CN112036393A - Identification method based on shale gas field production single-pointer meter reading - Google Patents
Identification method based on shale gas field production single-pointer meter reading Download PDFInfo
- Publication number
- CN112036393A CN112036393A CN202010713388.5A CN202010713388A CN112036393A CN 112036393 A CN112036393 A CN 112036393A CN 202010713388 A CN202010713388 A CN 202010713388A CN 112036393 A CN112036393 A CN 112036393A
- Authority
- CN
- China
- Prior art keywords
- boundary
- image
- pointer
- instrument
- rectangle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/02—Recognising information on displays, dials, clocks
Abstract
The invention discloses a method for recognizing the reading of a single-pointer instrument produced based on a shale gas field, which relates to the field of intelligent instrument recognition, and is characterized in that a fuzzy mechanical instrument dial image shot by a camera on a production site is converted from an RGB space to an HSV space, then, V channel data of an image space is circularly traversed, and a pattern spot boundary under each V value is detected and searched; then constructing a central rectangle of a sub-image of the mechanical instrument, and filtering a formed pattern spot boundary; then, carrying out aspect ratio constraint on the filtered image spot boundary, carrying out constraint on the image spot boundary including a central rectangular central point, and then positioning the mechanical instrument pointer boundary; and finally, performing linear fitting and deflection angle calculation on the found pointer boundary, and calculating the meter reading, so that the data reading of the mechanical meter dial is realized by directly using a monitoring camera on a production field, and compared with the existing intelligent identification method, the accuracy of identifying the fuzzy picture is higher.
Description
Technical Field
The invention relates to the field of intelligent identification of instruments, in particular to an identification method based on the reading of a single-pointer instrument for shale gas field production.
Background
With the continuous increase of shale gas productivity, more and more unmanned inspection shale gas production platforms can be put into production sites. At present, a platform well which is critical in a shale gas production field is provided with a plurality of mechanical instruments with single pointers, the mechanical instruments are not connected with intelligent electronic equipment, reading cannot be directly remotely transmitted back to a rear end on-line platform, the existing production field generally adopts manual inspection, data are transmitted back to the rear end on-line platform through a terminal on-line platform after reading, labor cost is high, and data cannot be read at any time in real time.
And the shale gas production field is provided with cameras, and each camera can be provided with 256 preset positions at most for rapidly checking the field production equipment. These preset positions of present scene all are in the vacant state, and the effect of camera is just used for looking over the site conditions usefulness temporarily, and all the other times all are in no record monitor state, and camera utilization efficiency has very big promotion space.
In the prior art, a plurality of data identification technologies for instruments exist, but the data identification technologies are generally based on that instrument data shot by a camera is clear, because the monitoring camera of the shale gas production site mainly plays a monitoring role, the problem of the monitoring range is mainly considered during installation, the monitoring camera can not only face a mechanical instrument, moreover, the single pointer instrument picture of the shale gas production site station collected by the camera is generally not high in definition under the comprehensive influence of the factors of different rotation angles, different times, illumination of different weather images, different shooting distances, instrument panels with different sizes, instrument panels with different types and instrument panel panels of the camera and the like, the factors of the production site are not considered in the method disclosed in the prior art, if the prior art is directly used for identifying the instrument of the shale gas production site, the identification effect is not ideal.
Disclosure of Invention
The invention aims to: the method is used for processing images of a mechanical instrument with low definition acquired by a monitoring camera on a production site, computer vision analysis is carried out on a mechanical instrument dial plate, the reading of the instrument is accurately recognized, the purpose that reading of mechanical instrument data is finished by using a background on a camera matching line is achieved, the image with low definition acquired by the camera can be accurately recognized, and the problem that the reading of the instrument image with low definition cannot be accurately recognized by the existing data recognition technology for the instrument is solved.
The technical scheme adopted by the invention is as follows:
a method for identifying readings of a single-pointer instrument based on shale gas field production mainly comprises the following steps of:
step 2, image preprocessing: image preprocessing is carried out on the intercepted sub-image of the mechanical instrument, and image resampling is carried out;
and 3, image filtering: converting the resampled image from an RGB space to an HSV space, circularly traversing V channel data of the image space, and detecting and searching a spot boundary under each V value;
step 4, constructing a central rectangle: constructing a central rectangle of the sub-image of the mechanical instrument, and filtering the boundary of the image spot formed in the step 3 by using the constructed central rectangle;
step 5, positioning an instrument pointer: carrying out aspect ratio constraint on the filtered image spot boundary, carrying out constraint on the image spot boundary including a central rectangular central point, and then positioning the mechanical instrument pointer boundary;
and 6, identifying the degree of the pointer: and performing linear fitting and deflection angle calculation on the found pointer boundary, determining the rotation angle of the straight line, and calculating the reading of the instrument by combining the initial measuring range of the instrument.
The existing data identification technology for instruments is generally established on the basis that instrument image data shot by a camera is clear, and because a monitoring camera on a shale gas production site mainly plays a monitoring role, the problem of a monitoring range is mainly considered during installation and the monitoring camera cannot only face a mechanical instrument, so that the images shot by the camera of the mechanical instrument are fuzzy, the existing technology is difficult to identify the images shot by the existing monitoring camera, and the invention mainly aims at the problem.
The invention adopts the steps when processing the instrument image, wherein, the resampling in the step 2 extracts the instrument dial in the sub-image of the mechanical instrument; and in the step 3, the resampled image is transferred from the RGB space to the HSV space, an object with a certain color is easier to track than the RGB space in the HSV color space, for a fuzzy dial image, although the pointer and the dial are difficult to distinguish by eyes, the chromaticity H of the pointer and the chromaticity H of the dial are greatly different, but the brightness V values of the pointer and the dial are similar, so that the pointer can be easily found from the dial according to the characteristic that the H values of the dial and the pointer are greatly different. In addition, the scheme of the invention traverses the boundary of the image spot under different V values, and experience proves that the outline of the instrument pointer is related to a V channel, the variable is circularly traversed, an ideal target image is selected, which is a key ring for automatically identifying the pointer, and the quality of the target image determines whether the subsequent instrument reading can be successfully identified.
And 4, constructing the instrument center rectangle by taking the center of the resampled image as the center of a new rectangle and taking one half of the side length of the image as the side length of the rectangle, and constructing the new rectangle, so that the problem of searching the pointer in the whole image is converted into the problem of searching the pointer in the rectangle. The method can reduce the searching range of the pointer and make the foreground and background of the pointer obviously distinguished.
And 5, carrying out boundary searching and positioning on the filtered image to locate the pointer of the instrument, generating a minimum circumscribed rectangle of the pointer boundary, carrying out aspect ratio constraint on the generated minimum circumscribed rectangle and carrying out central rectangle central point constraint on the image spot boundary, and finding the pointer boundary meeting the conditions when the aspect ratio constraint and the central rectangle central point constraint of the image spot boundary meet preset constraint conditions. Boundary filtering is then performed, and the selected boundaries all fall within the central rectangle. And sequencing the inner boundaries of the central rectangle, wherein the boundary containing the most pixels is the pointer boundary.
And finally, using the step 6 to perform straight line fitting and deflection angle calculation on the found pointer boundary, fitting the straight line of the pointer through the straight line, determining the rotation angle of the fitted straight line, determining the rotation angle of the pointer, and calculating the reading of the instrument by combining the initial range of the corresponding instrument.
Further, the step 3 mainly comprises the following steps which are carried out in sequence:
step 3-1: transferring the resampled image from the RGB space to the HSV space;
step 3-2: and circularly traversing the V value of the HSV image space from 80 to 230, judging and searching the boundary of each image 150 times, and searching the boundary of the image spot.
It should be noted that HSV value ranges are different in different application scenarios, the invention selects image types by using cvSplit function in openCV, wherein H value range is 0-180, S value range is 0-255, V value range is 0-255, and error range of image spot boundary is too large when V value is too large and too small, so we select a V value range in a middle range to cycle through, and here we select V value range from 80 to 230.
Through analyzing the illumination of different rotation angles, different times, different weather images, instrument images under the conditions of different shooting distances, instrument panels with different sizes, instrument panels of different types and instrument panel panels and the like, the common characteristics of the images are abstracted out: that is, all images have a pointer, the color of the pointer is obviously different from that of the background of the dial plate, but the lightness of the pointer is similar. Therefore, the invention converts the reading problem of identifying the mechanical instrument into the rotation angle problem of identifying the instrument pointer. In addition, the invention selects HSV space to search the boundary of the pattern spot of the pointer, so that the identification method based on the reading of the single-pointer instrument for shale gas field production can identify the pointer, calculate the angle of the instrument and efficiently realize reading of the indication of the mechanical instrument under the comprehensive influences of the image light intensity, the distance between the instrument and a camera within a certain range, the installation angle of the instrument relative to the camera, the background shadow generated by sunlight and the like.
Further, during the loop traversal of the V-channel data in the image space in step 3, the impact factor of the loop termination mainly includes at least one of an aspect ratio of a minimum circumscribed rectangle of the boundary line, a side length ratio of a minimum side length of the minimum circumscribed rectangle to a central rectangle of the construction instrument, and an area ratio of the boundary line to the central rectangle of the construction instrument, and values of the impact factors are within a predefined threshold, the loop is terminated, otherwise, the loop traversal is continued to search for the blob boundary.
Further, in step 3, calling a CvtColor function to transfer the resampled image from the RGB space to the HSV space; in the process of circularly traversing the V-channel data of the image space, an inRange function is called to filter HSV image space target pixels for detecting and searching the image spot boundary under each V value.
Further, after the image spot boundary is found in the step 3, the image is converted from the HSV space to the RGB space.
Further, the step 4 mainly includes: and (3) constructing a central rectangle, circularly traversing the V-channel data in the step (3), adding constraints falling in the central rectangle and constraints of which the boundary center comprises the midpoint of a central sub-rectangle to the found image spot boundary when detecting and searching the image spot boundary under each V value, and sequencing the filtered image spot boundaries under each V value, wherein the image spot boundary with the largest number of boundaries is the specific image spot boundary.
Further, the step 5 mainly includes: and positioning the boundary of the found specific image spot to contain a pointer boundary, generating a minimum circumscribed rectangle of the found boundary, carrying out boundary filtering on three indexes, namely the height-width ratio of the found boundary circumscribed rectangle is greater than a preset threshold value, the number of boundary pixels is greater than the side length of the central rectangle of the image, and the area contained by the boundary is less than 30% of the area of the central rectangle of the image, wherein the selected boundaries all fall in the central rectangle, sequencing the inner boundaries of the central rectangle, and the boundary containing the most pixels is the pointer boundary.
And 4, 5, finding the pointer boundary, wherein the step 4 is used for finding out the optimal value in the image spot boundary corresponding to the V value when each V value in the value range is traversed circularly in the step 3, and using the optimal value as the specific image spot boundary corresponding to each V value, and the step 5 is used for screening all the specific image spot boundaries found in the step 4, and using the screened specific image spot boundaries as the pointer boundary.
Further, the step 6 of performing straight line fitting and calculation of the deflection angle on the found pointer boundary mainly includes: the method comprises the steps of taking two diagonal points of a minimum external rectangle as analysis objects, taking each point as a center, taking a short side of the minimum rectangle as a side length, constructing a new rectangle, then counting the number of boundary points falling on the inner sides of the two rectangles, taking the larger number of the boundary points as the tail of a pointer, taking the smaller number of the boundary points as the head of the pointer, fitting a pointer boundary straight line, taking the horizontal coordinates of the two diagonal points of the minimum external rectangle as variables, solving the corresponding vertical coordinates on the straight line, taking the solved two vertical coordinates as calculation objects, and calculating the deflection angle of the pointer.
When the minimum circumscribed rectangle of the pointer is generated, the rotation angle of the width side of the minimum circumscribed rectangle is usually expressed in the recognition result, but the angle is affected by the boundary of the recognition pointer, and the error is large. In order to refine the reading precision of the pointer, two diagonal points of the minimum external rectangle are used as analysis objects, each point is used as a center, and the short side of the minimum rectangle is used as the side length, so that a new rectangle is constructed. The number of boundary points falling within the two rectangles is then counted. The more the number is the tail of the pointer, and the less the number is the head of the pointer. And fitting a pointer boundary straight line, solving corresponding vertical coordinates on the straight line by taking the horizontal coordinates of two diagonal points of the minimum external rectangle as variables, and calculating the deflection angle of the pointer by taking the newly solved two vertical coordinates as calculation objects.
Further, for the image positioned and captured in the step 1, the width and height of the mechanical instrument dial are not less than 100 pixels, and the included angle between the mechanical instrument dial and the camera is not less than 30 degrees.
Although the invention aims at the image with low definition to recognize the reading of the instrument, if the definition is too low, the width or height of the image of the mechanical instrument dial is less than 100 pixels, or the plane included angle between the camera and the mechanical instrument dial is less than 30 degrees, the method of the invention can not recognize the reading of the pointer of the instrument more accurately.
And it should be noted that, the Resize function is called in the step 2 to resample the input image, so as to compensate the angle difference generated by the different installation angles of the instrument relative to the camera; calling a FindContours function in the step 4 to search, filter and search the image spot boundary formed in the step 3; in the step 6, a FitLine function is called to fit the boundary of the image spot into a straight line, then an Atan2 function is called to calculate the angle of the straight line fitted by the boundary of the image spot, and finally a MinAreaRect function is called to find out a rectangle with the minimum boundary and used for calculating the initial direction of the straight line.
A method for identifying the reading of a single-pointer instrument mainly comprises the following steps in sequence:
step 1: detecting and positioning the position of the mechanical instrument in the snapshot image by using a mechanical instrument model for deep learning training, intercepting subimages of the instrument, and performing computer vision analysis on the mechanical instrument;
step 2: image preprocessing is carried out on the intercepted sub-image of the mechanical instrument, and image resampling is carried out;
and step 3: converting the resampled image from an RGB space to an HSV space, circularly traversing V channel data of the image space, and detecting and searching a spot boundary under each V value;
and 4, step 4: constructing a central rectangle of the sub-image of the mechanical instrument, and filtering the boundary of the image spot formed in the step 3 by using the constructed central rectangle;
and 5: carrying out aspect ratio constraint on the filtered image spot boundary, carrying out constraint on the image spot boundary including a central rectangular central point, and then positioning the mechanical instrument pointer boundary;
step 6: and performing linear fitting and deflection angle calculation on the found pointer boundary, determining the rotation angle of the straight line, and calculating the reading of the instrument by combining the initial measuring range of the instrument.
The method is developed on the background of reading identification of a single-pointer mechanical instrument on a shale gas production site, but the method has wide universality and mainly aims at identifying the pointer of a dial plate aiming at pictures with low pointer or dial plate definition, so that the method can be used for identifying the reading of the instrument aiming at the single-pointer mechanical instrument on the shale gas production site, the single-pointer mechanical instrument on the production site and the picture with low definition, and the reading of the instrument is identified only if the width and the height of the picture of the mechanical instrument dial plate in the picture are not less than 100 pixels.
In summary, due to the adoption of the technical scheme, the invention has the beneficial effects that:
1. the invention relates to a method for identifying the reading of a single-pointer instrument based on shale gas field production, which comprises the steps of converting an image from an RGB space to an HSV space for a fuzzy mechanical instrument dial image shot by a camera on a production site, then carrying out computer vision analysis on the mechanical instrument dial to identify the reading of the instrument, and improving the identification rate for the fuzzy mechanical instrument dial image;
2. the invention relates to a method for identifying the reading of a single-pointer instrument based on shale gas field production, which is characterized in that a fuzzy mechanical instrument dial plate image shot by a camera on a production site is converted from an RGB space to an HSV space, then the mechanical instrument dial plate is subjected to computer vision analysis, and the reading of the instrument is identified, so that the purpose of directly using a monitoring camera on the production site to read the data of the mechanical instrument dial plate is realized, the data of the mechanical instrument is read by comparing with the existing manual inspection, the labor cost is saved, and the real-time efficiency is higher.
Drawings
The invention will now be described, by way of example, with reference to the accompanying drawings, in which:
FIG. 1 is a functional block diagram of the present invention;
FIG. 2 is a diagram of the key function calls of the present invention;
FIG. 3 is a schematic representation of an embodiment of the present invention after resampling;
FIG. 4 is a schematic representation of an embodiment of the present invention after spatial transformation;
FIG. 5 is a schematic representation of an embodiment of the present invention after plaque boundary detection;
FIG. 6 is a schematic diagram of a central sub-rectangle after an add constraint in accordance with an embodiment of the present invention;
FIG. 7 is a schematic illustration of one embodiment of the present invention after determining the direction of the line;
FIG. 8 is a schematic illustration of a determination of pointer reading calculation according to an embodiment of the present invention;
FIG. 9 is a schematic representation of another embodiment of the present invention after resampling;
FIG. 10 is a schematic representation of another embodiment of the present invention where the V value takes 70 after the spatial transformation;
FIG. 11 is a diagram illustrating a V value of 130 after a spatial transformation according to another embodiment of the present invention;
FIG. 12 is a diagram illustrating a V value of 160 after a spatial transformation according to another embodiment of the present invention;
FIG. 13 is a schematic representation of another embodiment of the present invention when the V value is taken to be 220 after spatial transformation;
FIG. 14 is a schematic diagram of another embodiment of the present invention after a center sub-rectangle addition constraint is constructed;
FIG. 15 is a schematic illustration of another embodiment of the present invention after determining the direction of the line;
FIG. 16 is a schematic illustration of a determination of pointer reading calculation according to another embodiment of the present invention.
Detailed Description
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it should be understood that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments, and therefore should not be considered as a limitation to the scope of protection. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
In the description of the present invention, it is to be noted that, unless otherwise explicitly specified or limited, the terms "disposed," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.
The present invention will be described in detail with reference to fig. 1 to 16.
Example 1
A method for identifying readings of a single-pointer instrument based on shale gas field production mainly comprises the following steps in sequence as shown in figure 1:
step 2, image preprocessing: image preprocessing is carried out on the intercepted sub-image of the mechanical instrument, and image resampling is carried out;
and 3, image filtering: converting the resampled image from an RGB space to an HSV space, circularly traversing V channel data of the image space, and detecting and searching a spot boundary under each V value;
step 4, constructing a central rectangle: constructing a central rectangle of the sub-image of the mechanical instrument, and filtering the boundary of the image spot formed in the step 3 by using the constructed central rectangle;
step 5, positioning an instrument pointer: carrying out aspect ratio constraint on the filtered image spot boundary, carrying out constraint on the image spot boundary including a central rectangular central point, and then positioning the mechanical instrument pointer boundary;
and 6, identifying the degree of the pointer: and performing linear fitting and deflection angle calculation on the found pointer boundary, determining the rotation angle of the straight line, and calculating the reading of the instrument by combining the initial measuring range of the instrument.
The existing data identification technology for instruments is generally established on the basis that instrument image data shot by a camera is clear, and because a monitoring camera on a shale gas production site mainly plays a monitoring role, the problem of a monitoring range is mainly considered during installation and the monitoring camera cannot only face a mechanical instrument, so that the images shot by the camera of the mechanical instrument are fuzzy, the existing technology is difficult to identify the images shot by the existing monitoring camera, and the invention mainly aims at the problem.
The working principle is as follows: the invention adopts the steps when processing the instrument image, wherein, the resampling in the step 2 extracts the instrument dial in the sub-image of the mechanical instrument; and in the step 3, the resampled image is transferred from the RGB space to the HSV space, an object with a certain color is easier to track than the RGB space in the HSV color space, for a fuzzy dial image, although the pointer and the dial are difficult to distinguish by eyes, the chromaticity H of the pointer and the chromaticity H of the dial are greatly different, but the brightness V values of the pointer and the dial are similar, so that the pointer can be easily found from the dial according to the characteristic that the H values of the dial and the pointer are greatly different. In addition, the scheme of the invention traverses the boundary of the image spot under different V values, and experience proves that the outline of the instrument pointer is related to a V channel, the variable is circularly traversed, an ideal target image is selected, which is a key ring for automatically identifying the pointer, and the quality of the target image determines whether the subsequent instrument reading can be successfully identified.
And 4, constructing the instrument center rectangle by taking the center of the resampled image as the center of a new rectangle and taking one half of the side length of the image as the side length of the rectangle, and constructing the new rectangle, so that the problem of searching the pointer in the whole image is converted into the problem of searching the pointer in the rectangle. The method can reduce the searching range of the pointer and make the foreground and background of the pointer obviously distinguished.
And 5, carrying out boundary searching and positioning on the filtered image to locate the pointer of the instrument, generating a minimum circumscribed rectangle of the pointer boundary, carrying out aspect ratio constraint on the generated minimum circumscribed rectangle and carrying out central rectangle central point constraint on the image spot boundary, and finding the pointer boundary meeting the conditions when the aspect ratio constraint and the central rectangle central point constraint of the image spot boundary meet preset constraint conditions. Boundary filtering is then performed, and the selected boundaries all fall within the central rectangle. And sequencing the inner boundaries of the central rectangle, wherein the boundary containing the most pixels is the pointer boundary.
And finally, using the step 6 to perform straight line fitting and deflection angle calculation on the found pointer boundary, fitting the straight line of the pointer through the straight line, determining the rotation angle of the fitted straight line, determining the rotation angle of the pointer, and calculating the reading of the instrument by combining the initial range of the corresponding instrument.
The clear image which can be accurately identified by the prior art is compared with the blurred image which can not be accurately identified by the prior art, the pointer and the dial plate in the clear image have obvious difference in RGB color space, the difference of the blurred image in the RGB color space is very complicated and difficult to distinguish by using a formula, so if the method is continuously improved in the RGB color space, it is only useless, in HSV color space, the chromaticity H of the pointer and the dial plate of the clear image is greatly different, the brightness V values are close and larger, and the chroma H values of the pointer and the dial plate of the blurred image are also greatly different, but the brightness V values are close and small, so that only the V value of the whole blurred image needs to be controlled, when the chromaticity difference between the pointer and the dial plate is maximum, the boundary of the pointer is found out, and then the pointer can be positioned to read the reading of the dial plate.
The method is developed on the background of reading identification of a single-pointer mechanical instrument on a shale gas production site, but the method has wide universality and mainly aims at identifying the pointer of a dial plate aiming at pictures with low pointer or dial plate definition, so that the method can be used for identifying the reading of the instrument aiming at the single-pointer mechanical instrument on the shale gas production site, the single-pointer mechanical instrument on the production site and the picture with low definition, and the reading of the instrument is identified only if the width and the height of the picture of the mechanical instrument dial plate in the picture are not less than 100 pixels.
Example 2
On the basis of the above embodiment 1, the step 3 mainly includes the following steps performed in sequence:
step 3-1: transferring the resampled image from the RGB space to the HSV space;
step 3-2: and circularly traversing the V value of the HSV image space from 80 to 230, judging and searching the boundary of each image 150 times, and searching the boundary of the image spot.
It should be noted that HSV value ranges are different in different application scenarios, as shown in fig. 2, the invention selects an image type by using cvSplit function in openCV, where H value ranges from 0 to 180, S value ranges from 0 to 255, and V value ranges from 0 to 255, and an error range of a map-patch boundary is too large when V value is too large and too small, so that we select a V value range in a middle range for circular traversal, where we select a V value range from 80 to 230.
The working principle is as follows: through analyzing the illumination of different rotation angles, different times, different weather images, instrument images under the conditions of different shooting distances, instrument panels with different sizes, instrument panels of different types and instrument panel panels and the like, the common characteristics of the images are abstracted out: that is, all images have a pointer, the color of the pointer is obviously different from that of the background of the dial plate, but the lightness of the pointer is similar. Therefore, the invention converts the reading problem of identifying the mechanical instrument into the rotation angle problem of identifying the instrument pointer. In addition, the invention selects HSV space to search the boundary of the pattern spot of the pointer, so that the identification method based on the reading of the single-pointer instrument for shale gas field production can identify the pointer, calculate the angle of the instrument and efficiently realize reading of the indication of the mechanical instrument under the comprehensive influences of the image light intensity, the distance between the instrument and a camera within a certain range, the installation angle of the instrument relative to the camera, the background shadow generated by sunlight and the like.
Example 3
On the basis of the foregoing embodiment 1 or 2, in the step 3, when the V-channel data in the image space is subjected to loop traversal, the impact factor of loop termination mainly includes at least one of an aspect ratio of a minimum circumscribed rectangle of the boundary line, a side length ratio of a minimum circumscribed rectangle minimum side length ratio to a central rectangle of the construction instrument, and an area ratio of the boundary line to the central rectangle of the construction instrument, and values of the impact factors are within a predefined threshold, the loop is terminated, otherwise, the loop traversal is continued to search for the blob boundary.
Example 4
On the basis of the above embodiment 1 or 2, as shown in fig. 2, in step 3, a CvtColor function is called to transfer the resampled image from the RGB space to the HSV space; in the process of circularly traversing the V-channel data of the image space, an inRange function is called to filter HSV image space target pixels for detecting and searching the image spot boundary under each V value.
And 3, converting the image from the HSV space to the RGB space after the image spot boundary is found in the step 3. And the ConvertTo function is called to convert the image containing the filtered spot boundaries from the HSV space to the RGB space.
Example 5
On the basis of the foregoing embodiment 1, in steps 4 and 5 of embodiment 1, it is necessary to screen a pointer boundary that we need from a large number of patch boundaries corresponding to all traversed V values, first we screen a specific patch boundary from the patch boundaries, we judge the patch boundary found under each V value, construct a central sub-rectangle, and retain the patch boundaries that fall within the central sub-rectangle, so as to reduce the number of patch boundaries, and sort the patch boundaries that satisfy the following judgment conditions:
judgment condition 1: the quotient of the length divided by the width of the smallest circumscribed rectangle of the pattern spot boundary is greater than 3;
judgment condition 2: the polygon formed by the image spot boundary comprises the central point of the central sub-rectangular area;
the one with the largest number of boundaries after sorting is selected as the specific spot boundary.
During the circulation of the V channel data in the image space, the judgment and the sequencing are carried out on the image spot boundaries of the images corresponding to each V value, a specific image spot boundary is found, the average value, the variance and the standard deviation are calculated according to the pixel number of all the specific boundaries found out from the images corresponding to all the V values, the minimum circumscribed rectangle of the found specific image spot boundary is generated for the image of each specific image spot boundary, and the specific image spot boundary with the largest pixel number and meeting the following judgment conditions is selected as a pointer boundary:
judgment condition 3: falls within one standard deviation;
judgment condition 4: the number of pixels of the boundary of the specific image spot is larger than the side length of the central rectangle of the image;
judgment condition 5: the area encompassed by the particular spot boundary is less than 30% of the area of the central sub-rectangle.
Example 6
On the basis of the foregoing embodiment 1, the performing of straight line fitting and calculation of a deflection angle on the pointer boundary found in step 6 mainly includes: the method comprises the steps of taking two diagonal points of a minimum external rectangle as analysis objects, taking each point as a center, taking a short side of the minimum rectangle as a side length, constructing a new rectangle, then counting the number of boundary points falling on the inner sides of the two rectangles, taking the larger number of the boundary points as the tail of a pointer, taking the smaller number of the boundary points as the head of the pointer, fitting a pointer boundary straight line, taking the horizontal coordinates of the two diagonal points of the minimum external rectangle as variables, solving the corresponding vertical coordinates on the straight line, taking the solved two vertical coordinates as calculation objects, and calculating the deflection angle of the pointer.
The working principle is as follows: when the minimum circumscribed rectangle of the pointer is generated, the rotation angle of the width side of the minimum circumscribed rectangle is usually expressed in the recognition result, but the angle is affected by the boundary of the recognition pointer, and the error is large. In order to refine the reading precision of the pointer, two diagonal points of the minimum external rectangle are used as analysis objects, each point is used as a center, and the short side of the minimum rectangle is used as the side length, so that a new rectangle is constructed. The number of boundary points falling within the two rectangles is then counted. The more the number is the tail of the pointer, and the less the number is the head of the pointer. And fitting a pointer boundary straight line, solving corresponding vertical coordinates on the straight line by taking the horizontal coordinates of two diagonal points of the minimum external rectangle as variables, and calculating the deflection angle of the pointer by taking the newly solved two vertical coordinates as calculation objects.
And (3) calculating the reading of the meter: the angle between the default instrument starting scale and the default instrument ending scale is 90 degrees, the default instrument panel full scale is 270 degrees, the default instrument panel starting scale is 225 degrees relative to the X axis, the instrument pointer is in a clamping value range of (-180, 180 degrees) relative to the axis, and then the instrument reading is as follows:
example 7
On the basis of the embodiment 1, for the image positioned and captured in the step 1, the width and the height of the image of the mechanical instrument dial are not less than 100 pixels, and the included angle between the mechanical instrument dial and the camera is not less than 30 degrees.
Although the invention aims at the image with low definition to recognize the reading of the instrument, if the definition is too low, the width or height of the image of the mechanical instrument dial is less than 100 pixels, or the plane included angle between the camera and the mechanical instrument dial is less than 30 degrees, the method of the invention can not recognize the reading of the pointer of the instrument more accurately.
And it should be noted that, the Resize function is called in the step 2 to resample the input image, so as to compensate the angle difference generated by the different installation angles of the instrument relative to the camera; calling a FindContours function in the step 4 to search, filter and search the image spot boundary formed in the step 3; in the step 6, a FitLine function is called to fit the boundary of the image spot into a straight line, then an Atan2 function is called to calculate the angle of the straight line fitted by the boundary of the image spot, and finally a MinAreaRect function is called to find out a rectangle with the minimum boundary and used for calculating the initial direction of the straight line.
Example 8
Referring to fig. 3 to 8, the present invention is a process image for identifying a pressure gauge dial of a shale gas wellhead based on the identification method for reading single pointer instrument in shale gas field production, wherein:
as shown in fig. 3, after step 2 is performed, the image of the back bottom is resampled, and at this time, the image is in RGB space, and the color of the pointer and the color of the dial are similar;
as shown in fig. 4, after step 3-1 is performed, the space is transformed into HSV space, and then step 3-2 is performed, when the V value of HSV space is traversed, the image when the V value is 113 can be obviously seen by naked eyes, and the boundary of the chart spot of the dial plate is clear after the pointer is compared with the background;
step 4, calculating the graph to determine the boundary of the pointer, which is one of the specific boundaries and is the final determined boundary of the pointer; we transform it back into RGB space, as in fig. 5, which is the diagram of the dial containing the pointer boundaries;
as shown in fig. 6, the diagram of the pointer boundary finally determined by the dial after the step 5 is performed contains the pointer boundary, the minimum bounding rectangle and the center sub-rectangle, in fig. 6, the larger rectangle is the center sub-rectangle, and the smaller rectangle falling within the center sub-rectangle is the minimum bounding rectangle, and fig. 6 satisfies the following conditions:
judgment condition 1: the quotient of the length divided by the width of the minimum bounding rectangle of the pointer boundary is greater than 3;
judgment condition 2: the polygon formed by the pointer boundary comprises the central point of the central sub-rectangular area;
judgment condition 3: the pointer boundary falls within one standard deviation of the specific blob boundary for all V values;
judgment condition 4: the number of pixels of the pointer boundary is larger than the side length of the image center rectangle;
judgment condition 5: the pointer boundaries comprise less than 30% of the area of the central sub-rectangle.
As shown in fig. 7, the straight line more than that shown in fig. 6 is the straight line corresponding to the pointer fitted after the straight line fitting calculation in step 6 is continued, the two diagonal points circled in the minimum circumscribed rectangle in fig. 7 are used, the abscissa of the two diagonal points is used as a variable to calculate the corresponding ordinate on the straight line, the two newly calculated ordinates are used as calculation objects to calculate the deflection angle of the pointer, and the deflection angle of the pointer in the figure with respect to the X axis is 154 °.
Fig. 8 is a diagram when the final deflection angle calculation and the meter reading calculation in step 6 are continued, and for the straight line corresponding to the pointer fitted in fig. 7, the included angle between the start scale and the end scale of the pressure meter is 90 °, the range is 0-16 Mpa, the full scale of the default instrument panel is 270 °, the angle of the start scale of the default instrument panel relative to the X axis is 225 °, the straight lines of the third quadrant and the fourth quadrant in the coordinate system established in fig. 8 are the start scale and the end scale of the image dial, and finally the start scale and the end scale of the image dial are found through the formula
Namely, it is
The calculation can result in the pointer reading being 4.21Mpa and the actual pointer reading being 4.20 Mpa.
Example 9
Fig. 9-16 are process images of identifying a pressure gauge dial of a shale gas H5-4 wellhead based on the identification method of the reading of the single pointer instrument for shale gas field production according to the present invention, wherein:
fig. 9 is the image of the back bottom after the resampling in step 2, at this time, the image is in RGB space, and the color of the pointer and the dial are similar;
as shown in fig. 10-13, after performing step 3-1, the space is transformed into HSV space, and then performing step 3-2, when traversing the V value of HSV space, the image when the V value is different, wherein fig. 10 is the image when the V value is 70, fig. 11 is the image when the V value is 130, fig. 12 is the image when the V value is 160, and fig. 13 is the image when the V value is 220.
Step 4, the calculated and determined image spot boundary of the pointer is one of the specific image spot boundaries and is the finally solved pointer boundary; after we transform it back to RGB space again, as in fig. 14, which is a diagram of this dial containing the pointer boundaries, the minimum bounding rectangle and the central sub-rectangle, where the larger rectangle is the central sub-rectangle and the smaller rectangle falling within the central sub-rectangle is the minimum bounding rectangle, fig. 14 satisfies the following condition:
judgment condition 1: the quotient of the length divided by the width of the minimum bounding rectangle of the pointer boundary is greater than 3;
judgment condition 2: the polygon formed by the pointer boundary comprises the central point of the central sub-rectangular area;
judgment condition 3: the pointer boundary falls within one standard deviation of the specific blob boundary for all V values;
judgment condition 4: the number of pixels of the pointer boundary is larger than the side length of the image center rectangle;
judgment condition 5: the pointer boundaries comprise less than 30% of the area of the central sub-rectangle.
As shown in fig. 15, the straight line more than that shown in fig. 14 is the straight line corresponding to the pointer fitted after the straight line fitting calculation in step 6 is continued, the two diagonal points circled in the minimum circumscribed rectangle in fig. 7 are used, the abscissa of the two diagonal points is used as a variable to calculate the corresponding ordinate on the straight line, the two newly calculated ordinates are used as calculation objects to calculate the deflection angle of the pointer, and the deflection angle of the pointer in the figure with respect to the X axis is 203 °.
Fig. 16 is a diagram of the final deflection angle calculation and the meter reading calculation in step 6, and for the straight line corresponding to the pointer fitted in fig. 15, the included angle between the start scale and the end scale of the pressure meter is 90 °, the range is 0-25 Mpa, the full scale of the default instrument panel is 270 °, the angle of the start scale of the default instrument panel relative to the X axis is 225 °, the straight lines of the third quadrant and the fourth quadrant in the coordinate system established in fig. 8 are the start scale and the end scale of the image dial, and finally the calculation is performed by the formula
Namely, it is
The calculation can result in the pointer reading being 2.04Mpa and the actual reading being 1.97 Mpa.
Through the application of the embodiments 8 and 9 and the actual production process, the error rate is below 5 percent, and the production requirement is met.
The above description is only a preferred embodiment of the present invention, and is not intended to limit the present invention in any way, and all simple modifications and equivalent variations of the above embodiments according to the technical spirit of the present invention are included in the scope of the present invention.
Claims (10)
1. A method for recognizing the reading of a single-pointer instrument based on shale gas field production is characterized by comprising the following steps: mainly comprises the following steps which are carried out in sequence:
step 1: detecting and positioning the position of the mechanical instrument in the snapshot image by using a mechanical instrument model for deep learning training, intercepting subimages of the instrument, and performing computer vision analysis on the mechanical instrument;
step 2: image preprocessing is carried out on the intercepted sub-image of the mechanical instrument, and image resampling is carried out;
and step 3: converting the resampled image from an RGB space to an HSV space, circularly traversing V channel data of the image space, and detecting and searching a spot boundary under each V value;
and 4, step 4: constructing a central rectangle of the sub-image of the mechanical instrument, and filtering the boundary of the image spot formed in the step 3 by using the constructed central rectangle;
and 5: carrying out aspect ratio constraint on the filtered image spot boundary, carrying out constraint on the image spot boundary including a central rectangular central point, and then positioning the mechanical instrument pointer boundary;
step 6: and performing linear fitting and deflection angle calculation on the found pointer boundary, determining the rotation angle of the straight line, and calculating the reading of the instrument by combining the initial measuring range of the instrument.
2. The identification method based on shale gas field production single pointer meter reading according to claim 1, characterized in that: the step 3 mainly comprises the following steps which are carried out in sequence:
step 3-1: transferring the resampled image from the RGB space to the HSV space;
step 3-2: and circularly traversing the V value of the HSV image space from 80 to 230, judging and searching the boundary of each image 150 times, and searching the boundary of the image spot.
3. The identification method based on shale gas field production single pointer meter reading according to claim 1 or 2, characterized in that: and in the step 3, when the V channel data of the image space is circulated, the influence factors of the circulation termination mainly comprise at least one of the aspect ratio of the minimum external rectangle of the boundary line, the side length ratio of the minimum external rectangle to the central rectangle of the construction instrument and the area ratio of the boundary line to the central rectangle of the construction instrument, the circulation termination is carried out when the values of the influence factors are within the predefined threshold value, and otherwise, the circulation traversal is continued to search the boundary of the image spot.
4. The identification method based on shale gas field production single pointer meter reading according to claim 1 or 2, characterized in that: in step 3, calling a CvtColor function to transfer the resampled image from the RGB space to the HSV space; in the process of circularly traversing the V-channel data of the image space, an inRange function is called to filter HSV image space target pixels for detecting and searching the image spot boundary under each V value.
5. The identification method based on shale gas field production single pointer meter reading according to claim 1 or 2, characterized in that: and 3, converting the image from the HSV space to the RGB space after the image spot boundary is found in the step 3.
6. The identification method based on shale gas field production single pointer meter reading according to claim 1, characterized in that: the step 4 mainly comprises: and (3) constructing a central rectangle, circularly traversing the V-channel data in the step (3), adding constraints falling in the central rectangle and constraints of which the boundary center comprises the midpoint of a central sub-rectangle to the found image spot boundary when detecting and searching the image spot boundary under each V value, and sequencing the filtered image spot boundaries under each V value, wherein the image spot boundary with the largest number of boundaries is the specific image spot boundary.
7. The identification method based on shale gas field production single pointer meter reading according to claim 1, characterized in that: the step 5 mainly comprises: and positioning the boundary of the searched image spot to contain a pointer boundary, generating a minimum circumscribed rectangle of the searched boundary, carrying out boundary filtering on three indexes, namely the height-width ratio of the searched boundary circumscribed rectangle is greater than a preset threshold value, the number of boundary pixels is greater than the side length of the central rectangle of the image, and the area contained by the boundary is less than 30% of the area of the central rectangle of the image, wherein the selected boundaries all fall in the central rectangle, sequencing the inner boundaries of the central rectangle, and the boundary containing the most pixels is the pointer boundary.
8. The identification method based on shale gas field production single pointer meter reading according to claim 1, characterized in that: in step 6, the straight line fitting and deflection angle calculation of the found pointer boundary mainly comprises the following steps: the method comprises the steps of taking two diagonal points of a minimum external rectangle as analysis objects, taking each point as a center, taking a short side of the minimum rectangle as a side length, constructing a new rectangle, then counting the number of boundary points falling on the inner sides of the two rectangles, taking the larger number of the boundary points as the tail of a pointer, taking the smaller number of the boundary points as the head of the pointer, fitting a pointer boundary straight line, taking the horizontal coordinates of the two diagonal points of the minimum external rectangle as variables, solving the corresponding vertical coordinates on the straight line, taking the solved two vertical coordinates as calculation objects, and calculating the deflection angle of the pointer.
9. The identification method based on shale gas field production single pointer meter reading according to claim 1, characterized in that: for the image positioned and captured in the step 1, the width and the height of the image of the mechanical instrument dial are not less than 100 pixels, and the included angle between the mechanical instrument dial and the camera is not less than 30 degrees.
10. A method for identifying the reading of a pointer instrument is characterized in that: mainly comprises the following steps which are carried out in sequence:
step 1: detecting and positioning the position of the mechanical instrument in the snapshot image by using a mechanical instrument model for deep learning training, intercepting subimages of the instrument, and performing computer vision analysis on the mechanical instrument;
step 2: image preprocessing is carried out on the intercepted sub-image of the mechanical instrument, and image resampling is carried out;
and step 3: converting the resampled image from an RGB space to an HSV space, circularly traversing V channel data of the image space, and detecting and searching a spot boundary under each V value;
and 4, step 4: constructing a central rectangle of the sub-image of the mechanical instrument, and filtering the boundary of the image spot formed in the step 3 by using the constructed central rectangle;
and 5: carrying out aspect ratio constraint on the filtered image spot boundary, carrying out constraint on the image spot boundary including a central rectangular central point, and then positioning the mechanical instrument pointer boundary;
step 6: and performing linear fitting and deflection angle calculation on the found pointer boundary, determining the rotation angle of the straight line, and calculating the reading of the instrument by combining the initial measuring range of the instrument.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010713388.5A CN112036393B (en) | 2020-07-22 | 2020-07-22 | Shale gas field production single pointer instrument reading based identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010713388.5A CN112036393B (en) | 2020-07-22 | 2020-07-22 | Shale gas field production single pointer instrument reading based identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112036393A true CN112036393A (en) | 2020-12-04 |
| CN112036393B CN112036393B (en) | 2023-08-18 |
Family
ID=73582928
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010713388.5A Active CN112036393B (en) | 2020-07-22 | 2020-07-22 | Shale gas field production single pointer instrument reading based identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112036393B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116071736A (en) * | 2023-03-07 | 2023-05-05 | 北京全路通信信号研究设计院集团有限公司 | Meter reading method and device, electronic equipment and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080011025A1 (en) * | 2006-07-14 | 2008-01-17 | Lin Chu P | Combination and key operated padlock with device for indicating the lock has been opened by a key |
| CN104392206A (en) * | 2014-10-24 | 2015-03-04 | 南京航空航天大学 | Image processing method for automatic pointer-type instrument reading recognition |
| CN104751187A (en) * | 2015-04-14 | 2015-07-01 | 山西科达自控股份有限公司 | Automatic meter-reading image recognition method |
| CN106228161A (en) * | 2016-07-18 | 2016-12-14 | 电子科技大学 | A kind of pointer-type dial plate automatic reading method |
| CN106650697A (en) * | 2016-12-30 | 2017-05-10 | 亿嘉和科技股份有限公司 | Instrument scale recognition method |
| CN108573261A (en) * | 2018-04-17 | 2018-09-25 | 国家电网公司 | A kind of read out instrument recognition methods suitable for Intelligent Mobile Robot |
| JP2019096304A (en) * | 2017-11-17 | 2019-06-20 | キヤノン株式会社 | Image processing device, control method thereof, and program |
| CN111414934A (en) * | 2020-01-21 | 2020-07-14 | 国网浙江省电力有限公司湖州供电公司 | Pointer type meter reading automatic identification method based on fast R-CNN and U-Net |
-
2020
- 2020-07-22 CN CN202010713388.5A patent/CN112036393B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080011025A1 (en) * | 2006-07-14 | 2008-01-17 | Lin Chu P | Combination and key operated padlock with device for indicating the lock has been opened by a key |
| CN104392206A (en) * | 2014-10-24 | 2015-03-04 | 南京航空航天大学 | Image processing method for automatic pointer-type instrument reading recognition |
| CN104751187A (en) * | 2015-04-14 | 2015-07-01 | 山西科达自控股份有限公司 | Automatic meter-reading image recognition method |
| CN106228161A (en) * | 2016-07-18 | 2016-12-14 | 电子科技大学 | A kind of pointer-type dial plate automatic reading method |
| CN106650697A (en) * | 2016-12-30 | 2017-05-10 | 亿嘉和科技股份有限公司 | Instrument scale recognition method |
| JP2019096304A (en) * | 2017-11-17 | 2019-06-20 | キヤノン株式会社 | Image processing device, control method thereof, and program |
| CN108573261A (en) * | 2018-04-17 | 2018-09-25 | 国家电网公司 | A kind of read out instrument recognition methods suitable for Intelligent Mobile Robot |
| CN111414934A (en) * | 2020-01-21 | 2020-07-14 | 国网浙江省电力有限公司湖州供电公司 | Pointer type meter reading automatic identification method based on fast R-CNN and U-Net |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116071736A (en) * | 2023-03-07 | 2023-05-05 | 北京全路通信信号研究设计院集团有限公司 | Meter reading method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112036393B (en) | 2023-08-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102663360B (en) | Method for automatic identifying steel slab coding and steel slab tracking system | |
| CN109900711A (en) | Workpiece, defect detection method based on machine vision | |
| CN110807355A (en) | Pointer instrument detection and reading identification method based on mobile robot | |
| CN107729853A (en) | A kind of automatic identifying method suitable for the narrow tuning drive gear formula instrument of transformer station | |
| CN109447062A (en) | Pointer-type gauges recognition methods based on crusing robot | |
| CN104749801B (en) | High Precision Automatic optical detecting method and system | |
| CN108182682A (en) | A kind of detection method based on Computer Image Processing and pattern-recognition | |
| CN113688817A (en) | Instrument identification method and system for automatic inspection | |
| CN111563896A (en) | Image processing method for catenary anomaly detection | |
| CN109389165A (en) | Oil level gauge for transformer recognition methods based on crusing robot | |
| CN111666880A (en) | Intelligent identification system for fire extinguisher pointer instrument | |
| CN112036391B (en) | Electronic inspection method and system based on shale gas field production station | |
| CN112036393B (en) | Shale gas field production single pointer instrument reading based identification method | |
| CN114820439A (en) | PCB bare board defect detection system and method based on AOI | |
| CN114494169A (en) | Industrial flexible object detection method based on machine vision | |
| CN114187253A (en) | Circuit board part installation detection method | |
| CN108254380A (en) | PCB circuit board template matching method based on Digital Image Processing | |
| CN109784257B (en) | Transformer thermometer detection and identification method | |
| CN115082509B (en) | Method for tracking non-feature target | |
| CN114638805A (en) | Track slab crack detection method and system and storage medium | |
| CN116245793A (en) | Color ring resistor resistance value detection method based on vision | |
| CN111652055B (en) | Intelligent switch instrument identification method based on two-stage positioning | |
| CN109858474B (en) | Detection and identification method for transformer oil surface temperature controller | |
| CN106971391A (en) | A kind of multi-template positioning wicking detection method based on colour recognition | |
| CN114235815A (en) | Method for detecting surface defects of outdoor electrical equipment of converter station based on scene filtering |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |