Corrosion detection method, device and system for array-type metal connectors

Abstract

The present invention relates to a method, device and system for detecting corrosion of array-type metal connectors. The detection method comprises: posting a marker containing connector array information on a visible position on or near the connector array; acquiring an image containing the connector array and the corresponding marker by means of a camera; extracting information from the marker, and performing perspective transformation on the acquired image based on the acquired information to achieve image correction; based on the corrected image, identifying the corrosion pixel area of ​​the connector in the image by means of an image segmentation algorithm; and performing standardized recording by means of the ratio of the area of ​​the corrosion pixel area to the area of ​​the connector. Compared with the prior art, the present invention realizes the accurate acquisition of connector images and corresponding information by a mobile or fixed camera platform, and realizes the rapid feature extraction of corrosion pixel features, so that the corrosion degree detection of metal connectors is intelligent and automated.

Description

Technical Field

[0001] The present invention relates to the field of structural damage detection, and in particular to a method, device and system for detecting corrosion of array-type metal connectors. Background Art

[0002] Array connectors are often used to connect components. Common array connectors are mostly used in steel truss bridges, steel buildings, and mechanical structures. Due to the large number of connectors and the complexity of components, corrosion detection faces multiple challenges. First, manual detection is inefficient and the detection frequency is low; second, the connectors are not easily accessible; third, there is a problem with standardized records. The corrosion area and degree are difficult to record and describe in a standardized manner.

[0003] CN108416771A provides a method and terminal for detecting metal corrosion, including: obtaining a target image of a metal surface to be detected; extracting a corrosion image in the target image; converting the image coordinates corresponding to the corrosion image into space coordinates to obtain the position of the corrosion on the metal surface to be detected. In the above method, the terminal obtains a target image of a metal surface to be detected; extracts a corrosion image in the target image; converts the image coordinates corresponding to the corrosion image into space coordinates to obtain the position of the corrosion on the metal surface to be detected.

[0004] The defects of the above technical solution are: 1. It only measures the area of ​​corrosion on the metal surface to be detected. For complex metal surfaces, it is not as good as the flange plate with connectors. It is not possible to specifically divide the specific components where the corrosion is located. 2. It only quantitatively describes the corrosion position of a single metal surface, and cannot automatically identify the components to be detected. It lacks ease of use for complex scenes that are difficult to record manually. 3. No perspective transformation is performed, the collected image is distorted, and the measured area is inaccurate. Summary of the invention

[0005] The purpose of the present invention is to overcome the defects of the above-mentioned prior art and to provide a rust detection method for array-type metal connectors, a rust detection device for array-type metal connectors and an array-type metal connector system with rust monitoring function, which enables a mobile or fixed camera platform to accurately obtain connector images and corresponding information, and realizes rapid feature extraction of rust pixel features, so that the rust degree detection of metal connectors can be intelligent and automated, which is not only suitable for large-scale scenes, but also has high accuracy.

[0006] The purpose of the present invention can be achieved by the following technical solutions:

[0007] The first object of the present invention is to protect a method for detecting corrosion of an array of metal connectors, comprising the following steps:

[0008] S1: posting a marker containing information of the connector array on a visible position on or near the connector array;

[0009] S2: Acquire an image including the connector array and corresponding identifiers by using a camera;

[0010] S3: extracting information from the marker, and performing perspective transformation on the image obtained in S2 based on the obtained information to achieve image correction;

[0011] S4: Based on the rectified image, the rusted pixel area of ​​the connector in the image is identified by an image segmentation algorithm;

[0012] S5: Standardized recording is performed by the ratio of the rusted pixel area to the area of ​​the connector.

[0013] Furthermore, the photographing device includes a carrier and a camera arranged on the carrier.

[0014] Furthermore, the camera is a multi-eye camera, and each multi-eye camera obtains images of the connector array at different angles in a specific posture.

[0015] Furthermore, the marker comprises a carrier and a code visually marked on the carrier.

[0016] Furthermore, the coding includes information coding and structure coding, wherein specifically:

[0017] Information encoding, including the number and size of the connector array, the arrangement and spacing of the connector array, and the size of a single connector;

[0018] Structural encoding includes the calibration information required for perspective transformation and the specific parameters and thresholds required by the image segmentation algorithm.

[0019] The second object of the present invention is to protect a rust detection device for array-type metal connectors, comprising a movable platform and a plurality of multi-eye cameras arranged on the movable platform, each of which acquires images of different angles of the connector array in a specific posture;

[0020] The movable platform is also provided with a first communication component, which is a wireless or wired communication component.

[0021] Furthermore, the movable platform includes a platform body and an extendable component connected to the platform body, and the multi-eye camera is arranged on the platform body and / or the extendable component.

[0022] Furthermore, the extendable component includes a plurality of extendable units connected in sequence, adjacent extendable units can perform relative colinear / parallel displacement or relative rotational displacement, and a multi-eye camera is provided at the extendable end of the extendable component.

[0023] Furthermore, the movable platform includes a platform body and a walking assembly.

[0024] The third object of the present invention is to protect an array type metal connector system with a corrosion monitoring function, including connectors, identification carriers, monitoring equipment and background components, wherein specifically:

[0025] The connecting parts are arranged in an array structure on a specific structural body;

[0026] An identification carrier having a visual code thereon;

[0027] The monitoring device is the above-mentioned rust detection device, which takes images of the array-type metal connectors at a specific path and a specific distance;

[0028] The background component includes a second communication component connected to the monitoring device by wireless or wired communication, a ROM, a RAM and a microprocessor.

[0029] Compared with the prior art, the present invention has the following technical advantages:

[0030] 1) The mobile or fixed camera platform is used to accurately obtain the images and corresponding information of the connectors, and the fast feature extraction of the rust pixel features is realized, so that the rust degree detection of metal connectors can be intelligent and automated. Specifically, the technical solution uses coding, which has the advantage of providing component information and facilitating archiving and recording. On the other hand, the coding system can provide the point coordinates required for perspective correction, perform perspective correction on the image, and thus greatly reduce errors. In addition, for array connectors, this method can easily locate the defects of specific connectors and quantitatively describe the degree of damage of a single connector.

[0031] 2) In terms of accuracy, by posting markers, the pixel size and physical size of the object image can be correspondingly established according to the real size of the markers and the real points of the image actually detected, and the perspective transformation of the entire image can be performed to eliminate the perspective error of the flange plate surface caused by shooting. At the same time, the real rust area can be directly obtained, instead of measuring the number of pixels of the entire metal part and then calculating the rust percentage.

[0032] 3) In terms of detection strategy, by solving the information of the markers, the numbering information and connector information of the metal surface can be obtained. Combined with the results of perspective transformation, on the one hand, the rust distribution of other structural components on the flange surface, such as connectors and combined structures, can be automatically obtained; on the other hand, in large-scale scenarios, such as the node plates of steel truss bridges, there are thousands of node plates, most of which are repeated and difficult to automatically locate and record. The size and position information of the node plates can be automatically located through the present invention, and the rust on the bolts can be matched one by one, and they can be automatically archived after detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a schematic diagram of the process of the corrosion detection method of the array type metal connector in the technical solution;

[0034] Figure 2 This is a schematic diagram of the corrosion detection principle of the array-type metal connector in this technical solution;

[0035] Figure 3 This is a schematic structural diagram of the first corrosion detection device for array-type metal connectors in this technical solution;

[0036] Figure 4 This is a schematic structural diagram of the second type of rust detection device for array-type metal connectors in this technical solution.

[0037] In the figure: 0, structural body, 1, connecting part, 2, identification carrier, 3, photographing equipment, 4, movable platform, 41, platform body, 42, extendable component, 43, walking component, 5, multi-eye camera, 6, propeller. DETAILED DESCRIPTION

[0038] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0039] Example 1

[0040] The corrosion detection method for the array metal connector in this embodiment is shown in Figure 1 , including the following steps:

[0041] S1: Posting a marker containing information of the connector array on a visible position on or near the connector array, wherein the position should be designed in advance and included in the coded information of the marker;

[0042] S2: An image including the connector array and the corresponding identifier is obtained by a photographing device 3, wherein the photographing device 3 includes a carrier and a camera disposed on the carrier. The camera is a multi-eye camera, and each multi-eye camera obtains images of the connector array at different angles in a specific posture.

[0043] S3: Extract the information in the marker, perform perspective transformation on the image obtained in S2 based on the information obtained, generate a front view, and achieve image correction. The marker is a carrier and a code on the carrier that is visually marked. The code includes information coding and structural coding. Specifically, the information coding includes the number and size of the connector array, the arrangement form and spacing of the connector array, and the size of a single connector, wherein the size of a single connector includes the orthographic projection area of ​​each face of the connector during service, etc. The structural coding includes the calibration information required for perspective transformation, and the specific parameters and thresholds required by the image segmentation algorithm.

[0044] S4: Based on the rectified image, the connector is identified through the target detection algorithm. After identification, the connector anchor frame is cut out into a sub-image, and the image segmentation algorithm is used on the sub-image image to identify the rusted pixel area of ​​the connector in the image; in addition, in order to solve the rust problem of the flange plate itself, the sliding window method is used to sequentially use the sliding window sub-image as the input of the image segmentation algorithm to achieve high-precision results.

[0045] S5: Based on the real coordinates and size of the marker and the pixel size of the marker in the image, the correspondence between the image pixel point and the size of the real physical world can be directly established, and the real area of ​​the rusted pixel area can be directly recorded in a standardized manner.

[0046] The present invention constructs a rust detection device for array-type metal connectors, including a movable platform 4 and a plurality of multi-eye cameras 5 arranged on the movable platform 4, each of which acquires images of different angles of the connector array in a specific posture, so that not only the rust degree of the front of the connector can be acquired, but also the rust degree of the side of the connector ring can be acquired. A first communication component is also provided on the movable platform 4, and the first communication component is a wireless or wired communication component. In specific implementation, the first communication component transmits the captured image information to the second communication component in real time by wireless or wired signals.

[0047] The movable platform 4 includes a platform body 41 and an extendable component 42 connected to the platform body 41, wherein the platform body 41 can be a mobile vehicle, a drone body, or a hot air balloon body, see Figure 3 and Figure 4When the platform body 41 is provided with a propeller 6, a drone-type shooting platform can be realized, and a close-up shooting process can be achieved. The multi-eye camera 5 is arranged on the platform body 41 and / or the extendable component 42. In specific implementation, the extendable component 42 can be a mechanical arm. If necessary for cost considerations, it can be selected as a telescopic frame or a telescopic rod. The extendable component 42 includes a plurality of sequentially connected extension units, and adjacent extension units can perform relative colinear / parallel displacement or relative rotational displacement. A multi-eye camera 5 is arranged at the extension end of the extendable component 42. In specific implementation, the extension unit is a telescopic frame or a telescopic rod unit, and the mutual extension or rotation is achieved by the output rod of the cylinder or the electric cylinder.

[0048] The movable platform 4 includes a platform body 41 and a walking assembly 43. The walking assembly 43 in this embodiment includes a driving wheel, a universal wheel and other structures, which can be arranged on the top or top of the platform body 41, so as to realize image capture on the upper surface, lower surface and nearby area of ​​the structure.

[0049] The present invention also constructs an array type metal connector system with a corrosion monitoring function, including a connector 1, a marking carrier 2, a monitoring device and a background component, wherein specifically:

[0050] The connecting member 1 is arranged in an array structure on a specific structural body, and the identification carrier 2 is provided with a visual code, such as a combination of one or more of other forms such as a QR code, a one-dimensional code, a character string, etc., such as a combination of a QR code, a one-dimensional code and a checkerboard, wherein the image correction information, the node plate image segmentation information and other calibration information are structurally encoded in the form of a checkerboard in this embodiment, so as to realize single-target positioning or multi-target positioning during image processing, wherein the QR code, one-dimensional code, or character string is used as the information encoding.

[0051] The monitoring device is the above-mentioned rust detection device, which takes images of the array-type metal connectors at a specific path and a specific distance;

[0052] The backend component includes a second communication component, ROM, RAM and a microprocessor connected to the monitoring device by wireless or wired communication. The backend component is a key part for realizing preliminary image processing and subsequent processing, wherein the first communication component and the second communication component are both conventional signal transceiver components. The ROM stores an image segmentation algorithm, including a threshold-based segmentation method, a region-based segmentation method, and an edge-based segmentation method. The microprocessor can be selected from one of the ARM architecture or the x86 architecture. The microprocessor solves the artificial markers in the image after perspective transformation to obtain the arrangement information, quantity information, area information, etc. of the connector. At the same time, the ROM also stores a database commonly used in the prior art for image segmentation. The rusted pixel area of ​​the connector in the corrected image is identified based on the image segmentation algorithm, and finally the ratio of the rusted pixel area to the connector area is standardized and recorded. The calculated conclusion is stored in the ROM. The user sets the threshold of the comparison value, and the determination of the maintenance area to be replaced can be realized based on the calculated conclusion.

[0053] In this embodiment, the bolts in the connector 1 are used as an example for explanation, and other array-type connectors are also applicable to this technical solution.

[0054] Specific process of bolt detection:

[0055] 1. Design an artificial marker containing the connector array information, through which the arrangement information, quantity information, area information, etc. of the connector can be obtained, and perspective projection transformation can be realized.

[0056] 2. Stick the artificial marker on the plane where the connector is located.

[0057] 3. When inspection is required, use a camera to capture an image of the entire connecting component surface.

[0058] 4. Solve the artificial markers in the image after perspective transformation to obtain the arrangement information, quantity information, area information, etc. of the connectors.

[0059] 5. Identify the rusted pixel areas of the connector in the corrected image based on the image segmentation algorithm.

[0060] 6. Standardized recording is performed by the ratio of the rust pixel area to the area of ​​the connector.

[0061] The above description of the embodiments is to facilitate the understanding and use of the invention by those skilled in the art. It is obvious that those skilled in the art can easily make various modifications to these embodiments and apply the general principles described herein to other embodiments without creative work. Therefore, the present invention is not limited to the above embodiments, and improvements and modifications made by those skilled in the art based on the disclosure of the present invention without departing from the scope of the present invention should be within the scope of protection of the present invention.

Claims

1. A method for detecting corrosion of array-type metal connectors, characterized in that: The following steps are involved: S1: posting a marker containing information of the connector array on a visible position on or near the connector array; S2: acquiring an image including the connector array and corresponding identifiers by means of a photographing device (3); S3: extracting information from the marker, and performing perspective transformation on the image obtained in S2 based on the obtained information to achieve image correction; S4: Based on the rectified image, the rusted pixel area of ​​the connector in the image is identified by an image segmentation algorithm; S5: Standardized recording is performed by the ratio of the corroded pixel area to the area of ​​the connector; The photographing device (3) comprises a carrier and a camera arranged on the carrier, wherein the camera is a multi-eye camera, and each multi-eye camera respectively acquires images of the connecting member array at different angles; The marker is a code including a carrier and a visual mark on the carrier; The encoding includes: Information encoding, including the number and size of the connector array, the arrangement and spacing of the connector array, and the size of a single connector; Structural coding includes the calibration information required for perspective transformation, parameters and thresholds required by the image segmentation algorithm.

2. A rust detection device for array-type metal connectors for implementing the rust detection method described in claim 1, characterized in that: It comprises a movable platform (4) and a plurality of multi-eye cameras (5) arranged on the movable platform (4), each of the multi-eye cameras respectively acquiring images of the connecting member array at different angles; The movable platform (4) is also provided with a first communication component, which is a wireless or wired communication component.

3. A corrosion detection device for array-type metal connectors according to claim 2, characterized in that: The movable platform (4) comprises a platform body (41) and an extendable component (42) connected to the platform body (41), and the multi-eye camera (5) is arranged on the platform body (41) and / or the extendable component (42).

4. A corrosion detection device for array-type metal connectors according to claim 3, characterized in that: The extendable component (42) comprises a plurality of extendable units connected in sequence, and adjacent extendable units are capable of relative co-linear / parallel displacement or relative rotational displacement. A multi-eye camera (5) is provided at the extendable end of the extendable component (42).

5. A corrosion detection device for array-type metal connectors according to claim 4, characterized in that: The movable platform (4) comprises a platform body (41) and a walking assembly (43).

6. An array type metal connector system with corrosion monitoring function, characterized in that: include: The connecting members (1) are arranged in an array structure on the target structure body; An identification carrier (2) having a visual code thereon; A monitoring device, comprising: a rust detection device according to any one of claims 2 to 5, wherein the rust detection device takes images of the array-type metal connectors at a preset path and a preset distance; The background component includes a second communication component connected to the monitoring device by wireless or wired communication, a ROM, a RAM and a microprocessor.

Images