Steel bar fracture data acquisition and analysis method and system

Abstract

The invention discloses a steel bar fracture data acquisition and analysis method and system, and the system comprises a rack module which is used for fixing a feeding module, a conveying module, an acquisition module, a display module and a distribution module; the feeding module is used for placing reinforcing steel bars and conveying the reinforcing steel bars to the conveying module; the conveying module is used for conveying reinforcing steel bars to pass through the collecting module; the acquisition module is used for acquiring steel bar images; the analysis module is used for analyzing the steel bar image to obtain steel bar fracture data and transmitting an allocation command to the allocation module based on the steel bar fracture data; the display module is used for displaying the steel bar fracture data; the distribution module is used for distributing the broken reinforcing steel bars to the corresponding storage modules; and the storage module is used for storing broken steel bars and normal steel bars. The device has the beneficial effects that the steel bar fracture data of the steel bars can be detected, and the detected steel bars can be classified and placed based on the steel bar fracture data.

Description

Technical Field

[0001] The present invention relates to the technical field of steel bar detection, and in particular to a steel bar fracture data acquisition and analysis method and system. Background Art

[0002] Rebar testing refers to the process of testing and evaluating the quality of rebars. The importance of rebar testing lies in ensuring the safety of building structures. With the development of the construction industry, rebars, as a commonly used building material, are often subjected to strong mechanical effects during the construction process. If there are problems with the quality of the rebars, it may cause safety accidents such as the breakage or collapse of the building structure, causing huge losses to personnel and property. Through rebar testing, rebar quality problems can be discovered and resolved early, thereby improving the safety of building structures.

[0003] In the prior art, the fractures of steel bars are usually measured by manual observation to see if there are any, which requires a lot of manpower and material resources, and there may be errors in the manual measurement of the steel bar fracture data; or the steel bar fracture data is measured by professional equipment, but the equipment cost is high and is not suitable for testing multiple steel bars at the same time, which has certain limitations. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present application proposes a steel bar fracture data collection and analysis method and system, which can detect the steel bar fracture data of steel bars and classify and place the detected steel bars based on the steel bar fracture data.

[0005] The following is a technical solution of the present invention, a steel bar fracture data acquisition and analysis system, comprising:

[0006] Rack module, used to fix the loading module, conveying module, collection module, display module and distribution module;

[0007] The loading module is used to place the steel bars and transport them to the conveying module, which is connected to the rack module;

[0008] The conveying module is used to convey the steel bars through the collecting module and connect the rack module and the feeding module;

[0009] The acquisition module is used to acquire steel bar images and connect the rack module;

[0010] The analysis module is used to analyze the steel bar image to obtain the steel bar fracture data, transmit the distribution command to the distribution module based on the steel bar fracture data, and connect to the acquisition module;

[0011] Display module, used to display steel bar fracture data, connecting the rack module and the analysis module;

[0012] A distribution module, used to distribute broken steel bars to corresponding storage modules, connecting the rack module and the analysis module;

[0013] The storage module is used to place broken steel bars and normal steel bars and is connected to the conveying module.

[0014] Preferably, the loading module is provided with a first conveyor belt, a pressure sensing belt is provided on the first conveyor belt, a signal indicator light is provided on the rack module on one side of the first conveyor belt, and the pressure sensing belt signal is connected to the signal indicator light.

[0015] Preferably, the conveying module includes several second conveyor belts and a flipping structure, the first second conveyor belt and the first conveyor belt are arranged opposite to each other, a flipping mechanism is arranged between two adjacent second conveyor belts, and a distribution module is arranged on the rack modules on both sides of the last second conveyor belt.

[0016] Preferably, the acquisition module includes a base, a lifting column, a camera, a lamp housing and a fill light group. The base is fixedly connected to the rack module, the lifting column is fixedly connected to the base and the lamp housing, a camera is arranged in the lamp housing, and a fill light group is arranged at the lamp housing below the camera.

[0017] Preferably, the distribution module includes a first distribution device and a second distribution device, and the first distribution device and the second distribution device are connected to the analysis module by signal;

[0018] The storage module includes a first storage container, a second storage container and a third storage container. The first storage container and the first distribution device are located on the same side, the second storage container and the second distribution device are located on the same side, and the third storage container is located between the first storage container and the second storage container.

[0019] A method for collecting and analyzing steel bar fracture data comprises the following steps:

[0020] S1. The steel bars are placed in the loading module, the loading module transports the steel bars to the conveying module, the conveying module transports the steel bars and turns them over, and the acquisition module acquires images of the steel bars;

[0021] S2, the analysis module obtains the steel bar image, and obtains the steel bar fracture data based on the steel bar image;

[0022] S3, transmitting a distribution command to a distribution module based on the steel bar fracture data;

[0023] S4. The distribution module distributes the steel bars to the storage module based on the distribution command.

[0024] Preferably, S2 comprises the following steps:

[0025] S21, the analysis module obtains the steel bar image collected by the collection module;

[0026] S22, preprocessing the steel bar image;

[0027] Among them, the preprocessing of the steel bar image includes image enhancement processing and image noise reduction processing. The image enhancement processing includes adjusting the brightness, contrast and sharpening, and the image noise reduction processing is performed by filtering;

[0028] S23, performing region segmentation on the processed steel bar image;

[0029] Among them, the areas include base metal area, weld area, and joint area;

[0030] S24, performing data detection on the steel bar fracture to obtain steel bar fracture data, where the steel bar fracture data includes partition, fracture shape and size parameters;

[0031] S25. Save the steel bar fracture data.

[0032] Preferably, determining the partition of the steel bar fracture data comprises the following steps:

[0033] S2401, obtaining a steel bar fracture image, and partitioning the steel bar fracture image by manual annotation;

[0034] S2402, performing image preprocessing on the annotated steel bar image;

[0035] Among them, image preprocessing includes denoising, grayscale, resizing and normalization;

[0036] S2403, selecting a deep learning model and using a labeled training set to train the deep learning model, and evaluating the performance of the deep learning model;

[0037] Among them, the deep learning model adopts a fully convolutional neural network or a DeepLab model, and the evaluation indicators include one or more of the accuracy, recall rate and F1 score;

[0038] S2404. Input the steel bar image into the deep learning model for training and output the partitioning result.

[0039] Preferably, determining the fracture morphology of the steel bar fracture data comprises the following steps:

[0040] S2411, collecting steel bar fracture images, and marking the steel bar fracture images;

[0041] Among them, the steel bar fracture images include brittle fracture samples and ductile fracture samples;

[0042] S2412, performing image preprocessing on the annotated steel bar fracture image;

[0043] Among them, image preprocessing includes resizing, normalization and contrast enhancement;

[0044] S2413. Select a deep learning model and use the labeled training set to train the deep learning model, and evaluate the performance of the deep learning model;

[0045] Among them, the deep learning model uses convolutional neural network or VGG model;

[0046] S2414. Input the steel bar image into the deep learning model for training and output the fracture morphology.

[0047] Preferably, S3 comprises the following steps:

[0048] S31, if there is a data in the size parameter that exceeds the standard threshold, proceed to step S32, otherwise, do not generate a distribution command;

[0049] S32. Determine the fracture form. If it is a brittle fracture, generate and transmit a distribution command to the first distribution device. If it is a ductile fracture, generate and transmit a distribution command to the second distribution device.

[0050] The beneficial effects of the present invention are as follows: the present invention can convey steel bars and detect whether the steel bars have fractures, can detect steel bar fracture data such as the partition, fracture shape and size parameters of the steel bars, and can classify and place the detected steel bars based on the steel bar fracture data. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of module connection of the present invention;

[0052] Figure 2 It is a top view schematic diagram of the present invention;

[0053] Figure 3 It is a schematic diagram of the acquisition module of the present invention;

[0054] Figure 4 is a flow chart of the present invention;

[0055] Figure 5 A flow chart for obtaining steel bar fracture data in the present invention;

[0056] Figure 6 Obtain partition data flow chart for the present invention;

[0057] Figure 7 A flow chart for obtaining fracture morphology for the present invention;

[0058] In the figure: 1. rack module; 2. loading module; 201. first conveyor belt; 202. pressure sensing belt; 203. signal indicator light; 3. conveying module; 301. flip mechanism; 302. second conveyor belt; 4. acquisition module; 401. base; 402. lifting column; 403. lamp housing; 404. camera; 405. fill light group; 5. analysis module; 6. display module; 7. distribution module; 701. first distribution device; 702. second distribution device; 8. storage module; 801. first storage container; 802. second storage container; 803. third storage container. DETAILED DESCRIPTION

[0059] In order to make the technical problems solved by the present invention, the technical solutions adopted and the technical effects achieved clearer, the technical solutions of the embodiments of the present invention will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0060] Embodiment 1

[0061] like Figures 1 to 3 As shown, a steel bar fracture data acquisition and analysis system comprises:

[0062] The rack module 1 is used to fix the loading module 2, the conveying module 3, the collecting module 4, the display module 6 and the distribution module 7;

[0063] The loading module 2 is used to place the steel bars and transport them to the conveying module 3, which is connected to the rack module 1;

[0064] The conveying module 3 is used to convey the steel bars through the collecting module 4 and connect the rack module 1 and the feeding module 2;

[0065] The acquisition module 4 is used for acquiring steel bar images and is connected to the rack module 1;

[0066] The analysis module 5 is used to analyze the steel bar image to obtain the steel bar fracture data, transmit the distribution command to the distribution module 7 based on the steel bar fracture data, and connect to the acquisition module 4;

[0067] Display module 6, used for displaying steel bar fracture data, connecting the rack module 1 and the analysis module 5;

[0068] A distribution module 7, used for distributing broken steel bars to corresponding storage modules 8, connecting the rack module 1 and the analysis module 5;

[0069] The storage module 8 is used to place broken steel bars and normal steel bars and is connected to the conveying module 3 .

[0070] The rack module 1 is used to fix the feeding module 2, the conveying module 3, the acquisition module 4, the display module 6 and the distribution module 7. The conveying module 3 is laid on the rack module 1, and the steel bars provided by the feeding module 2 are conveyed by the conveying module 3, so that the steel bars pass through the acquisition module 4 for image acquisition, the display module 6 displays the steel bar fracture data, and the distribution module 7 distributes the broken steel bars to the corresponding storage module 8.

[0071] The loading module 2 is used to place the steel bars and convey the steel bars to the conveying module 3, and is connected to the rack module 1. Specifically, the loading module 2 is provided with a first conveying belt 201, and the steel bars are arranged on the first conveying belt 201; a pressure sensing belt 202 is provided on the first conveying belt 201, and a signal indicator light 203 is provided on the rack module 1 on one side of the first conveying belt 201. The signal of the pressure sensing belt 202 is connected to the signal indicator light 203. When the steel bars are placed on the first conveying belt 201, the pressure sensing belt 202 transmits a signal to the signal indicator light 203, and the signal indicator light 203 lights up after receiving the signal, otherwise, the signal indicator light 203 is in an off state, so that the staff can observe whether the steel bars are ready.

[0072] The conveying module 3 is used to convey the steel bars through the collection module 4, and is connected to the rack module 1 and the loading module 2. Specifically, the conveying module 3 includes a plurality of second conveyor belts 302 and a flipping structure, wherein the first second conveyor belt 302 and the first conveyor belt 201 are arranged opposite to each other, and when the first conveyor belt 201 is working, the steel bars on the first conveyor belt 201 are conveyed to the second conveyor belt 302, and a flipping mechanism 301 is arranged between the two second conveyor belts 302, and the flipping mechanism 301 is used to flip the steel bars, and the rack modules 1 on both sides of the last second conveyor belt 302 are provided with a distribution module 7.

[0073] The acquisition module 4 is used to acquire the steel bar image and is connected to the rack module 1. Specifically, the acquisition module 4 includes a base 401, a lifting column 402, a camera 404, a lamp housing 403 and a fill light group 405. The base 401 is fixedly connected to the rack module 1 and is used to fix the acquisition module 4 on the rack module 1. The lifting column 402 is fixedly connected to the base 401 and the lamp housing 403. The height of the camera 404 is adjusted by the lifting column 402. The lamp housing 403 is provided with a camera 404, which is a high-resolution camera 404. The image of the steel bar is acquired by the camera 404. The lamp housing 403 below the camera 404 is provided with a fill light group 405 to ensure that the shooting environment is bright and uniform. The fill light group 405 provides light to facilitate improving the clarity of the steel bar image.

[0074] The analysis module 5 is used to analyze the steel bar image to obtain the steel bar fracture data, transmit the distribution command to the distribution module 7 based on the steel bar fracture data, and connect the acquisition module 4. Specifically, the analysis module 5 obtains the steel bar image collected by the camera 404 in the acquisition module 4; pre-processes the steel bar image: performs image enhancement processing on the steel bar image, and the image enhancement processing includes adjusting the brightness, contrast and sharpening, thereby improving the clarity and recognizability of the image, removing the image noise, and removing the noise in the image by filtering to reduce the interference factors of the image; segmenting the processed steel bar image: segmenting different areas in the steel bar image based on the image segmentation technology, and the areas include the parent material area, the weld area, the joint area, etc., and extracting the main part of the steel bar fracture from the segmented image for subsequent analysis; performing data detection on the steel bar fracture: determining the partition of the steel bar, extracting the size parameters from the partitioned image, analyzing the fracture morphology of the steel bar, and judging whether it is a brittle fracture or a ductile fracture based on the fracture characteristics, and the fracture characteristics include necking, crack direction, etc.; saving the steel bar fracture data, and the steel bar fracture data includes partition, fracture morphology and size parameters.

[0075] Based on image segmentation technology, different regions in the steel bar image are segmented, and deep learning is usually used to identify the features of different regions. First, a large number of steel bar fracture images are collected, and the steel bar fracture images are partitioned through manual annotation, and the annotated steel bar images are preprocessed. Image preprocessing includes denoising, graying, resizing and normalization. In the field of image segmentation, deep learning models such as full convolutional neural networks and DeepLab models are usually used. The deep learning model is trained with annotated training sets, and the model performance is optimized by adjusting the batch size and the number of iterations. Overfitting is prevented by data enhancement and regularization. The performance of the deep learning model is evaluated using a test set. The evaluation indicators include accuracy, recall rate, F1 score, etc. If the deep learning model performance is not good, the deep learning model structure or training strategy is adjusted until the deep learning model meets the expected use. The steel bar image is input into the deep learning model for training, and the partitioning result is output.

[0076] Extract dimensional parameters such as crack length, width, depth, etc. from the partitioned image. Measure and mark the dimensions using the measurement tools in the image processing software. For example, use the straight line measurement tool to mark along the extension direction of the crack to calculate the length of the crack; select two points at the widest part of the crack to mark and calculate the distance between the two points as the width of the crack; the depth of the crack can be estimated through 3D reconstruction.

[0077] Collect steel bar fracture images, which include brittle fracture samples and ductile fracture samples, annotate the steel bar fracture images based on fracture features, and perform image preprocessing on the annotated steel bar fracture images, which includes resizing, normalization, contrast enhancement, etc.; usually use deep learning models such as convolutional neural networks and VGG models, and use annotated training sets to train the deep learning models; train the deep learning models, input steel bar images into the deep learning models for training, and output the fracture morphology, that is, whether it is brittle fracture or ductile fracture.

[0078] Based on the steel bar fracture data, the distribution command is transmitted to the distribution module 7. Specifically, the size parameter is used as the primary classification standard. If one of the length, width, and depth exceeds the standard threshold, the steel bar is determined to be a broken steel bar, and the distribution command is transmitted to the distribution module 7. Otherwise, no distribution is performed. The fracture morphology is used as the secondary classification standard. If it is a brittle fracture, the distribution command is transmitted to the first distribution device 701. If it is a ductile fracture, the distribution command is transmitted to the second distribution device 702.

[0079] The display module 6 is used to display the steel bar fracture data, and is connected to the rack module 1 and the analysis module 5. Specifically, the display module 6 is fixedly arranged on the side platform of the rack module 1, and the display module 6 and the analysis module 5 are connected by signal. The display module 6 obtains the steel bar fracture data from the analysis module 5, and displays the partition, fracture shape and size parameters in the steel bar fracture data, so as to facilitate the observation of the operator.

[0080] The distribution module 7 is used to distribute the broken steel bars to the corresponding storage module 8, and is connected to the rack module 1 and the analysis module 5; the distribution module 7 includes a first distribution device 701 and a second distribution device 702, which are connected to the analysis module 5 by signal, and are respectively arranged on the rack module 1 on both sides of the second conveyor belt 302. When the first distribution device 701 receives the distribution command from the analysis module 5, the first distribution device 701 distributes the broken steel bars; when the second distribution device 702 receives the distribution command from the analysis module 5, the second distribution device 702 distributes the broken steel bars.

[0081] The storage module 8 is used to place broken steel bars and normal steel bars, and is connected to the conveying module 3. Specifically, the storage module 8 includes a first storage container 801, a second storage container 802, and a third storage container 803, wherein the first storage container 801 and the first distribution device 701 are located on the same side, the second storage container 802 and the second distribution device 702 are located on the same side, the first distribution device 701 distributes brittle broken steel bars to the second storage container 802, the second distribution device 702 distributes ductile broken steel bars to the first storage container 801, and the third storage container 803 is used to place undistributed steel bars, and the third storage container 803 is located between the first storage container 801 and the second storage container 802.

[0082] The control method of this embodiment is controlled by a controller. The controller controls the functional operations of the loading module 2, the conveying module 3, the collection module 4 and the distribution module 7. The control circuit of the controller can be realized by simple programming by technicians in this field. The provision of power is also common knowledge in this field. In addition, this embodiment is used to protect the mechanical device, so this embodiment will not explain the control method and circuit connection in detail.

[0083] Embodiment 2

[0084] like Figures 4 to 7 As shown, a method for collecting and analyzing steel bar fracture data comprises the following steps:

[0085] S1, placing steel bars in the loading module 2, the loading module 2 conveys the steel bars to the conveying module 3, the conveying module 3 conveys the steel bars and turns them over, and the acquisition module 4 collects steel bar images;

[0086] S2, the analysis module 5 obtains the steel bar image, and obtains the steel bar fracture data based on the steel bar image;

[0087] S3, based on the steel bar fracture data, transmits the distribution command to the distribution module 7;

[0088] S4. The distribution module 7 distributes the steel bars to the storage module 8 based on the distribution command.

[0089] In step S1, the steel bar is placed on the loading module 2, and the loading module 2 conveys the steel bar to the conveying module 3, the conveying module 3 conveys the steel bar and turns the steel bar, and the acquisition module 4 collects the steel bar image. Specifically, the loading module 2 is provided with a first conveyor belt 201, and the steel bar is arranged on the first conveyor belt 201; the first conveyor belt 201 is provided with a pressure sensing belt 202, and the rack module 1 on one side of the first conveyor belt 201 is provided with a signal indicator 203, and the pressure sensing belt 202 signal is connected to the signal indicator 203. When the steel bar is placed on the first conveyor belt 201, the pressure sensing belt 202 transmits a signal to the signal indicator 203, and the signal indicator 203 lights up after receiving the signal, otherwise, the signal indicator 203 is in an off state, which is convenient for the staff to observe whether the steel bar is ready. The conveying module 3 includes several second conveyor belts 302 and a flipping structure, wherein the first second conveyor belt 302 and the first conveyor belt 201 are arranged opposite to each other. When the first conveyor belt 201 is working, the steel bars on the first conveyor belt 201 are transported to the second conveyor belt 302. A flipping mechanism 301 is arranged between the two second conveyor belts 302, and the flipping mechanism 301 is used to flip the steel bars. A distribution module 7 is arranged on the rack modules 1 on both sides of the last second conveyor belt 302.

[0090] In step S2, the analysis module 5 obtains the steel bar image and obtains the steel bar fracture data based on the steel bar image, including the following steps:

[0091] S21, the analysis module 5 obtains the steel bar image collected by the collection module 4;

[0092] S22, preprocessing the steel bar image;

[0093] The preprocessing of steel bar images includes image enhancement and image noise reduction;

[0094] Among them, image enhancement processing includes adjusting brightness, contrast and sharpening to improve the clarity and recognizability of the image; noise reduction processing is performed on the steel bar image through filtering to reduce the interference factors of the steel bar image;

[0095] S23, segmenting the processed steel bar image;

[0096] Based on the image segmentation technology, different areas in the steel bar image are segmented, including the base material area, weld area, joint area, etc. The main part of the steel bar fracture is extracted from the segmented image to facilitate subsequent analysis;

[0097] S24, performing data detection on the steel bar fracture to obtain steel bar fracture data, where the steel bar fracture data includes partition, fracture shape and size parameters;

[0098] Determine the partition of the steel bar, extract the size parameters from the steel bar image after partition, analyze the fracture morphology of the steel bar, and judge whether it is a brittle fracture or a ductile fracture based on the fracture characteristics, including necking, crack direction, etc.;

[0099] S25. Save the steel bar fracture data.

[0100] In step S24, the partition of the steel bar is determined. Different regions in the steel bar image are segmented based on image segmentation technology, and deep learning is usually used to identify the features of different regions, including the following steps:

[0101] S2401, obtaining a steel bar fracture image, and partitioning the steel bar fracture image by manual annotation;

[0102] The number of steel bar fracture images in each partition ranges from 500 to 1500;

[0103] S2402, performing image preprocessing on the annotated steel bar image;

[0104] Image preprocessing includes denoising, grayscale, resizing and normalization;

[0105] S2403, selecting a deep learning model and using a labeled training set to train the deep learning model, and evaluating the performance of the deep learning model;

[0106] In the field of image segmentation, deep learning models such as fully convolutional neural networks and DeepLab models are usually used. The deep learning models are trained using labeled training sets, and the model performance is optimized by adjusting the batch size and number of iterations. Overfitting is prevented through data enhancement and regularization.

[0107] Evaluate the performance of deep learning models. Use test sets to evaluate the performance of deep learning models. Evaluation indicators include accuracy, recall, F1 score, etc. If the performance of the deep learning model is poor, adjust the deep learning model structure or adjust the training strategy until the deep learning model meets the expected performance.

[0108] S2404. Input the steel bar image into the deep learning model for training and output the partitioning result.

[0109] In step S24, dimension parameters such as the length, width, and depth of the crack are extracted from the partitioned steel bar image. The dimensions are measured and marked using the measurement tools in the image processing software. For example, a straight line measurement tool is used to mark along the extension direction of the crack to calculate the length of the crack; two points are selected at the widest part of the crack to mark, and the distance between the two points is calculated as the width of the crack; the depth of the crack can be estimated through three-dimensional reconstruction.

[0110] In step S24, the fracture morphology of the steel bar is analyzed, including the following steps:

[0111] S2411, collecting steel bar fracture images, and marking the steel bar fracture images;

[0112] The steel bar fracture images include brittle fracture samples and ductile fracture samples, and the number of images of each sample ranges from 500 to 1500;

[0113] S2412, performing image preprocessing on the annotated steel bar fracture image;

[0114] Image preprocessing includes resizing, normalization, and contrast enhancement;

[0115] S2413. Select a deep learning model and use the labeled training set to train the deep learning model, and evaluate the performance of the deep learning model;

[0116] Usually, deep learning models such as convolutional neural networks and VGG models are used. The deep learning models are trained using labeled training sets. The performance of the deep learning models is evaluated using test sets. If the performance of the deep learning models is poor, the structure of the deep learning models or the training strategy is adjusted until the deep learning models meet the expectations.

[0117] S2414. Input the steel bar image into the deep learning model for training and output the fracture morphology.

[0118] In step S3, a distribution command is transmitted to the distribution module 7 based on the steel bar fracture data. Specifically, the size parameter is used as the primary classification standard. If one of the length, width, and depth exceeds the standard threshold, the steel bar is determined to be a broken steel bar, and the distribution command is transmitted to the distribution module 7. Otherwise, no distribution is performed. The fracture morphology is used as the secondary classification standard. If it is a brittle fracture, the distribution command is transmitted to the first distribution device 701. If it is a ductile fracture, the distribution command is transmitted to the second distribution device 702. The following steps are included:

[0119] S31, if there is a data in the size parameter that exceeds the standard threshold, proceed to step S32, otherwise, do not generate a distribution command;

[0120] S32 , determining the fracture form, if it is a brittle fracture, generating and transmitting a distribution command to the first distribution device 701 , if it is a ductile fracture, generating and transmitting a distribution command to the second distribution device 702 .

[0121] In step S4, the distribution module 7 distributes the steel bars to the storage module 8 based on the distribution command. The distribution module 7 includes a first distribution device 701 and a second distribution device 702, which are connected to the analysis module 5 by signal, and the first distribution device 701 and the second distribution device 702 are respectively arranged on the rack module 1 on both sides of the second conveyor belt 302. When the first distribution device 701 receives the distribution command of the analysis module 5, the first distribution device 701 distributes the broken steel bars; when the second distribution device 702 receives the distribution command of the analysis module 5, the second distribution device 702 distributes the broken steel bars. The storage module 8 includes a first storage container 801, a second storage container 802 and a third storage container 803, wherein the first storage container 801 and the first distribution device 701 are located on the same side, the second storage container 802 and the second distribution device 702 are located on the same side, the first distribution device 701 distributes the brittle fracture steel bars to the second storage container 802, the second distribution device 702 distributes the ductile fracture steel bars to the first storage container 801, and the third storage container 803 is used to place undistributed steel bars, and the third storage container 803 is located between the first storage container 801 and the second storage container 802.

[0122] The present invention can convey steel bars and detect whether the steel bars have fractures, can detect steel bar fracture data such as the partition, fracture shape and size parameters of the steel bars, and can classify and place the detected steel bars based on the steel bar fracture data.

[0123] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0124] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A steel bar fracture data acquisition and analysis system, characterized in that: include: Rack module, used to fix the loading module, conveying module, collection module, display module and distribution module; The loading module is used to place the steel bars and transport them to the conveying module, which is connected to the rack module; The conveying module is used to convey the steel bars through the collecting module and connect the rack module and the feeding module; The acquisition module is used to acquire steel bar images and connect the rack module; The analysis module is used to analyze the steel bar image to obtain the steel bar fracture data, transmit the distribution command to the distribution module based on the steel bar fracture data, and connect to the acquisition module; Display module, used to display steel bar fracture data, connecting the rack module and the analysis module; The distribution module is used to distribute the broken steel bars to the corresponding storage modules, connecting the rack module and the analysis module; the storage module is used to place the broken steel bars and normal steel bars, and connecting the conveying module.

2. A steel bar fracture data acquisition and analysis system according to claim 1, characterized in that: The feeding module is provided with a first conveyor belt, a pressure sensing belt is provided on the first conveyor belt, a signal indicator light is provided on the rack module on one side of the first conveyor belt, and the pressure sensing belt signal is connected to the signal indicator light.

3. A steel bar fracture data acquisition and analysis system according to claim 1, characterized in that: The conveying module includes several second conveyor belts and a flipping structure. The first second conveyor belt is arranged opposite to the first conveyor belt, a flipping mechanism is arranged between two adjacent second conveyor belts, and a distribution module is arranged on the rack modules on both sides of the last second conveyor belt.

4. A steel bar fracture data acquisition and analysis system according to claim 1, characterized in that: The acquisition module includes a base, a lifting column, a camera, a lamp housing and a fill light group. The base is fixedly connected to the rack module, the lifting column is fixedly connected to the base and the lamp housing, a camera is arranged in the lamp housing, and a fill light group is arranged at the lamp housing below the camera.

5. A steel bar fracture data acquisition and analysis system according to claim 1, characterized in that: The distribution module includes a first distribution device and a second distribution device, and the first distribution device and the second distribution device are connected to the analysis module by signal; The storage module includes a first storage container, a second storage container and a third storage container. The first storage container and the first distribution device are located on the same side, the second storage container and the second distribution device are located on the same side, and the third storage container is located between the first storage container and the second storage container.

6. A method for collecting and analyzing steel bar fracture data, applicable to a steel bar fracture data collection and analysis system according to any one of claims 1 to 5, characterized in that: The following steps are involved: S1. The steel bars are placed in the loading module, the loading module transports the steel bars to the conveying module, the conveying module transports the steel bars and turns them over, and the acquisition module acquires images of the steel bars; S2, the analysis module obtains the steel bar image, and obtains the steel bar fracture data based on the steel bar image; S3, transmitting a distribution command to a distribution module based on the steel bar fracture data; S4. The distribution module distributes the steel bars to the storage module based on the distribution command.

7. A steel bar fracture data collection and analysis method according to claim 6, characterized in that: S2 includes the following steps: S21, the analysis module obtains the steel bar image collected by the collection module; S22, preprocessing the steel bar image; Among them, the preprocessing of the steel bar image includes image enhancement processing and image noise reduction processing. The image enhancement processing includes adjusting the brightness, contrast and sharpening, and the image noise reduction processing is performed by filtering; S23, performing region segmentation on the processed steel bar image; Among them, the areas include base metal area, weld area, and joint area; S24, performing data detection on the steel bar fracture to obtain steel bar fracture data, where the steel bar fracture data includes partition, fracture shape and size parameters; S25. Save the steel bar fracture data.

8. A steel bar fracture data collection and analysis method according to claim 7, characterized in that: Determining the partition of steel bar fracture data includes the following steps: S2401, obtaining a steel bar fracture image, and partitioning the steel bar fracture image by manual annotation; S2402, performing image preprocessing on the annotated steel bar image; Among them, image preprocessing includes denoising, grayscale, resizing and normalization; S2403, selecting a deep learning model and using a labeled training set to train the deep learning model, and evaluating the performance of the deep learning model; Among them, the deep learning model adopts a fully convolutional neural network or a DeepLab model, and the evaluation indicators include one or more of the accuracy, recall rate and F1 score; S2404. Input the steel bar image into the deep learning model for training and output the partitioning result.

9. A steel bar fracture data collection and analysis method according to claim 7, characterized in that: Determine the fracture morphology of the steel bar fracture data, The following steps are involved: S2411, collecting steel bar fracture images, and marking the steel bar fracture images; Among them, the steel bar fracture images include brittle fracture samples and ductile fracture samples; S2412, performing image preprocessing on the annotated steel bar fracture image; Among them, image preprocessing includes resizing, normalization and contrast enhancement; S2413. Select a deep learning model and use the labeled training set to train the deep learning model, and evaluate the performance of the deep learning model; Among them, the deep learning model uses convolutional neural network or VGG model; S2414. Input the steel bar image into the deep learning model for training and output the fracture morphology.

10. A steel bar fracture data collection and analysis method according to claim 9, characterized in that: S3 includes the following steps: S31, if there is a data in the size parameter that exceeds the standard threshold, proceed to step S32, otherwise, do not generate a distribution command; S32. Determine the fracture form. If it is a brittle fracture, generate and transmit a distribution command to the first distribution device. If it is a ductile fracture, generate and transmit a distribution command to the second distribution device.

Images