence of the deformation area (i.e., the pit area value or the protrusion area value), the concavo-convex deformation degree can be calculated by combining the two parameters to characterize the deformation degree. Considering the error influence and different quality requirements, a third standard value is set here. When the concavo-convex deformation degree is less than the third standard value, it can also be identified as qualified, that is, a certain degree of concavo-convex deformation defect is allowed, so as to realize the accurate detection and identification of the third deformation feature.

[0099] It should be noted that the total concavo-convex deformation degree of the entire target angle steel 110 is obtained by adding the concavo-convex deformation degrees calculated for the two side plates respectively; when it is recognized that the target depth values that are greater than and less than the standard depth value at most exist on the same side plate at the same time, it means that there are both pits and protrusions. Then, after calculating the pit area value and the protrusion area value respectively, the concavo-convex deformation degrees corresponding to the corresponding pits and protrusions are calculated respectively, and the two concavo-convex deformation degrees are added to obtain the final total concavo-convex deformation degree of the side plate.

[0100] As an alternative embodiment, let the concavo-convex deformation degree be E, then the expression of E is:

[0101] E = λ1 * |H - h| + λ2 * S;

[0102] In the formula, H is the standard depth value, h is the target depth value, S is the pit area value and / or the protrusion area value, λ1 is the first adjustment coefficient, and λ2 is the second adjustment coefficient.

[0103] In this embodiment, since the target depth value is the value that is greater than or less than the standard depth value at most, the absolute value of the difference between the two is used here to represent the concavo-convex depth value, that is, |H - h|. At the same time, considering that the pit area value and the protrusion area value and the concavo-convex depth value belong to values with different unit attributes, the first adjustment coefficient λ1 and the second adjustment coefficient λ2 are used here for conversion and adjustment respectively, so that the two parameters can be superimposed on each other to obtain the final concavo-convex deformation degree E, which can be quantitatively calculated and improves the characterization accuracy of the concavo-convex defect features.

[0104] As an alternative embodiment, screening out the standard depth value among multiple measured depth values includes:

[0105] Divide equal measured depth values into the same set respectively to obtain multiple groups of depth data sets;

[0106] Obtain the number of monomers of the measured depth values in each group of depth data sets;

[0107] Output the measured depth value with the largest number of monomers as the standard depth value.

[0108] In this embodiment, when determining the standard depth value, since most parts of the side plates are flat parts without local deformation (i.e., pits or protrusions), here, multiple obtained measured depth values can be grouped equally to form multiple groups of depth data sets. The measured depth value with the most equal value occurrences (i.e., the largest number of monomers) can be directly output as the standard depth value. Therefore, the confirmation of the standard depth value can be achieved by using data characteristics. It is not necessary to place the angle steel in the exact middle of the conveyor 120. Just make the edge of the angle steel parallel to the conveying direction of the conveyor 120, which reduces the detection difficulty. Based on this, the standard depth values recognized by the two side plates may be different.

[0109] Embodiment 2

[0110] Based on the same inventive concept as the foregoing embodiment, referring to Figures 1 - 8 , this embodiment further provides an industrial Internet of Things-based workshop conveying management system, including a management platform, a sensing network platform, and an object platform that are communicatively connected in sequence. The management platform includes:

[0111] An image acquisition module, configured to acquire a top surface image of the target angle steel 110 in a set conveying state; wherein, the set conveying state is a state where the target angle steel 110 is placed in an inverted V shape on the conveyor 120;

[0112] A first feature recognition module, configured to recognize whether the target angle steel 110 has a first deformation feature according to the top surface image; wherein, the first deformation feature is a feature that the target angle steel 110 has a bending deformation in the length direction;

[0113] A first data processing module, configured to, if the first deformation feature exists, mark the target angle steel 110 as a first-class defective product and control the target angle steel 110 to be conveyed to the first rework area; if the first deformation feature does not exist, acquire an end surface image of the target angle steel 110 in the set conveying state;

[0114] A second feature recognition module, configured to recognize whether the target angle steel 110 has a second deformation feature according to the end surface image; wherein, the second deformation feature is a feature that the two side plates of the target angle steel 110 are deformed;

[0115] A second data processing module, configured to, if there is a second deformation feature, mark the target angle steel 110 as a second type of defective product and control the target angle steel 110 to be conveyed to a second rework area; if there is no second deformation feature, obtain a depth image of the target angle steel 110 in a set conveying state based on a preset perspective; wherein, the preset perspective is a perspective above the target angle steel 110 and perpendicular to the side plate of the target angle steel 110;

[0116] A third feature recognition module, configured to identify whether there is a third deformation feature on the target angle steel 110 according to the depth image; wherein, the third deformation feature is that the side plate of the target angle steel 110 has a pit or a protrusion;

[0117] A third data processing module, configured to, if there is a third deformation feature, mark the target angle steel 110 as a third type of defective product and control the target angle steel 110 to be conveyed to a third rework area; if there is no third deformation feature, mark the target angle steel 110 as a qualified product and control the target angle steel 110 to be conveyed to a qualified area.

[0118] It should be noted that the workshop conveying management system based on the industrial Internet of Things in this embodiment further includes a user platform and a service platform that are communicatively connected to each other, and the service platform is communicatively connected to the management platform, thus forming a standard five-platform structure of the Internet of Things. Among them, the physical entities of the user platform include various user terminals, such as mobile phones, computers, dedicated terminals, etc., and realize the services at the user end through the combination with the user information system software. The service platform is a functional platform for realizing service communication. The management platform is a functional platform for realizing the operation management of the Internet of Things system. In some embodiments, the management platform may include multiple management sub-platforms, and the management sub-platforms are respectively connected to the above-mentioned sensor network platform, and each management sub-platform respectively corresponds to include an image acquisition module, a first feature recognition module, a first data processing module, a second feature recognition module, a second data processing module, a third feature recognition module and a third data processing module. The sensor network platform is a functional platform for realizing sensor communication. The object platform is a functional platform for realizing perception control, and the object platform includes devices such as industrial cameras, conveyors, and depth cameras.

[0119] For the relevant explanations and examples of each module in the device of this embodiment, reference may be made to the method of the foregoing embodiment, which will not be elaborated here.

[0120] Embodiment 3

[0121] Based on the same inventive concept as the foregoing embodiment, this embodiment provides a computer device, which includes a memory and a processor, and a computer program is stored in the memory, and the processor executes the computer program to implement the above method.

[0122] Embodiment 4

[0123] Based on the same inventive concept as the foregoing embodiments, this embodiment provides a computer-readable storage medium, on which a computer program is stored, and a processor executes the computer program to implement the above method.

[0124] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the content of the specification and drawings of the present invention, or directly or indirectly applied to other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A workshop conveyor management method based on the industrial Internet of Things, characterized in that, The steps include: Obtain the top surface image of the target angle steel in a set conveying state; wherein, the set conveying state is the state where the target angle steel is placed in an inverted V shape on the conveyor; According to the top surface image, identify whether the target angle steel has a first deformation feature; wherein, the first deformation feature is the feature that the target angle steel has a bending deformation in the length direction; If the first deformation feature exists, mark the target angle steel as a first-class defective product and control the target angle steel to be conveyed to the first rework area; if the first deformation feature does not exist, obtain the end surface image of the target angle steel in the set conveying state; According to the end surface image, identify whether the target angle steel has a second deformation feature; wherein, the second deformation feature is the feature that the two side plates of the target angle steel are deformed; If the second deformation feature exists, mark the target angle steel as a second-class defective product and control the target angle steel to be conveyed to the second rework area; if the second deformation feature does not exist, obtain the depth image of the target angle steel in the set conveying state based on a preset perspective; wherein, the preset perspective is the perspective above the target angle steel and perpendicular to the side plate of the target angle steel; According to the depth image, identify whether the target angle steel has a third deformation feature; wherein, the third deformation feature is that the side plate of the target angle steel has a concave pit or a protrusion; If the third deformation feature exists, mark the target angle steel as a third-class defective product and control the target angle steel to be conveyed to the third rework area; if the third deformation feature does not exist, mark the target angle steel as a qualified product and control the target angle steel to be conveyed to the qualified area.

2. The workshop conveying management method based on industrial Internet of Things according to claim 1, wherein, According to the top surface image, identifying whether the target angle steel has a first deformation feature includes: According to the top surface image, extract the reference contour line of the vertex angle corresponding to the target angle steel and the edge contour line of any one side plate corresponding to the target angle steel; wherein, the reference contour line is the vertex angle contour line formed by the non-deformed part of the target angle steel; Select multiple measurement points in the edge contour line and obtain the first spacing value of each measurement point to the reference contour line; wherein, the measurement points at least include the two end points of the edge contour line, and the multiple measurement points are evenly spaced; Judge whether the multiple first spacing values are all equal to a preset first standard value. If so, identify that the target angle steel does not have a first deformation feature. If not, identify that the target angle steel has a first deformation feature.

3. The workshop conveying management method based on the industrial Internet of Things according to claim 2, characterized in that, Let the number of measurement points be n, then the expression of n is: n = K * L; In the formula, K is a proportionality coefficient, and L is the length of the target angle steel.

4. The workshop transportation management method based on industrial Internet of Things according to claim 1, wherein, According to the end surface image, identifying whether the target angle steel has a second deformation feature includes: According to the end surface image, identify the end surface contour of the target angle steel; Identify the second spacing value from the vertex angle point to the conveying horizontal line in the end surface contour; wherein, the vertex angle point is the common point of the included angle positions of the two side plates of the target angle steel, and the conveying horizontal line is the horizontal contour line formed by the plane where the conveyor contacts the target angle steel; Judge whether the second spacing value is equal to a preset second standard value. If so, identify that the target angle steel does not have a second deformation feature. If not, identify that the target angle steel has a second deformation feature.

5. The workshop conveying management method based on the industrial Internet of Things according to any one of claims 1-4, characterized in that According to the depth image, identifying whether the target angle steel has a third deformation feature includes: According to the depth image, obtain the measured depth values of multiple pixel points of the side plate corresponding to the target angle steel; Screen out the standard depth value among multiple measured depth values; wherein, the standard depth value is the pixel point depth value corresponding to the part of the side plate of the target angle steel without pits or protrusions. Screen out the target depth value among multiple measured depth values; wherein, the target depth value is the measured depth value that is greater than and / or less than the standard depth value. According to the depth image, obtain the pit area value and / or protrusion area value of the side plate of the corresponding target angle steel. Obtain the concave-convex deformation degree according to the target depth value, the pit area value and / or the protrusion area value. Judge whether the concave-convex deformation degree is less than a preset third standard value. If so, identify that the target angle steel does not have the third deformation feature. If not, identify that the target angle steel has the third deformation feature.

6. The workshop conveying management method based on industrial Internet of Things according to claim 5, characterized in that If the concave-convex deformation degree is E, the expression of E is: E = λ1 * |H - h| + λ2 * S; In the formula, H is the standard depth value, h is the target depth value, S is the pit area value and / or the protrusion area value, λ1 is the first adjustment coefficient, and λ2 is the second adjustment coefficient.

7. The workshop conveying management method based on industrial Internet of Things according to claim 5, characterized in that, Screening out the standard depth value among multiple measured depth values includes: Respectively divide the equal measured depth values into the same set to obtain multiple groups of depth data sets. Obtain the number of monomers of the measured depth values in each group of depth data sets. Output the measured depth value with the largest number of monomers as the standard depth value.

8. The workshop conveying management system based on the industrial Internet of Things is characterized in that, Including a management platform, a sensor network platform, and an object platform that are communicatively connected in sequence. The management platform includes: An image acquisition module for acquiring the top surface image of the target angle steel in a set conveying state; wherein, the set conveying state is the state where the target angle steel is placed in an inverted V shape on the conveyor. A first feature recognition module for recognizing whether the target angle steel has a first deformation feature according to the top surface image; wherein, the first deformation feature is the feature that the target angle steel has a bending deformation in the length direction. A first data processing module for, if there is a first deformation feature, marking the target angle steel as a first-class defective product and controlling the target angle steel to be conveyed to the first rework area; if there is no first deformation feature, acquiring the end surface image of the target angle steel in the set conveying state. A second feature recognition module for recognizing whether the target angle steel has a second deformation feature according to the end surface image; wherein, the second deformation feature is the feature that the two side plates of the target angle steel are deformed. A second data processing module for, if there is a second deformation feature, marking the target angle steel as a second-class defective product and controlling the target angle steel to be conveyed to the second rework area; if there is no second deformation feature, obtaining the depth image of the target angle steel in the set conveying state based on a preset perspective; wherein, the preset perspective is the perspective above the target angle steel and perpendicular to the side plate of the target angle steel. A third feature recognition module for recognizing whether the target angle steel has a third deformation feature according to the depth image; wherein, the third deformation feature is that the side plate of the target angle steel has pits or protrusions. A third data processing module for, if there is a third deformation feature, marking the target angle steel as a third-class defective product and controlling the target angle steel to be conveyed to the third rework area; if there is no third deformation feature, marking the target angle steel as a qualified product and controlling the target angle steel to be conveyed to the qualified area.

9. A computer device, characterized in that, The computer device includes a memory and a processor. A computer program is stored in the memory, and the processor executes the computer program to implement the workshop transportation management method based on the industrial Internet of Things as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and the processor executes the computer program to implement the workshop transportation management method based on the industrial Internet of Things as described in any one of claims 1-7.

Images