Method for identifying mechanical cabinet capacity image of gas cabinet

Abstract

The invention discloses a gas cabinet mechanical cabinet capacity image identification method, and relates to the technical field of gas, and the method comprises the steps: obtaining all cabinet capacity data in a preset precision range, constructing a cabinet capacity database, and each group of cabinet capacity data comprises a current cabinet capacity value and cabinet capacity indication image data; acquiring real-time cabinet capacity indication image data; and inputting the real-time cabinet capacity indication image data into the cabinet capacity database, judging whether a cabinet capacity value corresponding to the current cabinet capacity indication image data exists in the cabinet capacity database or not, if so, outputting the corresponding cabinet capacity value, and taking the corresponding cabinet capacity value as the current cabinet capacity value. According to the invention, the current cabinet capacity value can be identified by constructing the cabinet capacity database, carrying out image identification on the acquired cabinet capacity indication image data and then inputting the data into the cabinet capacity database for comparison, so that errors are effectively reduced, and production accidents caused by overlarge cabinet capacity errors of the gas cabinet radar are avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of gas, and particularly to a method for image recognition of the mechanical cabinet volume of a gas holder. Background Art

[0002] Coal gas is an important by-product produced in the iron-making and steel-making processes of iron and steel enterprises. Coal gas is also a combustible energy source with a high calorific value. After the coal gas is generated, it is transported to the gas holder through pipelines, and after being pressurized and blended to meet the standards, it is sent to users for production. The gas holder is an important device for storing coal gas and balancing the pressure of the coal gas pipeline network. The safe and stable operation of the gas holder can provide strong guarantee for production.

[0003] Currently, two sets of cabinet volume measuring devices are equipped for the gas holder, namely the radar cabinet volume and the mechanical cabinet volume. The radar cabinet volume is calculated by measuring the height of the piston of the gas holder by the radar of the gas holder cabinet volume, and is interlocked with the automatic control plc and displayed on the operation screen. However, the radar is greatly affected by factors such as noise, and the error is often large. The mechanical cabinet volume is that the piston of the gas holder is rigidly connected to the cabinet volume indicator panel by a steel wire rope, and through the cooperation of guide wheels, counterweights, etc., it can sensitively reflect the cabinet volume and is displayed on the indicator panel. It has high precision and good stability, but it cannot be interlocked with the plc and cannot be displayed on the operation screen, and can only be observed through monitoring. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the shortcomings of the prior art and provide a method for image recognition of the mechanical cabinet volume of a gas holder.

[0005] To solve the above technical problems, the technical solution of the present invention is as follows: A method for image recognition of the mechanical cabinet volume of a gas holder includes: Obtaining all cabinet volume data within a preset accuracy range, constructing a cabinet volume database, and each group of the cabinet volume data includes the current cabinet volume value and the cabinet volume indication image data; Collecting real-time cabinet volume indication image data; Inputting the real-time cabinet volume indication image data into the cabinet volume database, and judging whether there is a cabinet volume value corresponding to the current cabinet volume indication image data in the cabinet volume database. If it exists, output the corresponding cabinet volume value and use it as the current cabinet volume value.

[0006] As a preferred scheme of the method for image recognition of the mechanical cabinet volume of the gas holder according to the present invention, wherein: the obtaining all cabinet volume data within a preset accuracy range and constructing a cabinet volume database includes: Obtaining all cabinet volume values within a preset accuracy range; For any cabinet volume value, collecting the corresponding cabinet volume indication image data, and taking the current cabinet volume value and the current cabinet volume indication image data as a group of cabinet volume data; Obtaining all cabinet volume data within a preset accuracy range and constructing a cabinet volume database.

[0007] As a preferred embodiment of the method for identifying the mechanical cabinet volume image of the gas holder according to the present invention, wherein: collecting the image data corresponding to the cabinet volume indicator panel, and taking the current cabinet volume value and the current image data of the cabinet volume indicator panel as a set of cabinet volume data includes: Taking a fixed point at the top of the gas holder as the first endpoint, taking the midpoint of the top of the piston of the gas holder as the second endpoint, and taking a fixed point at the edge of the piston of the gas holder as the third endpoint to enclose a first target triangular region; Obtaining the angles of the three apex angles within the first target triangular region, and associating the angles of the three apex angles with the current cabinet volume to form a set of cabinet volume data.

[0008] As a preferred embodiment of the method for identifying the mechanical cabinet volume image of the gas holder according to the present invention, wherein: collecting the image data corresponding to the current cabinet volume indicator, and taking the current cabinet volume value and the current image data of the cabinet volume indicator as a set of cabinet volume data includes: Taking a fixed point at the top of the gas holder as the first endpoint, taking the midpoint of the top of the piston of the gas holder as the second endpoint, and taking a fixed point at the edge of the piston of the gas holder as the third endpoint, and obtaining the image of the first target triangular region enclosed by the first endpoint, the second endpoint and the third endpoint through the first camera assembly, and taking it as the current image data of the cabinet volume indicator; Obtaining the angles of the three apex angles within the first target triangular region, and associating the angles of the three apex angles with the current cabinet volume value to form a set of cabinet volume data.

[0009] After obtaining the angles of the three apex angles within the first target triangular region and associating the angles of the three apex angles with the current cabinet volume to form a set of cabinet volume data, it further includes: As a preferred embodiment of the method for identifying the mechanical cabinet volume image of the gas holder according to the present invention, wherein: taking the vertex of the pointer of the cabinet volume indicator panel as the fourth endpoint, taking the two endpoints of the cabinet volume indicator panel as the fifth endpoint and the sixth endpoint respectively, and obtaining the second target triangular region enclosed by the fourth endpoint, the fifth endpoint and the sixth endpoint through the second camera assembly; Obtaining the angles of the three apex angles within the second target triangular region, and associating the angles of the three apex angles within the first target triangular region and the angles of the three apex angles within the second target triangular region with the current cabinet volume value to form a set of cabinet volume data.

[0010] As a preferred embodiment of the method for identifying the mechanical cabinet volume image of the gas holder according to the present invention, wherein: the real-time image data of the cabinet volume indicator includes: Using a camera to obtain the image data of the piston at the top of the cabinet and the image data of the cabinet volume indicator panel at a fixed refresh rate.

[0011] As a preferred embodiment of the mechanical cabinet volume image recognition method of the present invention, the following steps are included: After inputting the real-time cabinet volume indication image data into the cabinet volume database and determining whether there is a corresponding cabinet volume value in the cabinet volume database, if there is, output the corresponding cabinet volume value and use it as the current cabinet volume value, the method further includes: If not, collect the next real-time cabinet volume indication image data and determine whether there is a corresponding cabinet volume value in the cabinet volume database until there is a corresponding cabinet volume value in the cabinet volume database for the current cabinet volume indication image data.

[0012] The beneficial effects of the present invention are as follows: (1) By constructing a cabinet volume database, the present invention performs image recognition on the collected cabinet volume indication image data and inputs it into the cabinet volume database for comparison, so as to identify the current cabinet volume value, effectively reducing errors and avoiding production accidents caused by excessive errors in the gas cabinet radar cabinet volume.

[0013] (2) By selecting multiple fixed endpoints to form two target triangular regions and associating the cabinet volume value with six angular values, the present invention effectively improves the accuracy of cabinet volume value recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0015] Figure 1 It is a schematic diagram of the distribution of the inner endpoint positions in the gas cabinet; Figure 2 It is a schematic diagram of the change of the first target triangle formed by the inner endpoints in the gas cabinet with the cabinet volume; Figure 3 It is a schematic diagram of the position of the pointer in the cabinet volume indicator and the second target triangle. DETAILED DESCRIPTION OF THE INVENTION

[0016] In order to make the content of the present invention more clearly understood, the following will further describe the present invention in detail according to the specific embodiments and in combination with the drawings.

[0017] The embodiment of the present application provides a mechanical cabinet volume image recognition method for a gas cabinet, which specifically includes the following steps: Step S101: Obtain all cabinet volume data within a preset accuracy range, construct a cabinet volume database, and each group of the cabinet volume data includes the current cabinet volume value and the cabinet volume indication image data.

[0018] Specifically, for the accuracy range required by the user, all cabinet volume data is obtained, and a cabinet volume database is constructed based on the cabinet volume data. Each group of cabinet volume data includes the current cabinet volume value and the cabinet volume indication image data.

[0019] It should be noted that the cabinet volume value data can be obtained according to the piston position in the gas holder and the cabinet volume value indicator panel. Therefore, the above-mentioned cabinet volume indication image data includes the piston position image data in the gas holder and the cabinet volume indicator panel image data. The specific operation steps are as follows: Step S101a: Obtain all cabinet volume values within the preset accuracy range.

[0020] Taking the accuracy value 1 as an example, all cabinet volume values within the accuracy range required by the user are 0, 1, 2... N, where N is the maximum cabinet volume.

[0021] Step S101b: For any cabinet volume value, collect the corresponding cabinet volume indication image data, and use the current cabinet volume value and the current cabinet volume indication image data as a group of cabinet volume data.

[0022] Specifically, first, the piston position in the gas holder will be described. Refer to Figure 1 、 Figure 2 , at the position of the railing on the top of the gas holder, a fixed point is taken as the first endpoint O, the midpoint of the top of the piston in the gas holder is taken as the second endpoint C, and a fixed point is taken on the edge of the piston in the gas holder as the third endpoint D. These three endpoints are connected by a straight line to enclose the first target triangular region X. A first camera assembly is installed directly above the position of the railing on the top of the gas holder to identify the positions of the above three endpoints. As the cabinet volume of the gas holder changes, the piston moves up and down reciprocally. Except for the top first endpoint O remaining stationary, the second endpoint C and the third endpoint D both move with the movement of the piston. Then, the three angles of the first target triangular region X change continuously with the movement of the piston, and any one cabinet volume value corresponds to only one unique triangle. By obtaining the angles of the three vertex angles within the first target triangular region and associating the angles of the three vertex angles with the current cabinet volume, a group of cabinet volume data can be formed. For example: Cabinet volume value 1 corresponds to X 1 : ∠OCD 1 、∠ODC 1 、∠DOC 1 Cabinet volume value 2 corresponds to X 2 : ∠OCD 2 、∠ODC 2 、∠DOC 2 Cabinet volume value 3 corresponds to X 3 : ∠OCD 3 、∠ODC 3 、∠DOC 3 The cabinet volume value 4 corresponds to X 4 : ∠OCD 4 , ∠ODC 4 , ∠DOC 4 .

[0023] After that, the image data of the cabinet volume indicator is used for explanation. See Figure 3 . Taking the vertex of the pointer of the cabinet volume indicator as the fourth endpoint P, and taking the two endpoints of the cabinet volume indicator as the fifth endpoint A and the sixth endpoint B respectively, connect these three endpoints through a straight line to enclose the second target triangular region Y. The second imaging component is used to obtain the second target triangular region Y enclosed by the fourth endpoint, the fifth endpoint and the sixth endpoint. As the cabinet volume changes, the cabinet volume steel wire rope pulls the cabinet volume indicator to move left and right, and the vertex P of the triangular region Y also moves left and right, that is, the three angles of the triangle change with the change of the cabinet volume and any cabinet volume value corresponds to a unique indicator value corresponding to a unique triangle. By obtaining the angles of the three top angles in the second target triangular region and associating the angles of the three top angles with the current cabinet volume, a set of cabinet volume data can be formed. For example: The cabinet volume value 1 corresponds to Y 1 : ∠PAB 1 , ∠PBA 1 , ∠APB 1 The cabinet volume value 2 corresponds to Y 1 : ∠PAB 2 , ∠PBA 2 , ∠APB 2 The cabinet volume value 3 corresponds to Y 1 : ∠PAB 3 , ∠PBA 3 , ∠APB 3 The cabinet volume value 4 corresponds to Y 1 : ∠PAB 4 , ∠PBA 4 , ∠APB 4 .

[0024] To sum up, any cabinet volume value can obtain two unique triangles X and Y, a total of six angles. By retaining the precision and converting the angles and numerical values, a set of data can be obtained.

[0025] Connect the two imaging components to the computer. The two imaging components respectively take pictures and collect the image data of the piston position in the gas holder and the image data of the cabinet volume indicator. The computer uses a preset program to identify the six vertex angle data in the first target triangular region X and the second target triangular region Y from the collected image data. Convert the six vertex angle data into a piece of data in pure digital format and associate it with the current cabinet volume value to form a set of cabinet volume data.

[0026] Step S101c: Obtain all cabinet capacity data within a preset accuracy range and construct a cabinet capacity database.

[0027] Specifically, by traversing all the cabinet capacity data within the accuracy range, the cabinet capacity database can be constructed.

[0028] Step S102: Collect real-time cabinet capacity indication image data.

[0029] Specifically, a camera is used to obtain the image data of the top piston of the cabinet and the image data of the cabinet capacity indicator disk at a fixed refresh rate.

[0030] Step S103: Input the real-time cabinet capacity indication image data into the cabinet capacity database, and determine whether there is a cabinet capacity value corresponding to the current cabinet capacity indication image data in the cabinet capacity database. If it exists, output the corresponding cabinet capacity value and use it as the current cabinet capacity value; if it does not exist, collect the next real-time cabinet capacity indication image data and determine whether there is a cabinet capacity value corresponding to the current cabinet capacity indication image data in the cabinet capacity database until there is a cabinet capacity value corresponding to the current cabinet capacity indication image data in the cabinet capacity database.

[0031] Specifically, two cameras are used to simultaneously capture the real-time image data of the corresponding cabinet capacity indicator disk at a fixed refresh rate and hand it over to the computer for recognition. The computer recognizes the angle values of the three apex angles of the target triangle, forms data in a preset format and compares it with the data in the database. If the comparison is successful, the cabinet capacity value of the gas holder is output. If the comparison fails, obtain the next real-time image data of the cabinet capacity indicator disk, and after computer recognition, compare it again until the comparison is successful.

[0032] Thus, through the technical solution of this application, by constructing a cabinet capacity database, after image recognition of the collected cabinet capacity indication image data and inputting it into the cabinet capacity database for comparison, the current cabinet capacity value can be recognized, effectively reducing errors and avoiding production accidents caused by excessive errors in the gas holder radar cabinet capacity.

[0033] In addition to the above embodiments, the present invention may have other embodiments; all technical solutions formed by equivalent replacement or equivalent transformation fall within the protection scope required by the present invention.

Claims

1. A method for recognizing the mechanical capacity of a gas tank, characterized by: include: Acquire all cabinet capacity data within a preset accuracy range and construct a cabinet capacity database, wherein each set of cabinet capacity data includes a current cabinet capacity value and cabinet capacity indication image data; Collect real-time cabinet capacity indication image data; The real-time cabinet capacity indication image data is input into the cabinet capacity database, and it is determined whether there is a cabinet capacity value corresponding to the current cabinet capacity indication image data in the cabinet capacity database. If there is, the corresponding cabinet capacity value is output and used as the current cabinet capacity value.

2. The method for recognizing the mechanical capacity of a gas tank according to claim 1, characterized in that: The acquisition of all cabinet capacity data within a preset accuracy range and the construction of a cabinet capacity database comprises: Get all cabinet capacity values ​​within the preset accuracy range; For any cabinet capacity value, the corresponding cabinet capacity indication image data is collected, and the current cabinet capacity value and the current cabinet capacity indication image data are used as a set of cabinet capacity data; Obtain all cabinet capacity data within the preset accuracy range and build a cabinet capacity database.

3. The method for recognizing the mechanical capacity of a gas tank according to claim 2, characterized in that: The collecting of corresponding cabinet capacity indication image data and taking the current cabinet capacity value and the current cabinet capacity indication image data as a set of cabinet capacity data includes: A fixed point is taken at the top of the gas cabinet as the first endpoint, the midpoint of the top of the gas cabinet piston is taken as the second endpoint, and a fixed point is taken at the edge of the gas cabinet piston as the third endpoint. The first target triangle area image surrounded by the first endpoint, the second endpoint and the third endpoint is acquired through the first camera assembly, and the image is used as the current cabinet capacity indication image data; The angles of the three vertices in the first target triangular area are obtained, and the angles of the three vertices are associated with the current cabinet capacity value to form a set of cabinet capacity data.

4. The method for recognizing the mechanical capacity of a gas tank according to claim 3, characterized in that: After obtaining the angles of the three vertices in the first target triangular area and associating the angles of the three vertices with the current cabinet capacity to form a set of cabinet capacity data, the method further includes: Taking the pointer vertex of the cabinet capacity indicator as the fourth endpoint, taking the two end points of the cabinet capacity indicator as the fifth endpoint and the sixth endpoint respectively, acquiring the second target triangular area enclosed by the fourth endpoint, the fifth endpoint and the sixth endpoint through the second camera assembly; The angles of the three vertices in the second target triangular area are obtained, and the angles of the three vertices in the first target triangular area and the three vertices in the second target triangular area are associated with the current cabinet capacity value to form a set of cabinet capacity data.

5. The method for recognizing the mechanical capacity of a gas tank according to claim 1, characterized in that: The real-time cabinet capacity indication image data includes: A camera is used to obtain the cabinet top piston image data and cabinet capacity indicator disk image data at a fixed refresh rate.

6. The method for recognizing the mechanical capacity of a gas tank according to claim 1, characterized in that: After inputting the real-time cabinet capacity indication image data into the cabinet capacity database and determining whether there is a cabinet capacity value corresponding to the current cabinet capacity indication image data in the cabinet capacity database, and if there is, outputting the corresponding cabinet capacity value and using it as the current cabinet capacity value, the method further includes: If not, collect the next real-time cabinet capacity indication image data, and determine whether there is a cabinet capacity value corresponding to the current cabinet capacity indication image data in the cabinet capacity database, until there is a cabinet capacity value corresponding to the current cabinet capacity indication image data in the cabinet capacity database.

Images