CN112507838B - Pointer meter identification method and device and electric power inspection robot - Google Patents
Pointer meter identification method and device and electric power inspection robot Download PDFInfo
- Publication number
- CN112507838B CN112507838B CN202011388525.9A CN202011388525A CN112507838B CN 112507838 B CN112507838 B CN 112507838B CN 202011388525 A CN202011388525 A CN 202011388525A CN 112507838 B CN112507838 B CN 112507838B
- Authority
- CN
- China
- Prior art keywords
- pointer
- image
- index
- area
- region
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000007689 inspection Methods 0.000 title claims abstract description 32
- 238000009826 distribution Methods 0.000 claims abstract description 25
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 16
- 238000004364 calculation method Methods 0.000 claims abstract description 16
- 238000001514 detection method Methods 0.000 claims abstract description 12
- 239000011159 matrix material Substances 0.000 claims description 16
- 238000012216 screening Methods 0.000 claims description 8
- 230000001629 suppression Effects 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 7
- 238000003708 edge detection Methods 0.000 claims description 7
- 238000003860 storage Methods 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 3
- 239000003550 marker Substances 0.000 claims description 2
- 239000000758 substrate Substances 0.000 claims 1
- 238000012937 correction Methods 0.000 abstract description 4
- 238000002372 labelling Methods 0.000 abstract description 4
- 230000008569 process Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 238000013507 mapping Methods 0.000 description 4
- 238000003702 image correction Methods 0.000 description 3
- VQKWAUROYFTROF-UHFFFAOYSA-N arc-31 Chemical compound O=C1N(CCN(C)C)C2=C3C=C4OCOC4=CC3=NN=C2C2=C1C=C(OC)C(OC)=C2 VQKWAUROYFTROF-UHFFFAOYSA-N 0.000 description 2
- 238000005520 cutting process Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
-
- 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
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/242—Aligning, centring, orientation detection or correction of the image by image rotation, e.g. by 90 degrees
-
- 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
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- 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
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a pointer meter identification method and device and an electric power inspection robot. The method comprises the following steps: acquiring a first index mark image generated by shooting an index mark by the electric power inspection robot in an elevation angle view; converting the first index mark image into a second index mark image of a front view angle; determining a pointer region from the second pointer-marked image based on maximum-connected-amount region detection, and fitting a pointer straight line based on the pointer region; in the remaining area of the second index image after the pointer area is removed, a plurality of scale value areas are determined based on pixel value distribution calculation, and a dial arc is fitted based on the plurality of scale value areas; and determining the reading of the pointer table based on the intersection point of the pointer straight line and the circular arc of the dial. The embodiment of the invention provides an automatic correction algorithm for images, provides good preconditions for pointer meter identification, also remarkably reduces manual labeling work, and simultaneously ensures the accuracy of reading identification.
Description
Technical Field
The embodiment of the invention relates to the technical field of intelligent patrol, in particular to a pointer meter identification method and device and an electric patrol robot.
Background
Along with the economic development and the improvement of the living standard of people, the social electricity consumption is continuously improved, higher challenges are provided for the running stability of the power grid, and the inspection requirements of all links of the power system such as power transformation, power transmission, power distribution and the like are further improved.
In recent years, the national electric power mechanism greatly promotes the intelligent inspection robot, and the robot assists electric power personnel to complete state monitoring of various instruments and digital reading of a pointer meter. This measure has accelerated the intelligent process of power equipment management and has promoted power institution work efficiency greatly. The main difference between all kinds of electric inspection robots is whether the robots are provided with lifting structures. Large-scale robot with elevation structure (like arm) can send the camera cloud platform to the eminence, often can gather the instrument image in the eminence cubical switchboard with good angle, is favorable to image recognition module to the reading discernment work of gauge. However, the addition of the lifting structure also increases the overall cost of the robot. And robots without lifting structures, although at lower manufacturing costs. However, the switch image at the high position collected by pitching of the pan-tilt camera often has certain distortion, which is not beneficial to the recognition of the meter reading.
At present, most pointer meter identification methods default or require a camera to acquire an image of a switch at a 'front view' angle, and a small inspection machine can only acquire the image at the 'bottom view' angle for a switch at a high place due to the fact that the small inspection machine has no lifting structure. Thus, such methods cannot deal with the problem of distortion of the dial caused by "looking up" the acquired image, and thus cannot accurately read the reading of the pointer.
In addition, current meter identification algorithms typically require a significant amount of dial marking work to be performed prior to use. Firstly, a clear meter picture is required to be collected as a template, the maximum and minimum scale directions and numbers of the dial are marked in the template, and some of the meter pictures even require to mark all scale lines for subsequent comparison of the template. Therefore, the earlier work of the method is more complicated.
Disclosure of Invention
The embodiment of the invention provides a pointer meter identification method and device and an electric power inspection robot.
The technical scheme of the embodiment of the invention is as follows:
a pointer-table identification method, the method comprising:
acquiring a first index mark image generated by shooting an index mark by the electric power inspection robot in an elevation angle view;
converting the first index mark image into a second index mark image of a front view angle;
determining a pointer region from the second pointer-marked image based on maximum-connected-amount region detection, and fitting a pointer straight line based on the pointer region;
in the remaining area of the second index image after the pointer area is removed, a plurality of scale value areas are determined based on pixel value distribution calculation, and a dial arc is fitted based on the plurality of scale value areas;
and determining the reading of the pointer table based on the intersection point of the pointer straight line and the circular arc of the dial.
In one embodiment, the converting the first index image into the second index image of the front view angle includes:
performing edge detection on the first pointer representation image to determine an edge of the pointer representation;
determining a homography matrix based on vertex coordinates of a quadrilateral surrounding the edge and vertex coordinates of the second index representation image;
and converting the first index marked image into a second index marked image of the front view angle based on the homography matrix.
In one embodiment, the determining a plurality of scale value regions based on pixel value distribution calculation includes:
converting the residual region into a binarized image;
scanning the binarized image by using a scanning window, and calculating the pixel value distribution of each scanning area;
adding a scanning area with a pixel point with a pixel value of 255 exceeding a preset threshold value into a scale value candidate area set;
screening the scale value candidate region set based on a preset reading range of the index table;
and removing the overlapped area from the screened scale value candidate area set by applying a non-maximum value suppression algorithm.
In one embodiment, the predetermined reading range is: the right lower corner of the residual area is taken as the origin, and the diameter isWherein w is the width of the remaining area, h is the height of the remaining area, and K1 is less than K2.
A pointer identification apparatus comprising:
the acquisition module is used for acquiring a first index mark image generated by the power inspection robot for shooting the index mark in an elevation angle view;
the conversion module is used for converting the first index mark image into a second index mark image with a front view angle;
the first fitting module is used for determining a pointer region from the second pointer mark image based on the maximum communication volume region detection and fitting out a pointer straight line based on the pointer region;
the second fitting module is used for determining a plurality of scale value areas based on pixel value distribution calculation in the remaining area of the second index mark image after the pointer area is removed, and fitting a dial arc based on the plurality of scale value areas;
and the determining module is used for determining the reading of the pointer table based on the intersection point of the pointer straight line and the circular arc of the dial.
In one embodiment, a conversion module performs edge detection on a first pointer representation image to determine an edge of the pointer representation; determining a homography matrix based on vertex coordinates of a quadrilateral surrounding the edge and vertex coordinates of the second index representation image; and converting the first index marked image into a second index marked image of the front view angle based on the homography matrix.
In one embodiment, a second fitting module is configured to convert the remaining region into a binarized image; scanning the binarized image by using a scanning window, and calculating the pixel value distribution of each scanning area; adding a scanning area with a pixel point with a pixel value of 255 exceeding a preset threshold value into a scale value candidate area set; screening the scale value candidate region set based on a preset reading range of the index table; and removing the overlapped area from the screened scale value candidate area set by applying a non-maximum value suppression algorithm.
A power inspection robot, comprising:
a camera assembly for capturing the index markers at an elevation view angle to obtain a first index marker image;
a processor for converting the first index image into a second index image of a front viewing angle; determining a pointer region from the second pointer-marked image based on maximum-connected-amount region detection, and fitting a pointer straight line based on the pointer region; in the remaining area of the second index image after the pointer area is removed, a plurality of scale value areas are determined based on pixel value distribution calculation, and a dial arc is fitted based on the plurality of scale value areas; and determining the reading of the pointer table based on the intersection point of the pointer straight line and the circular arc of the dial.
In one embodiment, the power inspection robot is a power inspection robot without a structure of a lifting camera assembly.
A computer readable storage medium having stored thereon a computer program which when executed by a processor implements the pointer representation identification method of any one of the above.
As can be seen from the above technical solution, in the embodiment of the present invention: acquiring a first index mark image generated by shooting an index mark by the electric power inspection robot in an elevation angle view; converting the first index mark image into a second index mark image of a front view angle; determining a pointer region from the second pointer-marked image based on maximum-connected-amount region detection, and fitting a pointer straight line based on the pointer region; in the remaining area of the second index image after the pointer area is removed, a plurality of scale value areas are determined based on pixel value distribution calculation, and a dial arc is fitted based on the plurality of scale value areas; and determining the reading of the pointer table based on the intersection point of the pointer straight line and the circular arc of the dial. Therefore, the embodiment of the invention provides an automatic image correction algorithm, and provides a good precondition for pointer gauge identification.
In addition, the embodiment of the invention also provides a reading identification algorithm of the pointer meter, which obviously reduces manual labeling work and ensures reading identification precision.
Drawings
FIG. 1 is a flow chart of an exemplary pointer table identification method of the present invention.
FIG. 2 is a schematic diagram of the overall process of pointer table identification according to the present invention.
FIG. 3 is a schematic diagram illustrating the process of pointer table identification according to the present invention.
FIG. 4 is a schematic representation of the read range of the present invention.
Fig. 5 is an exemplary structure diagram of the pointer table identification apparatus of the present invention.
Fig. 6 is a structural diagram of the power inspection robot of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings, in order to make the objects, technical solutions and advantages of the present invention more apparent.
For simplicity and clarity of description, the following description sets forth aspects of the invention by describing several exemplary embodiments. Numerous details in the embodiments are provided solely to aid in the understanding of the invention. It will be apparent, however, that the embodiments of the invention may be practiced without limitation to these specific details. Some embodiments are not described in detail in order to avoid unnecessarily obscuring aspects of the present invention, but rather only to present a framework. Hereinafter, "comprising" means "including but not limited to", "according to … …" means "according to at least … …, but not limited to only … …". The term "a" or "an" is used herein to refer to a number of components, either one or more, or at least one, unless otherwise specified.
In the embodiment of the invention, the problem that the image acquired by the 'look-up' angle of the camera has distortion is considered, an automatic image correction algorithm is provided, and a good precondition is provided for pointer gauge identification. Moreover, the embodiment of the invention also provides a reading identification algorithm of the pointer meter, which greatly reduces manual labeling work and ensures reading identification precision.
FIG. 1 is a flow chart of an exemplary pointer table identification method of the present invention.
As shown in fig. 1, the method includes:
step 101: and acquiring a first index mark image generated by the power inspection robot shooting the index mark in an elevation angle view.
Step 102: and converting the first index image into a second index image with a front view angle.
Step 103: and determining a pointer region from the second pointer image based on the maximum communication volume region detection, and fitting a pointer straight line based on the pointer region.
Step 104: in the remaining area of the second index image after the pointer area is removed, a plurality of scale value areas are determined based on pixel value distribution calculation, and a dial arc is fitted based on the plurality of scale value areas;
step 105: and determining the reading of the pointer table based on the intersection point of the pointer straight line and the circular arc of the dial.
In one embodiment, converting the first index tab image to a second index tab image at the front viewing angle in step 102 includes: performing edge detection on the first pointer representation image to determine an edge of the pointer representation; determining a homography matrix based on vertex coordinates of a quadrilateral surrounding the edge and vertex coordinates of the second index representation image; and converting the first index marked image into a second index marked image of the front view angle based on the homography matrix.
In one embodiment, determining a plurality of scale value regions based on pixel value distribution calculations in step 103 includes: converting the residual region into a binarized image; scanning the binarized image by using a scanning window, and calculating the pixel value distribution of each scanning area; adding a scanning area with a pixel point with a pixel value of 255 exceeding a preset threshold value into a scale value candidate area set; screening the scale value candidate region set based on a preset reading range of the index table; and removing the overlapped area from the screened scale value candidate area set by applying a non-maximum value suppression algorithm.
Preferably, the predetermined reading range is: the right lower corner of the residual area is taken as the origin, and the diameter isWherein w is the width of the remaining area, h is the height of the remaining area, and K1 is less than K2. More preferably, K1 is 0.95; k2 is 1.05.
FIG. 2 is a schematic diagram of the overall process of pointer table identification according to the present invention.
A typical small power inspection robot lacks the structure to lift the camera assembly (typically including the camera and pan-tilt). The pointer meter processing procedure of the electric power inspection robot based on the structure without the lifting camera component is shown in fig. 2. The intelligent monitoring system mainly comprises a holder camera module, an instrument identification module, a meter reading identification module and a data storage module. The flow is as follows in the running process:
(1) The electric power inspection robot without the structure of the lifting camera component passes through the positioning system to come in front of the switch cabinet to be detected, and the cradle head camera module is called to collect images of the switch in the cabinet according to the set identification task.
(2) Registering the acquired image with the template image, and then identifying the state and the position of the acquired switch image by using a pre-trained instrument identification module.
(3) And if the identified switch is a pointer-type meter, calling a meter identification module, taking the meter image as input, interacting with a data storage module, and reading dial information of a pre-configured meter. After the recognition is finished, the result is finally stored in a database.
(4) If the identified switch is not a pointer-like meter, if no special processing is required, the identification result in (2) is directly stored in the database.
FIG. 3 is a schematic diagram illustrating the process of pointer table identification according to the present invention.
(1): the first step, an automatic deviation correcting process is executed:
because the small-sized power inspection robot generally has no lifting structure, if the meter to be collected is located at a higher position of the switch cabinet, the cradle head needs to maintain an elevation angle to ensure that the meter is collected. Therefore, the meter image acquired at the moment has certain distortion, and correction is needed to ensure accurate reading. The automatic deviation rectifying method provided by the embodiment of the invention is realized by deducing the homography matrix H AB The gauge is converted from an "elevation" view to a "elevation" view using homography.
Specifically: the step of deducing the homography matrix comprises: first, the gradient of the tabular image is calculated using the Canny edge detection algorithmAnd find the four outer frames of the meter by using Hough transform, thereby determining four vertexes (A 1 ,A 2 ,A 3 ,A 4 ). Subsequently, assuming a front view, gauge left upper corner vertex B 1 ＝A 1 Further, it was inferred that in the case of "front view", four vertices (B 1 ,B 2 ,B 3 ,B 4 ). Then according to the mapping relation between the four pairs of points, confirming the single mapping matrix H from A to B AB . Finally, coordinate mapping can be carried out on each point of the meter image according to the single mapping matrix, automatic correction is completed, and a meter image P under a 'front view' angle is obtained s 。
(2): secondly, executing the preprocessing process of the image:
in the image preprocessing, the corrected meter image P is first corrected S Conversion to a grey-scale map P G . Because the background of the gauge is in sharp contrast to the scale of the pointer, the background is typically white and the foreground, such as the pointer, is typically black. The embodiment of the invention can utilize expansion operation and median filtering to erase details such as a pointer, a scale and the like of the foreground from the image to obtain a background map p B . Then take P G -P B The absolute value of the meter is binarized, so that the scale, the pointer and the frame skeleton of the meter are completely extracted, and a skeleton map p is obtained M 。
(3): third, executing automatic cutting:
at this time, the hough transform is used again to find the corrected meter image P S Is then applied to the skeleton map p M Secondary cutting to cut out the content outside the frame to obtain an image P 'containing only the pointer and the pointer scale' M 。
(4) Fourth, executing meter pointer identification:
for images P 'containing only pointers and pointer scales' M It can be observed that the area occupied by the pointer is the largest, at P' M The location area for determining the pointer can be found by searching the maximum connection quantity within the range of (1). Performing straight line fitting on the pixel coordinates where the pointer is located to finally obtainTo the straight line l where the pointer is located.
(5) Fifth step, executing meter scale value identification:
here, the binarized image P 'can be scanned by a window scanning method using a scanning window W' M The remaining area after the pointer area is removed (pixel value is 0 or 255), scanning of a predetermined step size (for example, 10 pixels) is performed, and the pixel value distribution of each scanned area is calculated. When the pixel point with the pixel value of 255 in the window exceeds a certain threshold value, the scanning area is added to the candidate area set C of the scale value all 。
Subsequently, embodiments of the present invention pair candidate set C by the following rule all Screening is performed. Namely: the scale candidate should be within a predetermined reading range of the index mark, such as in image P' M The lower right corner is the origin and the diameter isWherein w, h is the width and height of the remaining regions. Finally, candidate set C is excluded from this range by a non-maximum suppression algorithm (NMS) all To obtain a set C of scale values in the dial num . For example, FIG. 4 is a schematic representation of the read range of the present invention. The circular area between the circular arc 31 and the circular arc 32 is the preset reading range. Wherein the diameter of the arc 31 is +.>The diameter of the circular arc 32 is
(6) Sixth, dial arc fitting is performed:
taking C num The central point coordinates of each window can be fitted with the circular arc of the dial. The intersection point of l and arc is obtained, and the reading of the meter can be finally determined according to the position of the intersection point. At this time, only the maximum scale value, the minimum scale value and the number of scale values of the meter are recorded in advance, and the intersection point can be usedThe location at which it is located determines the gauge reading.
Therefore, the embodiment of the invention provides a method for automatically identifying the pointer meter reading by a small robot with a structure without a lifting camera component. Even if the camera collects the meter image at a certain elevation angle, the method can realize automatic correction of the dial according to the extracted meter frame shape, thereby ensuring the accuracy of meter identification. Compared with the prior art, the embodiment of the invention has the advantages that the manual marking quantity is small, and only the maximum and minimum scale values of the meter and the number of the scale values of the meter are required to be recorded.
Fig. 5 is an exemplary structure diagram of the pointer table identification apparatus of the present invention.
As shown in fig. 5, the pointer identification apparatus includes:
the acquisition module is used for acquiring a first index mark image generated by the power inspection robot for shooting the index mark in an elevation angle view;
the conversion module is used for converting the first index mark image into a second index mark image with a front view angle;
the first fitting module is used for determining a pointer region from the second pointer mark image based on the maximum communication volume region detection and fitting out a pointer straight line based on the pointer region;
the second fitting module is used for determining a plurality of scale value areas based on pixel value distribution calculation in the remaining area of the second index mark image after the pointer area is removed, and fitting a dial arc based on the plurality of scale value areas;
and the determining module is used for determining the reading of the pointer table based on the intersection point of the pointer straight line and the circular arc of the dial.
In one embodiment, a conversion module performs edge detection on a first pointer representation image to determine an edge of the pointer representation; determining a homography matrix based on vertex coordinates of a quadrilateral surrounding the edge and vertex coordinates of the second index representation image; and converting the first index marked image into a second index marked image of the front view angle based on the homography matrix.
In one embodiment, a second fitting module is configured to convert the remaining region into a binarized image; scanning the binarized image by using a scanning window, and calculating the pixel value distribution of each scanning area; adding a scanning area with a pixel point with a pixel value of 255 exceeding a preset threshold value into a scale value candidate area set; screening the scale value candidate region set based on a preset reading range of the index table; and removing the overlapped area from the screened scale value candidate area set by applying a non-maximum value suppression algorithm.
Based on the above description, fig. 6 is a structural diagram of the power inspection robot of the present invention. The power inspection robot includes:
the camera component is used for shooting the index marks in an elevation angle view to obtain a first index mark image;
a processor for converting the first index image into a second index image of a front viewing angle; determining a pointer region from the second pointer-marked image based on maximum-connected-amount region detection, and fitting a pointer straight line based on the pointer region; in the remaining area of the second index image after the pointer area is removed, a plurality of scale value areas are determined based on pixel value distribution calculation, and a dial arc is fitted based on the plurality of scale value areas; and determining the reading of the pointer table based on the intersection point of the pointer straight line and the circular arc of the dial.
In one embodiment, the power inspection robot is a power inspection robot without a structure of a lifting camera assembly.
In summary, in the embodiments of the present invention: acquiring a first index mark image generated by shooting an index mark by the electric power inspection robot in an elevation angle view; converting the first index mark image into a second index mark image of a front view angle; determining a pointer region from the second pointer-marked image based on maximum-connected-amount region detection, and fitting a pointer straight line based on the pointer region; in the remaining area of the second index image after the pointer area is removed, a plurality of scale value areas are determined based on pixel value distribution calculation, and a dial arc is fitted based on the plurality of scale value areas; and determining the reading of the pointer table based on the intersection point of the pointer straight line and the circular arc of the dial. Therefore, the embodiment of the invention provides an automatic image correction algorithm, and provides a good precondition for pointer gauge identification. Moreover, the embodiment of the invention also provides a reading identification algorithm of the pointer meter, which greatly reduces manual labeling work and ensures reading identification precision.
The embodiment of the invention also provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements each process implemented in the above embodiments of the invention, and can achieve the same technical effects, so that repetition is avoided, and no further description is provided herein. Wherein the computer readable storage medium is selected from Read-Only Memory (ROM), random access Memory (Random Access Memory, RAM), magnetic disk or optical disk. From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present invention.
The embodiments of the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and many forms may be made by those having ordinary skill in the art without departing from the spirit of the present invention and the scope of the claims, which are to be protected by the present invention.
Claims (6)
1. A pointer-table identification method, the method comprising:
acquiring a first index table image generated by shooting an index table by an electric power inspection robot without a lifting camera component in an elevation angle view;
converting the first index mark image into a second index mark image of a front view angle;
determining a pointer region from the second pointer-marked image based on maximum-connected-amount region detection, and fitting a pointer straight line based on the pointer region;
in the remaining area of the second index image after the pointer area is removed, a plurality of scale value areas are determined based on pixel value distribution calculation, and a dial arc is fitted based on the plurality of scale value areas;
determining a reading of the pointer register based on an intersection of the pointer line and the circular arc of the dial;
the determining a plurality of scale value regions based on pixel value distribution calculation includes: converting the residual region into a binarized image; scanning the binarized image by using a scanning window, and calculating the pixel value distribution of each scanning area; adding a scanning area with the number of pixel points with the pixel value of 255 exceeding a preset threshold value into a scale value candidate area set; screening the scale value candidate region set based on a preset reading range of the index table; removing an overlapping region from the screened scale value candidate region set by applying a non-maximum value suppression algorithm;
the predetermined reading range is: the right lower corner of the residual area is taken as the origin, and the diameter is Wherein w is the width of the remaining area, h is the height of the remaining area, and K1 is less than K2.
2. The method of claim 1, wherein the step of determining the position of the substrate comprises,
converting the first index tab image into a second index tab image at the front viewing angle includes:
performing edge detection on a first pointer representation image to determine an edge of the pointer representation;
determining a homography matrix based on vertex coordinates of a quadrilateral surrounding the edge and vertex coordinates of the second index representation image;
and converting the first index marked image into a second index marked image of the front view angle based on the homography matrix.
3. A pointer identification apparatus, comprising:
the acquisition module is used for acquiring a first index mark image generated by the power inspection robot without the lifting camera component for shooting the index mark in an elevation angle view;
the conversion module is used for converting the first index mark image into a second index mark image with a front view angle;
the first fitting module is used for determining a pointer region from the second pointer mark image based on the maximum communication volume region detection and fitting out a pointer straight line based on the pointer region;
the second fitting module is used for determining a plurality of scale value areas based on pixel value distribution calculation in the remaining area of the second index mark image after the pointer area is removed, and fitting a dial arc based on the plurality of scale value areas;
the determining module is used for determining the reading of the pointer table based on the intersection point of the pointer straight line and the circular arc of the dial;
the second fitting module is used for converting the residual area into a binarized image; scanning the binarized image by using a scanning window, and calculating the pixel value distribution of each scanning area; adding a scanning area with the number of pixel points with the pixel value of 255 exceeding a preset threshold value into a scale value candidate area set; screening the scale value candidate region set based on a preset reading range of the index table; removing an overlapping region from the screened scale value candidate region set by applying a non-maximum value suppression algorithm;
the predetermined reading range is: the right lower corner of the residual area is taken as the origin, and the diameter is Wherein w is the width of the remaining area, h is the height of the remaining area, and K1 is less than K2.
4. The apparatus of claim 3, wherein the device comprises a plurality of sensors,
the conversion module is used for performing edge detection on the first index mark image to determine the edge of the index mark; determining a homography matrix based on vertex coordinates of a quadrilateral surrounding the edge and vertex coordinates of the second index representation image; and converting the first index marked image into a second index marked image of the front view angle based on the homography matrix.
5. The utility model provides a power inspection robot which characterized in that includes:
a camera assembly for capturing the index markers at an elevation view angle to obtain a first index marker image;
a processor for converting the first index image into a second index image of a front viewing angle; determining a pointer region from the second pointer-marked image based on maximum-connected-amount region detection, and fitting a pointer straight line based on the pointer region; in the remaining area of the second index image after the pointer area is removed, a plurality of scale value areas are determined based on pixel value distribution calculation, and a dial arc is fitted based on the plurality of scale value areas; determining a reading of the index mark based on an intersection of the pointer line and the dial arc;
the determining a plurality of scale value regions based on pixel value distribution calculation includes: converting the residual region into a binarized image; scanning the binarized image by using a scanning window, and calculating the pixel value distribution of each scanning area; adding a scanning area with the number of pixel points with the pixel value of 255 exceeding a preset threshold value into a scale value candidate area set; screening the scale value candidate region set based on a preset reading range of the index table; removing an overlapping region from the screened scale value candidate region set by applying a non-maximum value suppression algorithm;
the predetermined reading range is: the right lower corner of the residual area is taken as the origin, and the diameter is Wherein w is the width of the remaining region, h is the height of the remaining region, and K1 is less than K2;
the electric power inspection robot is of a structure without a lifting camera component.
6. A computer readable storage medium, characterized in that the computer readable storage medium stores thereon a computer program, which when executed by a processor implements the pointer table identification method according to any one of claims 1 to 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011388525.9A CN112507838B (en) | 2020-12-02 | 2020-12-02 | Pointer meter identification method and device and electric power inspection robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011388525.9A CN112507838B (en) | 2020-12-02 | 2020-12-02 | Pointer meter identification method and device and electric power inspection robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112507838A CN112507838A (en) | 2021-03-16 |
| CN112507838B true CN112507838B (en) | 2024-03-22 |
Family
ID=74968511
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011388525.9A Active CN112507838B (en) | 2020-12-02 | 2020-12-02 | Pointer meter identification method and device and electric power inspection robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112507838B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114549282B (en) * | 2022-01-11 | 2023-12-12 | 深圳昱拓智能有限公司 | Method and system for realizing multi-meter reading based on affine transformation |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3998215B1 (en) * | 2007-03-29 | 2007-10-24 | 国立大学法人山口大学 | Image processing apparatus, image processing method, and image processing program |
-
2020
- 2020-12-02 CN CN202011388525.9A patent/CN112507838B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3998215B1 (en) * | 2007-03-29 | 2007-10-24 | 国立大学法人山口大学 | Image processing apparatus, image processing method, and image processing program |
Non-Patent Citations (1)
| Title |
|---|
| 多个指针式仪表的识别;赵健 等;传感器与微系统;第39卷(第2期);第61-64页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112507838A (en) | 2021-03-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110599548A (en) | Camera calibration method and device, camera and computer readable storage medium | |
| CN108648169A (en) | The method and device of high voltage power transmission tower defects of insulator automatic identification | |
| CN111738229B (en) | Automatic reading method for scale of pointer dial | |
| CN114549835A (en) | Pointer instrument correction identification method and device based on deep learning | |
| CN113469178B (en) | Power meter identification method based on deep learning | |
| CN114511718B (en) | Intelligent management method and system for materials for building construction | |
| CN112560837A (en) | Reading method, device, equipment and storage medium of pointer instrument | |
| CN113688817A (en) | Instrument identification method and system for automatic inspection | |
| CN112325809A (en) | Method for detecting flatness of flange | |
| CN112507838B (en) | Pointer meter identification method and device and electric power inspection robot | |
| CN115601743A (en) | Power distribution room pointer type instrument registration automatic reading identification method | |
| CN113592839B (en) | Distribution network line typical defect diagnosis method and system based on improved fast RCNN | |
| CN113705350A (en) | Pointer instrument reading identification method and device for transformer substation, medium and electronic equipment | |
| CN117434568A (en) | Intelligent positioning system based on remote sensing satellite | |
| CN112419402A (en) | Positioning method and system based on multispectral image and laser point cloud | |
| CN114926417B (en) | Microscopic imaging detection method and system for weak scratches on surface of high-density flexible FPC | |
| Tu et al. | 2D in situ method for measuring plant leaf area with camera correction and background color calibration | |
| CN114037993A (en) | Substation pointer instrument reading method and device, storage medium and electronic equipment | |
| CN113792616A (en) | Remote meter reading system based on edge calculation and working method thereof | |
| CN108509845B (en) | Visual inspection method for transformer substation instrument equipment inspection based on feature fitting | |
| CN112288049A (en) | Pointer meter reading method based on template matching and limitation of connected domain pointer positioning | |
| CN113191336A (en) | Electric power hidden danger identification method and system based on image identification | |
| CN112907490A (en) | Pixel point extraction method, target information acquisition method and push bench control system | |
| CN116189192B (en) | Intelligent reading identification method and system for pointer instrument | |
| CN112446895A (en) | Checkerboard angular point automatic extraction method, system, equipment and medium |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 215000 room 2104, block a, Suzhou Central Plaza, No. 58, Suxiu Road, Suzhou Industrial Park, Jiangsu Province Applicant after: SUZHOU TOUCHAIR TECHNOLOGY Co.,Ltd. Address before: Room 801, building 6, creative industry park, 328 Xinghu street, Suzhou Industrial Park, Jiangsu Province 215000 Applicant before: SUZHOU TOUCHAIR TECHNOLOGY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |