A method and device for identifying the opening degree of a fine stem valve

Abstract

The application discloses a kind of fine rod type valve opening recognition method, the method includes: making identification band, and it is pasted on multiple valve components, the structural dimension of each the valve component is measured;Valve number represented by identification band, size information, initial switch state of valve component represented by identification band of identification band, color feature of identification band;Collect the image I of target valve;Image processing is carried out to the image I of target valve and is recognized, and the image area where identification band is located is fitted, and the actual opening of target valve is calculated;The valve opening recognition method of the application can mark and data entry to the valve component to be identified in advance, compare by image acquisition processing of inspection robot and input data, to calculate the actual opening size of valve, reduce human intervention, improve calculation accuracy;Enhance the applicability of inspection robot in actual inspection process.

Description

Technical Field

[0001] This invention belongs to the field of safety monitoring and image recognition technology, and specifically relates to a method and device for recognizing the opening degree of a thin-rod valve. Background Technology

[0002] Currently, valve opening degree identification mainly relies on the experience and manual operation of inspection personnel. This manual operation can affect the operational status of the equipment controlled by the valve. For large industrial areas with a high number of valves, this manual identification method not only consumes a significant amount of time and energy for inspection personnel but also poses certain safety hazards to both personnel and equipment operation. In the era of rapid development of artificial intelligence, more and more industrial parks are using inspection robots to assist inspection personnel. These robots utilize various sensors and visible light cameras, combined with image recognition technology, to replace manual inspections.

[0003] However, current inspection robots primarily rely on image recognition for reading various instruments and infrared temperatures, with limited technology for identifying valves, which significantly impact safety in production. Furthermore, purchasing valve condition monitoring equipment would be prohibitively expensive and present difficulties in installation and maintenance. Summary of the Invention

[0004] To address the above problems, this invention discloses a method for identifying the opening degree of a thin-rod valve, the method comprising:

[0005] Make identification tapes and affix them to multiple valve components, and measure the structural dimensions of each valve component;

[0006] Record the valve number, size information, initial opening and closing status of the valve component represented by the label, and color characteristics of the label;

[0007] Image I of the target valve was acquired;

[0008] Image processing is performed on image I of the target valve to identify and fit the image region containing the marker band, and the actual opening degree of the target valve is calculated; whereby...

[0009] The plurality of valve components include the target valve.

[0010] Furthermore, the identification tape includes striped identification tape and specific color identification tape; the valve component is a thin rod valve, and the valve opening is controlled by a single thin rod, the rotation range of the thin rod is 0-90°; the process of making the identification tape and affixing it to multiple valve components, and measuring the structural dimensions of each valve component includes the following specific steps:

[0011] Production of striped and color-coded identification tapes;

[0012] Apply the striped marking tape circumferentially along the outer surface of the fixing part at the end of the valve stem of the valve component;

[0013] Apply the specific color-coded label around the outer surface of the valve stem of the valve component.

[0014] Measure the maximum diameter L of the fastener to which the striped marking tape is attached. t With perimeter L;

[0015] Measure the length L of the valve stem where the specific color-coded label is attached. b The radial cross-sectional perimeter of the valve stem;

[0016] Measure the distance L between the center point F of the specific color-coded label sticker location and the center point O of the end face of the striped label sticker location. FO .

[0017] Furthermore, the width of the specific color-coded label is preset to be the perimeter of the radial cross-section of the valve stem to which the color-coded label is pasted.

[0018] Furthermore, the striped marking strip has multiple parallel stripes pre-set on it, and the multiple stripes are arranged with two different colors of stripes.

[0019] When multiple valve components appear in Image I, the two different colored stripes of the striped identification tape pasted on the multiple valve components are arranged in different orders.

[0020] Furthermore, the color of the specific color identification strip is set to one of two different colored stripes of the strip identification strip affixed to the same valve component.

[0021] Furthermore, the steps of recording the valve number, size information, initial opening / closing state of the valve component represented by the identification band, and color characteristics of the identification band include:

[0022] The valve number, size information, initial opening and closing status of the valve component represented by the identification strip of multiple valve components, and color characteristics of the identification strip are entered into the inspection robot database.

[0023] in,

[0024] The size information includes the maximum diameter L of the fastener to which the striped identification tape is attached. t .

[0025] Furthermore, the acquisition of image I of the target valve includes the following specific steps:

[0026] Adjust the image acquisition angle of the acquisition unit so that the target valve, the specific color marking stripe on the target valve, and the stripe marking stripe on the target valve are within the image acquisition range of the acquisition unit;

[0027] Adjust the focal length and magnification of the acquisition unit;

[0028] The target centering algorithm is used to adjust the image area of ​​the target valve to the center of image I, and to make the straight line containing the valve stem center axis of the target valve parallel to the image plane captured by the acquisition unit;

[0029] Output the image I.

[0030] Furthermore, the image processing of image I of the target valve and the identification and fitting of the image region where the marker band is located includes the following specific steps:

[0031] Select the striped marker band in image I of the target valve and obtain the coordinates of the four corner points A, B, C, and D of the selected area;

[0032] Identify and read the target number information of the stripe identification band in the selected area of ​​the target valve in Image I, and match it with the stripe number information entered in the inspection robot database;

[0033] Based on the color characteristics of the striped markings entered into the inspection robot database, the shape of the color area of ​​the specific color marking in image I of the target valve is extracted to obtain image II of the target valve;

[0034] Image II of the target valve is preprocessed to obtain image III of the target valve;

[0035] The Hough transform algorithm is used to fit a straight line to the image of the target valve III, and the coordinates (X, Y) of the center point E of the fitted line are obtained. E ,Y E );

[0036] Furthermore, the process of identifying and reading the target number information of the stripe markers in the selected area of ​​image I of the target valve, and matching it with the stripe number information entered in the inspection robot database, includes the following specific steps:

[0037] The matching results are evaluated; among them,

[0038] If the match is successful, the coordinates of points A, B, C, and D in the target valve image I of the corresponding striped area are stored, along with the color characteristics of the specific color stripe.

[0039] If the match fails, the target number information of the stripe identification band in the next target valve image I is identified and matched in the order from left to right and from top to bottom.

[0040] Furthermore, the feature is that calculating the actual opening degree of the target valve includes the following specific steps:

[0041] Let O be the center point between points A and B, and let the coordinates of point O be (X, B, C). o ,Y o ), calculate point O(X) o ,Y o The coordinates of ).

[0042]

[0043] The center point of the pre-fitted line is set to point E. The image length L from point E to point O is calculated. EO The length L of the image between points A and B AB :

[0044]

[0045]

[0046] Among them, X A X B The x and y coordinates of points A and B are respectively. A Y B The ordinates of points A and B are X and Y, respectively. E X O Let Y be the x-coordinate of point E and point O. E Y O Let E and O be the ordinates;

[0047] Calculate the distance L' between the center point F of the specific color-coded label placement location and the bottom center point O of the striped label placement location in the target valve image. FO And the coordinates of the center point F of the specific color-coded marker in the target valve image where it is pasted:

[0048]

[0049] X F =X O ±L' FO Y F =Y O ;

[0050] Calculate the angle θ between line segments EO and FO:

[0051]

[0052] Where 0°≤θ≤90°;

[0053] Calculate the valve opening degree k:

[0054]

[0055] Specifically, if the initial switch state of the target valve recorded in the inspection robot database is "open", then the actual opening degree of the valve is 1-k; if the initial switch state of the target valve recorded in the inspection robot database is "closed", then the actual opening degree of the valve is k.

[0056] On another aspect, the present invention also discloses a thin-rod valve opening degree recognition device, which automatically identifies the actual opening degree of the thin-rod valve based on valve opening degree recognition preparation work. The device includes:

[0057] The acquisition unit is used to acquire image I of the target valve;

[0058] The image processing unit is used to perform image processing on image I of the target valve and to identify and fit the image region where the marker band is located; and

[0059] The calculation unit is used to calculate the actual opening degree of the target valve; wherein,

[0060] The valve opening degree identification preparation work specifically includes the following steps:

[0061] Make identification tapes and affix them to multiple valve components, and measure the structural dimensions of each valve component;

[0062] Record the valve number, size information, initial opening and closing status of the valve component represented by the label, and color characteristics of the label.

[0063] Furthermore, the specific steps for the acquisition unit to acquire image I of the target valve include:

[0064] Adjust the image acquisition angle of the acquisition unit so that the target valve, the specific color marking stripe on the target valve, and the stripe marking stripe on the target valve are within the image acquisition range of the acquisition unit;

[0065] Adjust the focal length and magnification of the acquisition unit;

[0066] The target centering algorithm is used to adjust the image area of ​​the target valve to the center of image I, and to make the straight line containing the valve stem center axis of the target valve parallel to the image plane captured by the acquisition unit;

[0067] Output the image I.

[0068] Furthermore, the specific steps by which the image processing unit performs image processing on image I of the target valve and identifies and fits the image region where the marker band is located include:

[0069] Select the striped marker band in image I of the target valve and obtain the coordinates of the four corner points A, B, C, and D of the selected area;

[0070] Identify and read the target number information of the stripe identification band in the selected area of ​​the target valve in Image I, and match it with the stripe number information entered in the inspection robot database;

[0071] Based on the color characteristics of the striped markings entered into the inspection robot database, the shape of the color area of ​​the specific color marking in image I of the target valve is extracted to obtain image II of the target valve;

[0072] Image II of the target valve is preprocessed to obtain image III of the target valve;

[0073] The Hough transform algorithm is used to fit a straight line to the image of the target valve III, and the coordinates (X, Y) of the center point E of the fitted line are obtained. E ,Y E );

[0074] Specifically, identifying and reading the target number information of the stripe identifier band in the selected area of ​​image I of the target valve, and matching it with the stripe number information entered in the inspection robot database, includes:

[0075] The matching results are evaluated; among them,

[0076] If the match is successful, the coordinates of points A, B, C, and D in the target valve image I of the corresponding striped area are stored, along with the color characteristics of the specific color stripe.

[0077] If the match fails, the target number information of the stripe identification band in the next target valve image I is identified and matched in the order from left to right and from top to bottom.

[0078] Furthermore, the calculation unit performs the calculation of the actual opening degree of the target valve, which includes the following specific steps:

[0079] Let O be the center point between points A and B, and let the coordinates of point O be (X, B, C). o ,Y o ), calculate point O(X) o ,Y o The coordinates of ).

[0080]

[0081] The center point of the pre-fitted line is set to point E. The image length L from point E to point O is calculated. EO The length L of the image between points A and B AB :

[0082]

[0083]

[0084] Among them, X A X B The x and y coordinates of points A and B are respectively. A Y B The ordinates of points A and B are X and Y, respectively. E X O Let Y be the x-coordinate of point E and point O. E Y O Let E and O be the ordinates;

[0085] Calculate the distance L' between the center point F of the specific color-coded label placement location and the bottom center point O of the striped label placement location in the target valve image. FO And the coordinates of the center point F of the specific color-coded marker in the target valve image where it is pasted:

[0086]

[0087] X F =X O ±L' FO Y F =Y O ;

[0088] Calculate the angle θ between line segments EO and FO:

[0089]

[0090] Where 0°≤θ≤90°;

[0091] Calculate the valve opening degree k:

[0092]

[0093] Specifically, if the initial switch state of the target valve recorded in the inspection robot database is "open", then the actual opening degree of the valve is 1-k; if the initial switch state of the target valve recorded in the inspection robot database is "closed", then the actual opening degree of the valve is k.

[0094] The valve opening recognition method of this invention can pre-mark and input data for the valve components to be identified. The actual valve opening is then calculated by comparing the image acquisition and processing data with that of the inspection robot, reducing human intervention and improving calculation accuracy. This invention solves the problem that inspection robots cannot identify the opening of thin-stem valves in practical applications, reducing the high cost associated with real-time valve opening detection using digital valves and enhancing the applicability of inspection robots in actual inspection processes.

[0095] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures pointed out in the description, claims and drawings. Attached Figure Description

[0096] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0097] Figure 1 A flowchart illustrating a method for identifying the opening degree of a thin-rod valve according to an embodiment of the present invention is shown.

[0098] Figure 2 A schematic diagram of the striped identification band unfolded according to an embodiment of the present invention is shown;

[0099] Figure 3 A front view of a striped marking strip affixed to a valve component according to an embodiment of the present invention is shown.

[0100] Figure 4 A front view of a specific color-coded stripe according to an embodiment of the present invention is shown. Detailed Implementation

[0101] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0102] To address the issue that existing inspection robots primarily focus on recognizing various instrument readings and infrared temperatures, with limited technology for identifying valves—a critical component in safe production—this invention addresses the problem of high costs and difficult installation and maintenance associated with purchasing valve status monitoring equipment. This invention proposes a thin-rod valve opening recognition device to execute the aforementioned thin-rod valve opening recognition method. (Refer to...) Figure 1 As shown, in one embodiment of the present invention, the device includes: an acquisition unit, an image processing unit, and a computing unit.

[0103] The system includes an acquisition unit for acquiring image I of the target valve; an image processing unit for processing image I of the target valve and identifying and fitting the image region where the marker band is located; and a calculation unit for calculating the actual opening degree of the target valve.

[0104] The plurality of valve components include the target valve.

[0105] In this embodiment, the acquisition unit, image processing unit, and computing unit are integrated inside the inspection and robot. Before executing the thin-rod valve opening recognition method, data entry and labeling of the valve component to be recognized are required, i.e., valve opening recognition preparation work is performed. The valve opening recognition preparation work includes the following steps:

[0106] Make identification tapes and affix them to multiple valve components, and measure the structural dimensions of each valve component;

[0107] Record the valve number, size information, initial opening and closing status of the valve component represented by the label, and color characteristics of the label;

[0108] The identification strip includes a striped identification strip and a specific color identification strip; the valve component is a thin rod valve, and the valve opening is controlled by a single thin rod, the rotation range of which is 0-90°.

[0109] Furthermore, such as Figure 2-4 As shown, the process of making the identification tape, affixing it to multiple valve components, and measuring the structural dimensions of each valve component includes the following specific steps:

[0110] Production of striped and color-coded identification tapes;

[0111] Apply the striped marking tape circumferentially along the outer surface of the fixing part at the end of the valve stem of the valve component;

[0112] Apply the specific color-coded label around the outer surface of the valve stem of the valve component.

[0113] Measure the maximum diameter L of the fastener to which the striped marking tape is attached.t With perimeter L;

[0114] Measure the length L of the valve stem where the specific color-coded label is attached. b The radial cross-sectional perimeter of the valve stem;

[0115] Measure the distance L between the center point F of the specific color-coded label sticker location and the center point O of the end face of the striped label sticker location. FO .

[0116] It should be noted that the fastener can be a screw or other fastener, and the fastener has a nut, on which the striped marking tape is circumferentially attached; the striped marking tape has multiple parallel stripes pre-set, and two different colors of stripes are arranged within the multiple stripes; the specific number of stripes can be limited according to the actual usage. For example, if the number of stripes is set to t, then the corresponding stripe arrangement is 2. t There are two types, each arrangement corresponding to one thin-stem valve, which can meet the requirement that the number of thin-stem valves on site does not exceed 2. t In the case of multiple valve components in Image I, the two different colored stripes of the striped identification tape affixed to the multiple valve components are arranged in different orders.

[0117] like Figure 2 As shown, in this embodiment, the multiple stripes can be five stripes, with two different colored stripes (black and gray) arranged within each five stripes. The function of the stripe marking stripe is to distinguish multiple thin-stem valves of the same type within the same target valve image I by varying the width of the black and gray stripe areas, thereby allowing for the identification of the opening degree of multiple thin-stem valves of the same type. The function of the specific color marking stripe is to indicate the position of the valve stem on the thin-stem valve.

[0118] The striped identification tape and the specific color identification tape can be made of any material. Preferably, the striped identification tape and the specific color identification tape are made of insulating polyvinyl chloride film. The width of the specific color identification tape is designed to be the perimeter of the radial cross-section of the valve stem to which it is attached; the color of the specific color identification tape is selected from either black or gray stripes in the striped identification tape.

[0119] The arrangement order of the two colors of the stripe on the stripe marking strip refers to the irregular arrangement of two different colors of stripes of the same width; wherein, the number represented by the identification of one color stripe is set to "1", for example, the number "1" is recorded after the black stripe is identified, and the number represented by the identification of the other color stripe is set to "0", for example, the number "0" is recorded after the gray stripe is identified, and the total actual width of the two colors of stripe is equal to the actual width d of the stripe marking strip.

[0120] In this embodiment, the corresponding numerical labels are arranged and combined according to the arrangement order of the two different colored stripes of the stripe identification strip to obtain the stripe number information of the stripe identification strip of multiple valve components; the stripe number information consists of five consecutive numerical labels, and the numerical labels are 0 or 1, and the stripe number information is recorded and stored as the valve number of the valve component.

[0121] In one embodiment of the present invention, the recording of the valve number, size information, initial opening / closing state of the valve component represented by the identification band, and color characteristics of the identification band includes the following specific steps:

[0122] The valve number, size information, initial open / closed state of the valve component represented by the identification strips, and color characteristics of the identification strips are entered into the inspection robot database; wherein, the size information includes the maximum diameter L of the fastener to which the striped identification strip is attached. t .

[0123] Specifically, the valve number information is used to filter out a specific target valve image from multiple identical valve images identified from the target valve image, determining the valve opening degree to be identified; the color feature information of the specific color-coded stripe is used to filter out the valve stem image area of ​​the target valve whose opening degree needs to be identified; and the maximum diameter L of the fixing piece at the end of the valve stem to which the striped marking stripe is attached. t The length L of the striped marking stripe in the acquired target valve image I is used to calculate the length L of the image region. AB The actual length L of the striped marking strip t Proportion.

[0124] In one embodiment of the present invention, the specific steps for the acquisition unit to acquire image I of the target valve include the following steps:

[0125] Adjust the image acquisition angle of the acquisition unit so that the target valve, the specific color marking stripe on the target valve, and the stripe marking stripe on the target valve are within the image acquisition range of the acquisition unit; adjust the focal length and magnification of the acquisition unit; use a target centering algorithm to adjust the image area of ​​the target valve to the center of image I, and make the straight line containing the central axis of the valve stem of the target valve parallel to the image plane captured by the acquisition unit; output image I.

[0126] The acquisition unit can be a visible light camera integrated into the pan-tilt head of the inspection robot. In practical use, the pan-tilt head angle of the inspection robot is adjusted so that the target valve, its specific color markings, and striped markings are within the camera's field of view. This ensures that the straight line containing the valve stem's central axis is parallel to the image plane captured by the visible light camera, facilitating the acquisition of a complete and clear image and reducing the impact of accumulated errors caused by the pan-tilt head rotation on the integrity and clarity of the image acquisition.

[0127] In one embodiment of the present invention, the specific steps of the image processing unit performing image processing on the image I of the target valve and identifying and fitting the image region where the marker band is located include:

[0128] Select the striped marker band in image I of the target valve and obtain the coordinates of the four corner points A, B, C, and D of the selected area;

[0129] Identify and read the target number information of the stripe identification band in the selected area of ​​the target valve in Image I, and match it with the stripe number information entered in the inspection robot database;

[0130] Based on the color characteristics of the striped markings entered into the inspection robot database, the shape of the color area of ​​the specific color marking in image I of the target valve is extracted to obtain image II of the target valve;

[0131] Image II of the target valve is preprocessed to obtain image III of the target valve;

[0132] The Hough transform algorithm is used to fit a straight line to the image of the target valve III, and the coordinates (X, Y) of the center point E of the fitted line are obtained. E ,Y E );

[0133] Specifically, identifying and reading the target number information of the stripe identifier band in the selected area of ​​image I of the target valve, and matching it with the stripe number information entered in the inspection robot database, includes:

[0134] The matching results are evaluated; among them,

[0135] If the match is successful, the coordinates of points A, B, C, and D in the target valve image I of the corresponding striped area are stored, along with the color characteristics of the specific color stripe.

[0136] If the match fails, the target number information of the stripe identification band in the next target valve image I is identified and matched in the order from left to right and from top to bottom.

[0137] Specifically, the target valve's target number information is identified based on the arrangement order of the two different colored stripes on the stripe marking stripe of the target valve. The read target number information is then compared with the valve number information stored in the inspection robot to determine the target valve and retrieve other stored information about the target valve.

[0138] Furthermore, in the process of preprocessing the target valve image II to obtain the target valve image III, the image processing software in the inspection robot's host computer needs to be used to preprocess the target valve image II. The specific preprocessing steps include: performing grayscale conversion, noise reduction, binarization, and image thinning on the target valve image II in sequence, and removing image features within the area selected by the striped marker band. In addition, the image in the target valve image III is the image after processing the area of ​​the specific color marker band.

[0139] Furthermore, the shape of the specific color-coded tape in Image I is extracted using color extraction to obtain Image II. Therefore, Image II contains two rectangular regions of different sizes, which are extracted and are in color. Image II is preprocessed to obtain Image III, which is a binarized image, facilitating line fitting and improving computational efficiency.

[0140] The calculation unit performs the calculation of the actual opening degree of the target valve, including the following specific steps:

[0141] Let O be the center point between points A and B, and let the coordinates of point O be (X, B, C). o ,Y o ), calculate point O(X) o ,Y o The coordinates of ).

[0142]

[0143] The center point of the pre-fitted line is set to point E. The image length L from point E to point O is calculated. EO The length L of the image between points A and B AB :

[0144]

[0145]

[0146] Among them, X A X B The x and y coordinates of points A and B are respectively. A Y B These are the ordinates of points A and B, respectively;

[0147] X E X OLet Y be the x-coordinate of point E and point O. E Y O Let E and O be the ordinates;

[0148] Calculate the distance L' between the center point F of the specific color-coded label placement location and the bottom center point O of the striped label placement location in the target valve image. FO And the coordinates of the center point F of the specific color-coded marker in the target valve image where it is pasted:

[0149]

[0150] X F =X O ±L' FO Y F =Y O ;

[0151] Calculate the angle θ between line segments EO and FO:

[0152]

[0153] Where 0°≤θ≤90°;

[0154] Calculate the valve opening degree k:

[0155]

[0156] Specifically, if the initial switch state of the target valve recorded in the inspection robot database is "open", then the actual opening degree of the valve is 1-k; if the initial switch state of the target valve recorded in the inspection robot database is "closed", then the actual opening degree of the valve is k.

[0157] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying the opening degree of a thin-rod valve, characterized in that, The method includes: Image I of the target valve was acquired; Image processing is performed on image I of the target valve to identify and fit the image region where the marker band is located, and the actual opening degree of the target valve is calculated. The image processing of image I of the target valve and the identification and fitting of the image region where the marker band is located include the following specific steps: Select the striped marker band in image I of the target valve and obtain the coordinates of the four corner points A, B, C, and D of the selected area; The process involves identifying and reading the target number information of the stripe markers in the selected area of ​​the target valve image I, and matching it with the stripe number information entered in the inspection robot database. This identification and matching includes the following specific steps: judging the matching result; if the match is successful, storing the coordinates of points A, B, C, and D in the target valve image I and the color characteristics of the specific color markers; if the match is unsuccessful, sequentially identifying the target number information of the stripe markers in the next target valve image I from left to right and top to bottom, and matching the stripe number information of the next stripe marker. Based on the color characteristics of the striped markings entered into the inspection robot database, the shape of the color area of ​​the specific color marking in image I of the target valve is extracted to obtain image II of the target valve; Image II of the target valve is preprocessed to obtain image III of the target valve; The Hough transform algorithm is used to fit a straight line to the image of the target valve III, and the coordinates of the center point E of the fitted line are obtained. ; The calculation of the actual opening degree of the target valve includes the following specific steps: Let O be the center point between points A and B, and let the coordinates of point O be ( ). ), calculate point O ( The coordinates of ). ， The center point of the pre-fitted line is set to point E. The image length from point E to point O is calculated. The length of the image between points A and B : in, These are the x-coordinates of points A and B, respectively. Let A and B be the ordinates of points B, respectively. Let E and O be the x-coordinates. Let E and O be the ordinates; Calculate the distance between the center point F of the specific color-coded label placement location and the bottom center point O of the striped label placement location in the target valve image. And the coordinates of the center point F of the specific color marker in the target valve image where it is pasted. The maximum diameter of the fastener to which the striped signage is attached: ； ， Calculate the angle between line segment EO and line segment FO. : ,in ; Calculate the valve opening degree k: Specifically, if the initial switch state of the target valve recorded in the inspection robot database is "open", then the actual opening degree of the valve is 1-k; if the initial switch state of the target valve recorded in the inspection robot database is "closed", then the actual opening degree of the valve is k.

2. The method for identifying the opening degree of a thin-rod valve according to claim 1, characterized in that, The method further includes the following steps before acquiring image I of the target valve: Make identification tapes and affix them to multiple valve components, and measure the structural dimensions of each valve component; Record the valve number, size information, initial open / closed state of the valve component represented by the label, and color characteristics of the label; among which, The plurality of valve components include the target valve.

3. The method for identifying the opening degree of a thin-rod valve according to claim 2, characterized in that, The marking strip includes a striped marking strip and a specific color marking strip; the valve component is a thin rod valve, and the valve opening is controlled by a single thin rod, the rotation range of which is 0-90°; The process of creating identification tapes and affixing them to multiple valve components, and measuring the structural dimensions of each valve component, includes the following specific steps: Production of striped and color-coded identification tapes; Apply the striped marking tape circumferentially along the outer surface of the fixing part at the end of the valve stem of the valve component; Apply the specific color-coded label around the outer surface of the valve stem of the valve component. Measure the maximum diameter of the fastener to which the striped marking tape is attached. With perimeter L; Measure the length of the valve stem to which the specific color-coded label is attached. The radial cross-sectional perimeter of the valve stem; Measure the distance between the center point F of the specific color-coded labeling tape's affixation location and the center point O of the end face of the striped labeling tape's affixation location. .

4. The method for identifying the opening degree of a thin-rod valve according to claim 3, characterized in that, The width of the specific color-coded strip is the perimeter of the radial cross-section of the valve stem to which the color-coded strip is attached.

5. The method for identifying the opening degree of a thin-rod valve according to claim 3, characterized in that, The striped marking strip has multiple parallel stripes pre-set, and the stripes are arranged with two different colors of stripes. When multiple valve components appear in Image I, the two different colored stripes of the striped identification tape pasted on the multiple valve components are arranged in different orders.

6. The method for identifying the opening degree of a thin-rod valve according to claim 5, characterized in that, The color of the specific color identification strip is set to one of two different colored stripes of the strip identification strip affixed to the same valve component.

7. The method for identifying the opening degree of a thin-rod valve according to claim 2, characterized in that, The steps of recording the valve number, size information, initial opening / closing state of the valve component represented by the identification band, and color characteristics of the identification band include: The valve number, size information, initial open / closed state of the valve component represented by the identification bands, and color characteristics of the identification bands are entered into the inspection robot database; among them, The size information includes the maximum diameter of the fastener to which the striped identification tape is attached. .

8. The method for identifying the opening degree of a thin-rod valve according to any one of claims 1-7, characterized in that, The specific steps for acquiring the target valve image I are as follows: Adjust the image acquisition angle of the acquisition unit so that the target valve, the specific color marking stripe on the target valve, and the stripe marking stripe on the target valve are within the image acquisition range of the acquisition unit; Adjust the focal length and magnification of the acquisition unit; The target centering algorithm is used to adjust the image area of ​​the target valve to the center of image I, and to make the straight line containing the valve stem center axis of the target valve parallel to the image plane captured by the acquisition unit; Output the image I.

9. A device for recognizing the opening degree of a thin-rod valve, which automatically recognizes the actual opening degree of the thin-rod valve based on valve opening degree recognition preparation work, characterized in that... The device includes: The acquisition unit is used to acquire image I of the target valve; The image processing unit is used to perform image processing on image I of the target valve and to identify and fit the image region where the marker band is located; and The calculation unit is used to calculate the actual opening degree of the target valve; wherein, The valve opening degree identification preparation work specifically includes the following steps: Make identification tapes and affix them to multiple valve components, and measure the structural dimensions of each valve component; Record the valve number, size information, initial opening and closing status of the valve component represented by the label, and color characteristics of the label; The specific steps by which the image processing unit performs image processing on image I of the target valve and identifies and fits the image region where the marker band is located include: Select the striped marker band in image I of the target valve and obtain the coordinates of the four corner points A, B, C, and D of the selected area; The system identifies and reads the target number information of the stripe markers in the selected area of ​​the target valve in image I, and matches it with the stripe number information entered in the inspection robot database. Specifically, this involves judging the matching result. If the match is successful, the coordinates of points A, B, C, and D in the target valve image I, along with the color characteristics of the specific color markers, are stored. If the match fails, the system sequentially identifies and matches the target number information of the stripe markers in the next target valve image I in a left-to-right, top-to-bottom order. Based on the color characteristics of the striped markings entered into the inspection robot database, the shape of the color area of ​​the specific color marking in image I of the target valve is extracted to obtain image II of the target valve; Image II of the target valve is preprocessed to obtain image III of the target valve; The Hough transform algorithm is used to fit a straight line to the image of the target valve III, and the coordinates of the center point E of the fitted line are obtained. ; The calculation unit performs the calculation of the actual opening degree of the target valve, including the following specific steps: Let O be the center point between points A and B, and let the coordinates of point O be ( ). ), calculate point O ( The coordinates of ). ， The center point of the pre-fitted line is set to point E. The image length from point E to point O is calculated. The length of the image between points A and B : in, These are the x-coordinates of points A and B, respectively. These are the ordinates of points A and B, respectively. Let E and O be the x-coordinates. Let E and O be the ordinates; Calculate the distance between the center point F of the specific color-coded label placement location and the bottom center point O of the striped label placement location in the target valve image. And the coordinates of the center point F of the specific color marker in the target valve image where it is pasted. The maximum diameter of the fastener to which the striped signage is attached: ； ， ； Calculate the angle between line segment EO and line segment FO. : ； ,in ; Calculate the valve opening degree k: Specifically, if the initial switch state of the target valve recorded in the inspection robot database is "open", then the actual opening degree of the valve is 1-k; if the initial switch state of the target valve recorded in the inspection robot database is "closed", then the actual opening degree of the valve is k.

10. The valve opening degree recognition device according to claim 9, characterized in that, The specific steps by which the acquisition unit acquires image I of the target valve include: Adjust the image acquisition angle of the acquisition unit so that the target valve, the specific color marking stripe on the target valve, and the stripe marking stripe on the target valve are within the image acquisition range of the acquisition unit; Adjust the focal length and magnification of the acquisition unit; The target centering algorithm is used to adjust the image area of ​​the target valve to the center of image I, and to make the straight line containing the valve stem center axis of the target valve parallel to the image plane captured by the acquisition unit; Output the image I.

Images