阅读说明：本技术 一种表计读数识别方法、装置、终端设备及存储介质 (Meter reading identification method and device, terminal equipment and storage medium ) 是由 田治仁 张贵峰 李锐海 廖永力 张巍 龚博 王俊锞 黄增浩 吴新桥 朱登杰 何锦 于 2019-09-25 设计创作，主要内容包括：本发明公开了一种表计读数识别方法、装置、终端设备及存储介质,包括：获取变电站的表计图像,并将所述表计图像组织成表计图像数据集；采用预设的共显著性检测算法,提取所述表计图像数据集的所有图像中表针和表盘的共显著性,得到共显著性图像；对所述共显著性图像进行表针直线识别,确定表针位置；根据所述表针位置,采用预先建立的表计读数模型进行计算,得到表针读数,能有效解决现有技术不同的指针式仪表存在于不同的环境背景干扰导致难以提取表计指针特征的问题,能有效从背景因素上避免变电站其他设备以及环境的干扰,能有效提高表针特征提取的准确度。(The invention discloses a meter reading identification method, a meter reading identification device, a terminal device and a storage medium, wherein the meter reading identification method comprises the following steps: acquiring meter images of a transformer substation, and organizing the meter images into a meter image data set; extracting the common saliency of a pointer and a dial plate in all images of the meter image data set by adopting a preset common saliency detection algorithm to obtain a common saliency image; performing pointer straight line identification on the common saliency image to determine a pointer position; according to the position of the pointer, a pointer reading model which is established in advance is adopted for calculation to obtain pointer reading, the problem that pointer characteristics of the pointer are difficult to extract due to the fact that different pointer type instruments in the prior art exist in different environment background interferences can be effectively solved, interferences of other equipment and environments of a transformer substation can be effectively avoided from background factors, and the accuracy of pointer characteristic extraction can be effectively improved.)

1. A meter reading identification method is characterized by comprising the following steps:

acquiring meter images of a transformer substation, and organizing the meter images into a meter image data set;

extracting the common saliency of a pointer and a dial plate in all images of the meter image data set by adopting a preset common saliency detection algorithm to obtain a common saliency image;

performing pointer straight line identification on the common saliency image to determine a pointer position;

and calculating by adopting a pre-established meter reading model according to the position of the meter pointer to obtain the meter pointer reading.

2. The meter reading identification method of claim 1, wherein the extracting the co-saliency of the hands and the dial in all the images of the meter image dataset by using a preset co-saliency detection algorithm to obtain a co-saliency image, further comprises:

obtaining a common saliency information cluster of the tabulated image data set according to formula (1):

wherein the content of the first and second substances,

obtaining the co-saliency image according to the co-saliency information cluster by adopting a formula (2):

wherein, ω is^d(k) Weight, var (q), is indexed for the consistency of the d-th cluster^k) Is the variance of the co-significant information cluster.

3. The meter reading identification method of claim 1, wherein the performing pointer straight line identification on the co-saliency image to determine a pointer position specifically comprises:

carrying out binarization processing on the common saliency image to obtain a binarization image;

detecting all straight lines and distances thereof in the binary image by adopting a preset hough algorithm;

and screening the straight line with the maximum distance as a pointer straight line according to the distance of the straight line, and obtaining the pointer position corresponding to the pointer straight line.

4. The meter-reading identification method of claim 1, wherein the meter-reading model is constructed by:

according to the pre-acquired dial plate range scales of the meter, a polar coordinate system is established by taking a dial plate central point as an original point and symmetrically establishing a maximum range scale and a minimum range scale of the dial plate range scales with respect to a y axis to obtain a meter reading model;

and determining the position information corresponding to the pointer position and the dial range scale according to the meter reading model.

5. The meter reading identification method according to claim 4, wherein the obtaining of the meter reading by calculating according to the meter reading position by using a pre-established meter reading model specifically comprises:

calculating the pointer reading according to equation (3):

Result＝(θ/ψ)(V_max-V_min)+V_min(3)

wherein Result is the pointer reading, V_maxIs the maximum range scale value, V, of the dial range scale_minAnd phi is an included angle between the connecting line of the original point and the initial scale mark and the connecting line of the original point and the ending scale mark of the dial range scale.

6. The meter reading identification method of claim 1, wherein the obtaining of the meter image of the substation and the organizing of the meter image into a meter image dataset specifically comprises:

acquiring the meter image of the transformer substation, and setting the image size of the meter image;

and constructing the meter image data set according to the set meter image.

7. A meter reading identification device, comprising:

the meter image data set construction module is used for acquiring meter images of the transformer substation and organizing the meter images into a meter image data set;

the common significance detection module is used for extracting the common significance of the hands and the dials in all the images of the meter image data set by adopting a preset common significance detection algorithm to obtain a common significance image;

the pointer position determining module is used for carrying out pointer linear identification on the common saliency image and determining the pointer position;

and the meter reading module is used for calculating by adopting a pre-established meter reading model according to the meter position to obtain the meter reading.

8. A meter reading identification terminal device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, the processor implementing the meter reading identification method of any of claims 1 to 6 when executing the computer program.

9. A computer-readable storage medium, comprising a stored computer program, wherein the computer program, when executed, controls an apparatus in which the computer-readable storage medium is located to perform the meter reading identification method according to any one of claims 1 to 6.

Technical Field

The invention relates to the technical field of image processing, in particular to a meter reading identification method and device, terminal equipment and a storage medium.

Background

A substation of an electric power system has a number of devices equipped with pointer meters, such as: barometer, thermometer, oil thermometer, lightning arrester, etc. The pointer instrument is widely applied to a power system due to the characteristics of good practicability, high reliability, low cost, good anti-electromagnetic interference capability and the like, and the purpose of monitoring equipment is achieved by identifying the readings of the instrument. With the pilot construction of the intelligent substation and the development and popularization of the inspection robot, the recording work of instrument data is improved towards the direction of automatic operation, and operators still need to participate in verification to confirm instrument state images acquired by the robot. Although the mode reduces the labor intensity of operators, the subjective judgment of field personnel is greatly dependent, and the promotion of intensive operation management modes such as unattended operation of a transformer substation is influenced. Under the background that the traditional manual inspection is time-consuming, labor-consuming and inefficient, the power inspection robot is produced by the way. The inspection robot replaces manual inspection, and reading of the readings of the instruments under different conditions is one of essential core functions.

Most of the instruments of the transformer substation are arranged outdoors, and instrument images collected by the inspection robot are generally influenced by the environment. For the existing recognition algorithm, the basic position and the area range of the instrument dial in the image are determined by utilizing the shape characteristics of the instrument dial through template matching or ellipse fitting. Although the algorithm has certain real-time performance and robustness, the algorithm is not suitable for transformer substation instrument images acquired by the inspection robot. This is because the structure of the substation equipment is complex, and different pointer type meters exist in different environmental background interferences, so that the meter pointer characteristics are difficult to extract.

Disclosure of Invention

The embodiment of the invention provides a meter reading identification method, a meter reading identification device, terminal equipment and a storage medium, which can effectively solve the problem that meter pointer characteristics are difficult to extract due to the fact that different pointer type meters exist in different environment background interferences in the prior art, can effectively avoid interferences of other equipment and environments of a transformer substation from background factors, and can effectively improve the accuracy of meter pointer characteristic extraction.

An embodiment of the present invention provides a meter reading identification method, including:

acquiring meter images of a transformer substation, and organizing the meter images into a meter image data set;

extracting the common saliency of a pointer and a dial plate in all images of the meter image data set by adopting a preset common saliency detection algorithm to obtain a common saliency image;

performing pointer straight line identification on the common saliency image to determine a pointer position;

and calculating by adopting a pre-established meter reading model according to the position of the meter pointer to obtain the meter pointer reading.

As an improvement of the above scheme, the extracting, by using a preset co-saliency detection algorithm, the co-saliency of the hands and the dials in all the images of the meter image dataset to obtain a co-saliency image further includes:

obtaining a common saliency information cluster of the tabulated image data set according to formula (1):

wherein the content of the first and second substances,

for the co-significant information cluster, C^kIs the kth cluster, n^kIs the total number of pixel points of the kth cluster, j is the total number of images of the meter image data set,

is the ith pixel point on the jth image, N_jFor the total chrominance of a single image,

mapping the ith pixel point on the jth image to a cluster first level;

obtaining the co-saliency image according to the co-saliency information cluster by adopting a formula (2):

wherein, ω is^d(k) Weight, var (q), is indexed for the consistency of the d-th cluster^k) Is the variance of the co-significant information cluster.

As an improvement of the above scheme, the performing pointer straight line recognition on the co-saliency image to determine the pointer position specifically includes:

carrying out binarization processing on the common saliency image to obtain a binarization image;

detecting all straight lines and distances thereof in the binary image by adopting a preset hough algorithm;

and screening the straight line with the maximum distance as a pointer straight line according to the distance of the straight line, and obtaining the pointer position corresponding to the pointer straight line.

As an improvement of the scheme, the meter reading model is constructed by the following steps:

according to the pre-acquired dial plate range scales of the meter, a polar coordinate system is established by taking a dial plate central point as an original point and symmetrically establishing a maximum range scale and a minimum range scale of the dial plate range scales with respect to a y axis to obtain a meter reading model;

and determining the position information corresponding to the pointer position and the dial range scale according to the meter reading model.

As an improvement of the above scheme, the calculating, according to the position of the pointer, by using a pre-established meter reading model to obtain the pointer reading specifically includes:

calculating the pointer reading according to equation (3):

Result＝(θ/ψ)(V_max-V_min)+V_min(3)

wherein Result is the pointer reading, V_maxIs the maximum range scale value, V, of the dial range scale_minAnd phi is an included angle between the connecting line of the original point and the initial scale mark and the connecting line of the original point and the ending scale mark of the dial range scale.

As an improvement of the above scheme, the acquiring a meter image of a substation and organizing the meter image into a meter image dataset specifically includes:

acquiring the meter image of the transformer substation, and setting the image size of the meter image;

and constructing the meter image data set according to the set meter image.

Compared with the prior art, the meter reading identification method disclosed by the embodiment of the invention comprises the steps of acquiring a meter image of a transformer substation, and organizing the meter image into a meter image data set; extracting the common saliency of a pointer and a dial plate in all images of the meter image data set by adopting a preset common saliency detection algorithm to obtain a common saliency image; performing pointer straight line identification on the common saliency image to determine a pointer position; and calculating by adopting a pre-established meter reading model according to the position of the meter pointer to obtain the meter pointer reading. The method adopts a common significance algorithm, can effectively solve the problem that the pointer characteristics of the meter are difficult to extract due to the fact that different pointer instruments exist in different environment background interferences in the prior art, can effectively avoid the interferences of other equipment and environments of a transformer substation from background factors, can effectively improve the accuracy and precision of the pointer characteristic extraction, and has high applicability.

Another embodiment of the present invention correspondingly provides a meter reading identification device, including:

the meter image data set construction module is used for acquiring meter images of the transformer substation and organizing the meter images into a meter image data set;

the common significance detection module is used for extracting the common significance of the hands and the dials in all the images of the meter image data set by adopting a preset common significance detection algorithm to obtain a common significance image;

the pointer position determining module is used for carrying out pointer linear identification on the common saliency image and determining the pointer position;

and the meter reading module is used for calculating by adopting a pre-established meter reading model according to the meter position to obtain the meter reading.

Compared with the prior art, the meter reading identification device disclosed by the embodiment of the invention is used for acquiring the meter image of the transformer substation through the meter image dataset construction module and organizing the meter image into the meter image dataset; the common significance detection module is used for extracting the common significance of the hands and the dials in all the images of the meter image data set by adopting a preset common significance detection algorithm to obtain a common significance image; the pointer position determining module is used for carrying out pointer linear identification on the common saliency image and determining the pointer position; and the meter reading module is used for calculating by adopting a pre-established meter reading model according to the meter position to obtain the meter reading. The device executes the common significance algorithm, can effectively solve the problem that the pointer characteristics of the meter are difficult to extract due to the fact that different pointer instruments exist in different environment background interferences in the prior art, can effectively avoid interferences of other equipment and environments of a transformer substation from background factors, can effectively improve the accuracy and precision of the pointer characteristic extraction, and has high applicability.

Another embodiment of the present invention provides a meter reading identification terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, where the processor implements the meter reading identification method according to the above embodiment of the present invention when executing the computer program.

Another embodiment of the present invention provides a storage medium, where the computer-readable storage medium includes a stored computer program, where when the computer program runs, the apparatus where the computer-readable storage medium is located is controlled to execute the meter reading identification method described in the above embodiment of the present invention.

Drawings

Fig. 1 is a schematic flowchart of a meter reading identification method according to an embodiment of the present invention;

fig. 2 is a schematic structural diagram of a meter reading identification device according to a second embodiment of the present invention.

Detailed Description

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 is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.