An electric power operation site safety monitoring object tracking method, system, device and medium

Abstract

The application discloses a power operation site safety supervision object tracking method, system, device and medium, obtains a plurality of site pictures corresponding to a power operation site, and constructs an initial background picture and an initial work picture corresponding to the power operation site according to the site pictures. Based on the matching degree corresponding to the initial background picture and the initial work picture, a target background picture and a target work picture corresponding to the power operation site are determined. The target background picture and the target work picture are subjected to difference fractal dimension calculation to determine an initial safety supervision object corresponding to the power operation site. Finally, the target safety supervision object corresponding to the power operation site is determined according to picture data corresponding to the site pictures and the initial safety supervision object. Through difference fractal dimension calculation, the fractal dimension features of the safety supervision object are extracted, the general requirements and individual special feature expressions of the safety supervision object are taken into account, and the continuous tracking under different monitoring angles and performance forms can be met.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power operation site safety supervision object tracking, and particularly relates to a power operation site safety supervision object tracking method, system, device and medium. BACKGROUND

[0002] In the process of power production, whether the power operation site follows the safety regulations will directly affect the safety of production. At present, the safety control requirements of power operation site of each power grid company are increasing.

[0003] Through the management of the safety supervision objects of the power operation site, the safety control of the power operation site is realized. Through the identification and tracking of the people, construction equipment and equipment in the camera video and image of the power operation site, the basis for the management focus of the safety supervision objects of the power operation site is provided, so as to determine whether the operation of the power operation site follows the safety regulations.

[0004] The existing power operation site safety supervision object tracking method is to set multiple cameras to collect pictures of safety supervision objects from different angles, and manually track and analyze the safety supervision objects in the pictures to determine whether the power operation site is safe. The safety supervision object tracking method adopts manual safety supervision object tracking analysis, which has large workload and low accuracy of analysis results. SUMMARY

[0005] The present application provides a power operation site safety supervision object tracking method, system, device and medium, which solves the technical problem of large workload and low accuracy of analysis results of the existing power operation site safety supervision object tracking method.

[0006] The power operation site safety supervision object tracking method provided by the present application comprises the following steps:

[0007] Obtaining multiple site pictures corresponding to the power operation site, and constructing an initial background picture and an initial working picture corresponding to the power operation site according to the site pictures;

[0008] According to the matching degree of the initial background picture and the initial working picture, determining a target background picture and a target working picture corresponding to the power operation site;

[0009] Carrying out difference fractal dimension calculation on the target background picture and the target working picture to determine an initial safety supervision object corresponding to the power operation site;

[0010] According to the picture data corresponding to the site pictures and the initial safety supervision object, determining a target safety supervision object corresponding to the power operation site.

[0011] Optionally, the step of constructing the initial background picture and the initial work picture corresponding to the power operation site according to the site picture comprises:

[0012] Respectively acquiring the work state corresponding to the site picture;

[0013] Classifying the site pictures according to the work state to generate a first site picture set and a second site picture set;

[0014] Performing picture splicing on the pictures in the first site picture set to generate the initial background picture corresponding to the power operation site;

[0015] Performing picture splicing on the pictures in the second site picture set to generate the initial work picture corresponding to the power operation site.

[0016] Optionally, the step of determining the target background picture and the target work picture corresponding to the power operation site according to the matching degree corresponding to the initial background picture and the initial work picture comprises:

[0017] Acquiring a first initial reference line set corresponding to the initial background picture and a second initial reference line set corresponding to the initial work picture;

[0018] Respectively calculating the cosine value corresponding to each reference line in the first initial reference line set and the second initial reference line set;

[0019] If the cosine value meets a preset threshold value, the initial background picture and the initial work picture are taken as the target background picture and the target work picture corresponding to the power operation site.

[0020] Optionally, the first initial reference line set comprises a plurality of first initial reference lines; the second initial reference line set comprises a plurality of second initial reference lines; the cosine value comprises a first cosine value and a second cosine value; the step of respectively calculating the cosine value corresponding to each reference line in the first initial reference line set and the second initial reference line set comprises:

[0021] Respectively selecting a first matching line corresponding to the first initial reference line in the initial work picture;

[0022] According to the gray value of the pixel point, respectively constructing a first color system vector and a second color system vector corresponding to the first initial reference line and the first matching line;

[0023] Calculating the cosine value corresponding to the first color system vector and the second color system vector to generate the first cosine value;

[0024] Respectively selecting a second matching line corresponding to the second initial reference line in the initial background picture;

[0025] constructing a third color system vector and a fourth color system vector corresponding to the second initial reference line and the second matching line respectively according to the gray value of the pixel point;

[0026] calculating a second cosine value corresponding to the third color system vector and the fourth color system vector.

[0027] Optionally, the step of performing difference fractal dimension calculation on the target background picture and the target work picture to determine the initial safety supervision object corresponding to the power operation site comprises:

[0028] performing feature comparison on the target background picture and the target work picture to generate a difference picture corresponding to the power operation site;

[0029] selecting a plurality of difference points and a plurality of islands in the difference picture according to a preset selection standard;

[0030] calculating a difference singular coefficient corresponding to the difference point respectively, and determining a plurality of difference singular spectra corresponding to the island in combination with a preset difference singular spectrum formula;

[0031] taking the maximum value in the difference singular spectrum as a first detection operator corresponding to the island;

[0032] calculating the sum value of all the difference singular spectra to generate a second detection operator corresponding to the island;

[0033] integrating the difference singular spectra respectively to generate a third detection operator corresponding to the island;

[0034] comparing the first detection operator, the second detection operator and the third detection operator with a corresponding standard safety supervision defect feature value respectively to generate a feature comparison result corresponding to the island;

[0035] determining the initial safety supervision object corresponding to the power operation site according to the feature comparison result.

[0036] Optionally, the step of calculating a difference singular coefficient corresponding to the difference point respectively, and determining a plurality of difference singular spectra corresponding to the island in combination with a preset difference singular spectrum formula comprises:

[0037] selecting a plurality of square regions with the difference point as the center according to a preset size;

[0038] substituting the maximum difference value in the square region into a preset difference singular coefficient formula to generate a difference singular coefficient corresponding to the square region;

[0039] The preset difference singular coefficient formula is:

[0040]

[0041] wherein x(m,n) is a difference singular coefficient; epsilon i is the side length of the square region; mu i (m,n) is the maximum difference value in the square region; (m,n) is the coordinate of the difference point;

[0042] The difference singular coefficient maximum value and the difference singular coefficient minimum value corresponding to the island are used to construct a singular coefficient interval corresponding to the island.

[0043] The singular coefficient interval is divided into a plurality of singular coefficient local intervals according to a standard spectrum segmentation number;

[0044] The singular value number corresponding to the singular coefficient local interval and the corresponding difference singular coefficient are respectively substituted into a preset difference singular spectrum formula to generate a plurality of difference singular spectra corresponding to the island.

[0045] The preset difference singular spectrum formula is:

[0046]

[0047] wherein P(alpha i ) is a difference singular spectrum; is the singular value number falling in the singular coefficient local interval; x α (m,n) is the difference singular coefficient of the corresponding difference point under the alpha measure; (m,n) is the coordinate of the difference point; alpha i represents the left boundary value of the i-th singular coefficient local interval.

[0048] Optionally, the step of determining the target safety supervision object corresponding to the power operation site according to the picture data corresponding to the field picture and the initial safety supervision object comprises:

[0049] Obtaining parameter data dimensions and sampling data corresponding to the initial safety supervision object in the picture data;

[0050] Using the parameter data dimensions and the sampling data to construct a symbolic dynamics expression corresponding to the initial safety supervision object;

[0051] The symbolic dynamics expression is used as the target safety supervision object corresponding to the power operation site.

[0052] The application further provides a power operation site safety supervision object tracking system, comprising:

[0053] The initial background image and initial working image construction module is used to acquire multiple on-site images corresponding to the power operation site, and construct the initial background image and initial working image corresponding to the power operation site based on the on-site images;

[0054] The target background image and target work image determination module is used to determine the target background image and target work image corresponding to the power operation site based on the matching degree between the initial background image and the initial work image.

[0055] The initial safety supervision object determination module is used to perform differential fractal dimension calculation on the target background image and the target working image to determine the initial safety supervision object corresponding to the power operation site.

[0056] The target safety supervision object determination module is used to determine the target safety supervision object corresponding to the power operation site based on the image data corresponding to the on-site image and the initial safety supervision object.

[0057] The present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of any of the above-described methods for tracking objects under safety supervision at power operation sites.

[0058] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed, implements any of the above-described methods for tracking safety monitoring objects at power operation sites.

[0059] As can be seen from the above technical solutions, the present invention has the following advantages:

[0060] This invention acquires multiple images of a power operation site and constructs initial background and working images based on these images. Based on the matching degree between the initial background and working images, target background and working images are determined for the power operation site. The initial safety monitoring object for the power operation site is determined by calculating the difference fractal dimension of the target background and working images. Finally, the target safety monitoring object for the power operation site is determined based on the image data corresponding to the site images and the initial safety monitoring object. This invention solves the technical problems of existing methods for tracking safety monitoring objects at power operation sites, which involve large workloads and low accuracy due to manual tracking and analysis. By calculating the difference fractal dimension, the fractal dimension features of the safety monitoring object are extracted, taking into account both the general requirements of the safety monitoring object and the expression of personalized special features, thus meeting the needs of continuous tracking under different monitoring angles and manifestations. Attached Figure Description

[0061] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0062] Figure 1 A step flow chart of a power operation site safety monitoring object tracking method provided for the first embodiment of the present application.

[0063] Figure 2 A step flow chart of a power operation site safety monitoring object tracking method provided for the second embodiment of the present application.

[0064] Figure 3 A structural block diagram of a power operation site safety monitoring object tracking system provided for the third embodiment of the present application. DETAILED DESCRIPTION

[0065] The embodiments of the present application provide a power operation site safety monitoring object tracking method, system, device and medium, which are used to solve the technical problems of large workload of manual safety monitoring object tracking analysis and low accuracy of analysis results in the existing power operation site safety monitoring object tracking mode.

[0066] In order to make the purposes, features and advantages of the present application more obvious and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the following described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0067] Please refer to Figure 1 , Figure 1 A step flow chart of a power operation site safety monitoring object tracking method provided for the first embodiment of the present application.

[0068] The power operation site safety monitoring object tracking method provided by the present application comprises:

[0069] Step 101, acquiring a plurality of site pictures corresponding to the power operation site, and constructing an initial background picture and an initial work picture corresponding to the power operation site according to the site pictures.

[0070] The site pictures refer to photos of different time periods and different angles collected by a plurality of cameras arranged at the power operation site.

[0071] The initial background picture refers to a background picture obtained by splicing pictures of a work site of an unattended substation under the condition that no work ticket is executed.

[0072] The initial work picture refers to a work picture obtained by splicing pictures of a work site of an electric power operation under the condition that an object of safety supervision works.

[0073] In the embodiment of the present application, a plurality of pictures corresponding to a work site of an electric power operation are acquired, and a work state corresponding to each picture is acquired, the pictures are classified according to the work state, a first picture set and a second picture set are generated. The pictures in the first picture set are spliced to generate an initial background picture corresponding to the work site of the electric power operation. The pictures in the second picture set are spliced to generate an initial work picture corresponding to the work site of the electric power operation.

[0074] Step 102, according to the matching degree of the initial background picture and the initial work picture, determine the target background picture and the target work picture corresponding to the work site of the electric power operation.

[0075] In the embodiment of the present application, a first initial reference line set corresponding to the initial background picture and a second initial reference line set corresponding to the initial work picture are acquired. The cosine value corresponding to each reference line in the first initial reference line set and the second initial reference line set is calculated. If the cosine value meets a preset threshold, the initial background picture and the initial work picture at the current time are taken as the target background picture and the target work picture corresponding to the work site of the electric power operation.

[0076] Step 103, the target background picture and the target work picture are subjected to difference fractal dimension calculation to determine the initial object of safety supervision corresponding to the work site of the electric power operation.

[0077] In the embodiment of the present application, the target background picture and the target work picture are subjected to feature comparison to generate a difference picture corresponding to the work site of the electric power operation. A plurality of difference points and a plurality of islands in the difference picture are selected according to a preset selection standard. The difference singular coefficient corresponding to each difference point is calculated, and a plurality of difference singular spectra corresponding to the islands are determined in combination with a preset difference singular spectrum formula. The maximum value in the difference singular spectrum is taken as a first detection operator corresponding to the island, the sum of all difference singular spectra is calculated to generate a second detection operator corresponding to the island. And the difference singular spectrum is integrated to generate a third detection operator corresponding to the island. By comparing the first detection operator, the second detection operator and the third detection operator with the corresponding standard safety supervision defect feature value respectively, a feature comparison result corresponding to each island is generated. Thus, according to the feature comparison result, the initial object of safety supervision corresponding to the work site of the electric power operation is determined.

[0078] Step 104, determining the target safety supervision object corresponding to the power operation site according to the picture data corresponding to the on-site picture and the initial safety supervision object.

[0079] In the embodiment of the present application, first, the parameter data dimension and the sampling data corresponding to the initial safety supervision object in the picture data are acquired. Then, the symbolic dynamics expression corresponding to the initial safety supervision object is constructed by using the parameter data dimension and the sampling data. Finally, the symbolic dynamics expression is taken as the target safety supervision object corresponding to the power operation site.

[0080] In the embodiment of the present application, by acquiring multiple on-site pictures corresponding to the power operation site, and constructing the initial background picture and the initial work picture corresponding to the power operation site according to the on-site pictures, the target background picture and the target work picture corresponding to the power operation site are determined based on the matching degree of the initial background picture and the initial work picture. The initial safety supervision object corresponding to the power operation site is determined by difference fractal dimension calculation of the target background picture and the target work picture. Finally, the target safety supervision object corresponding to the power operation site is determined according to the picture data corresponding to the on-site picture and the initial safety supervision object. The technical problem that the existing power operation site safety supervision object tracking method adopts manual safety supervision object tracking analysis with large workload and low analysis result accuracy is solved. By difference fractal dimension calculation, the fractal dimension features of the safety supervision object are extracted, the general requirements and individualized special feature expressions of the safety supervision object are taken into account, and the continuous tracking under different monitoring angles and performance forms can be met.

[0081] Please refer to Figure 2 , Figure 2 The steps of a power operation site safety supervision object tracking method provided for the first embodiment of the present application.

[0082] Another power operation site safety supervision object tracking method provided by the present application comprises:

[0083] Step 201, acquiring multiple on-site pictures corresponding to the power operation site, and constructing the initial background picture and the initial work picture corresponding to the power operation site according to the on-site pictures.

[0084] Further, step 201 can comprise the following sub-steps S11-S14:

[0085] S11, acquiring the work states corresponding to the on-site pictures respectively.

[0086] S12, classifying the on-site pictures according to the work states to generate a first on-site picture set and a second on-site picture set.

[0087] S13, performing picture splicing on the pictures in the first on-site picture set to generate the initial background picture corresponding to the power operation site.

[0088] S14, picture splicing is performed on the pictures in the second field picture set to generate an initial work picture corresponding to the power operation field.

[0089] In the embodiment of the present application, it is assumed that K1 cameras are arranged in the power operation field, and K2 safety supervision objects, i.e., safety supervision element objects, need to be managed. All the safety supervision element objects participating in the power field operation are in the identified area. The safety supervision element objects include two types of deformable objects (workers) and non-deformable objects (devices and equipment required for work). Due to deformation or occlusion of non-deformable objects, object recognition deviation may occur. Therefore, effective identification of the safety supervision objects is needed.

[0090] Since the field pictures are pictures collected by the cameras in the unattended substation when there is no work ticket / operation ticket execution in the work field area under the condition of no work and safety supervision element object work. Therefore, by determining the work state corresponding to the field picture and classifying the field picture according to the work state type, the first field picture set and the second field picture set are generated. The field pictures in the first field picture set and the second field picture set are spliced by using a feature extraction and registration combined image splicing fusion algorithm, so as to generate an initial background picture G1 and an initial work picture G2 corresponding to the power operation field.

[0091] Step 202, according to the matching degree corresponding to the initial background picture and the initial work picture, determine the target background picture and the target work picture corresponding to the power operation field.

[0092] Further, step 202 can include the following sub-steps S21-S23:

[0093] S21, obtain a first initial reference line set corresponding to the initial background picture and a second initial reference line set corresponding to the initial work picture.

[0094] S22, calculate the cosine value corresponding to each reference line in the first initial reference line set and the second initial reference line set, respectively.

[0095] S23, if the cosine value meets the preset threshold value, the initial background picture and the initial work picture are taken as the target background picture and the target work picture corresponding to the power operation field.

[0096] The preset threshold value refers to a critical value for detecting whether the cosine value corresponding to each reference line meets the requirement, which is usually set to 0.9.

[0097] In the embodiment of the present application, the uppermost line, the lowermost line and the middle line in the initial background picture and the initial working picture are selected respectively to construct the first initial reference line set and the second initial reference line set, that is, each initial reference line set includes three reference lines respectively. The cosine values corresponding to each reference line are calculated and compared with the preset threshold value respectively, if at least one reference line meets the preset threshold value, the initial background picture and the initial working picture at the current time are taken as the target background picture and the target working picture corresponding to the power operation site. That is, the matching line segment in the image is found from the six reference lines as the target reference line for accurate registration. If no matching line segment is found, it means that the camera data is wrong, and the program is exited after warning.

[0098] Further, the first initial reference line set includes a plurality of first initial reference lines; the second initial reference line set includes a plurality of second initial reference lines; the cosine value includes a first cosine value and a second cosine value, and step S22 can include the following sub-steps S221-S226:

[0099] S221, selecting a first matching line corresponding to the first initial reference line in the initial working picture respectively.

[0100] S222, constructing a first color system vector and a second color system vector corresponding to the first initial reference line and the first matching line according to the gray value of the pixel point respectively.

[0101] S223, calculating the cosine value corresponding to the first color system vector and the second color system vector to generate the first cosine value.

[0102] S224, selecting a second matching line corresponding to the second initial reference line in the initial background picture respectively.

[0103] S225, constructing a third color system vector and a fourth color system vector corresponding to the second initial reference line and the second matching line according to the gray value of the pixel point respectively.

[0104] S226, calculating the second cosine value corresponding to the third color system vector and the fourth color system vector.

[0105] The first matching line refers to the line segment in the initial working picture corresponding to the first initial reference line, that is, the line segment with the same position of the power operation site as the first initial reference line.

[0106] The second matching line refers to the line segment in the initial background picture corresponding to the second initial reference line, that is, the line segment with the same position of the power operation site as the second initial reference line.

[0107] In the embodiment of the present application, the first initial reference line and the first matching line refer to line segments corresponding to different working states at the same position in the power operation site, and the second initial reference line and the second matching line have the same meaning. The first matching line corresponding to the first initial reference line in the initial working picture is selected, and the first color system vector and the second color system vector corresponding to the first initial reference line and the first matching line are respectively constructed according to the gray values of the pixel points in the order of RGB. The cosine value corresponding to the first color system vector and the second color system vector is calculated by using the cosine value formula, so as to obtain the corresponding first cosine value. The second matching line corresponding to the second initial reference line in the initial background picture is selected, and the third color system vector and the fourth color system vector corresponding to the second initial reference line and the second matching line are respectively constructed according to the gray values of the pixel points in the order of RGB. The cosine value corresponding to the third color system vector and the fourth color system vector is calculated by using the cosine value formula, so as to obtain the corresponding second cosine value.

[0108] For example, the line segment f1((x1, y1), (x2, y2)) on the initial background picture G1 and the line segment f2((x3, y3), (x4, y4)) on the initial working picture G2 are selected, and the coordinate points are the starting point and the ending point of the line segment, wherein at least one of f1 and f2 is a reference line in the first initial reference line set or the second initial reference line set. The color system vector is constructed according to the gray values of the pixel points in the order of RGB and The cosine of the angle between the two color system vectors is taken as the criterion, n is min(x2-x1, x4-x3), and the cosine value formula is:

[0109]

[0110] The preset threshold is 0.9, when the cosine value cosθ≥0.9, it is considered that the two lines are matched, wherein n is the number of points on the line segment, j is the serial number of the point on the line segment, and is the RGB vector of point j.

[0111] Step 203, difference fractal dimension calculation is performed on the target background picture and the target working picture, and an initial safety supervision object corresponding to the power operation site is determined.

[0112] Further, step 203 can include the following sub-steps S31-S38:

[0113] S31, feature comparison is performed on the target background picture and the target working picture, and a difference picture corresponding to the power operation site is generated.

[0114] S32, a plurality of difference points and a plurality of isolated islands in the difference picture are selected according to the preset selection standard.

[0115] S33, respectively calculate the difference singular coefficient corresponding to the difference point, and combine the preset difference singular spectrum formula to determine the plurality of difference singular spectrum corresponding to the island.

[0116] S34, the maximum value in the difference singular spectrum is taken as the first detection operator corresponding to the island.

[0117] S35, the sum of all difference singular spectrum is calculated to generate the second detection operator corresponding to the island.

[0118] S36, respectively integrate the difference singular spectrum to generate the third detection operator corresponding to the island.

[0119] S37, the first detection operator, the second detection operator and the third detection operator are compared with the corresponding standard safety defect characteristic value respectively to generate the characteristic comparison result corresponding to the island.

[0120] S38, according to the characteristic comparison result, the initial safety object corresponding to the power operation site is determined.

[0121] The points of 0 in the target background picture and the target work picture indicate that there is no difference in the target background picture and the target work picture, and the points not equal to 0 indicate the object to be tracked, and the object to be tracked is certainly surrounded by the value of 0, that is, the preset selection standard refers to taking each part surrounded by 0 as an island. The points not equal to 0 are taken as difference points.

[0122] The standard safety defect characteristic value refers to the critical value corresponding to the first detection operator, the second detection operator and the third detection operator realized by setting, which is used to measure whether the first detection operator, the second detection operator and the third detection operator satisfy the early warning condition.

[0123] In the embodiment of the application, first, the target background picture and the target work picture are compared in features, the different parts of the target work picture and the target background picture are used to construct the difference picture G3 corresponding to the power operation site, that is, to construct the image area needing fractal analysis, which is equivalent to G3=G2-G1. And each part surrounded by 0 is taken as an island. The points not equal to 0 are taken as difference points. The difference singular coefficient corresponding to the difference point is calculated respectively, and the plurality of difference singular spectrum corresponding to the island is determined in combination with the preset difference singular spectrum formula. The maximum value in the difference singular spectrum is taken as the first detection operator corresponding to the island, and the first detection operator is Δ(α)=max(P(α i )). The sum of all difference singular spectrum is taken as the second detection operator, that is, the second detection operator is Ψ(α)=∑P(α i ). The difference singular spectrum is integrated respectively, and the integral result is taken as the third detection operator, that is, the third detection operator is Θ=∫P(α i )dα iThe feature comparison result corresponding to the island is generated by calculating the data of each image island respectively and comparing with the corresponding standard safety supervision defect characteristic value respectively. The feature comparison result is consistent, and the corresponding initial safety supervision object is obtained. In addition, for the image that cannot directly find the corresponding defect, it is possible that multiple defects are in one image. Therefore, the traversal method is used to find the possible combination of multiple safety supervision defects and list as the initial safety supervision object, i.e. the safety supervision early warning object.

[0124] Further, the step S33 can include the following sub-steps S331-S335:

[0125] S331, selecting a plurality of square regions centered on the difference point according to the preset size.

[0126] S332, respectively substituting the maximum difference value in the square region into the preset differential singular coefficient formula to generate the differential singular coefficient corresponding to the square region.

[0127] The preset differential singular coefficient formula is:

[0128]

[0129] Wherein, x(m,n) is the differential singular coefficient; ε i is the side length of the square region; μ i (m,n) is the maximum difference value in the square region; (m,n) is the coordinate of the difference point.

[0130] S333, using the maximum value and the minimum value of the differential singular coefficient corresponding to the island to construct the singular coefficient interval corresponding to the island.

[0131] S334, respectively dividing the singular coefficient interval according to the standard spectrum segmentation number to generate a plurality of singular coefficient local intervals.

[0132] S335, respectively substituting the singular value number corresponding to the singular coefficient local interval and the corresponding differential singular coefficient into the preset differential singular spectrum formula to generate a plurality of differential singular spectra corresponding to the island.

[0133] The preset differential singular spectrum formula is:

[0134]

[0135] Wherein, P(α i ) is the differential singular spectrum; is the number of singular values falling in the singular coefficient local interval; x α (m,n) is the differential singular coefficient of the corresponding difference point under the α measure; (m,n) is the coordinate of the difference point; α i represents the left boundary value of the i-th singular coefficient local interval.

[0136] The preset size refers to the size of the square region corresponding to ε i = 2i-1, i = 1, 2, 3, 4, and i is the serial number of the point on the target reference line.

[0137] In the embodiment of the present application, for any point p(m, n) of G3 that is not equal to 0, that is, a difference point, a square region with a size of ε i = 2i-1, i = 1, 2, 3, 4, and i is the serial number of the point on the target reference line. The maximum difference value μ i (m, n) = max(p(m, n)) | (m, n) ∈ ε i is calculated in each square region, respectively. The maximum difference value in each square region is substituted into the preset difference singular coefficient formula, thereby obtaining the difference singular coefficient corresponding to each square region.

[0138] For each island surrounded by 0 in G3, the maximum value α max and the minimum value α min of the difference singular coefficient in each island are obtained, and the singular coefficient interval [α max , α min ] corresponding to the island is constructed. The singular coefficient interval is evenly divided into N parts, a plurality of singular coefficient local intervals are generated, N is the standard spectrum segmentation number, which is 4 times the number of safety supervision defects, and is usually set to 40.

[0139] By respectively substituting the singular value number corresponding to the singular coefficient local interval and the corresponding difference singular coefficient into the preset difference singular spectrum formula, a plurality of difference singular spectra corresponding to the island are calculated.

[0140] Step 204, obtaining the parameter data dimension and sampling data corresponding to the initial safety supervision object in the picture data.

[0141] In the embodiment of the present application, after determining the initial safety supervision object, the parameter data dimension corresponding to the initial safety supervision object is obtained from the picture data, at least including its spatial coordinates. And the corresponding sampling data is obtained, including sampling start time, sampling serial number and sampling interval, etc.

[0142] Step 205, constructing the symbolic dynamics expression corresponding to the initial safety supervision object by using the parameter data dimension and the sampling data.

[0143] In the embodiment of the present application, the initial safety supervision object is expressed in the form of symbolic dynamics by using the parameter data dimension and the sampling data, and the symbolic dynamics expression obtained is:

[0144]

[0145] Ind is the parameter corresponding to the initial safety monitoring object, wherein m is the parameter data dimension of the initial safety monitoring object, at least including its spatial coordinates, t is the sampling starting time, k is the sampling serial number, and Δt is the sampling interval.

[0146] In step 206, the symbolic dynamics expression is taken as the target safety monitoring object corresponding to the power operation site.

[0147] In the embodiment of the present application, the symbolic dynamics expression is taken as the target safety monitoring object corresponding to the power operation site, so that the safety monitoring element object framework can be expanded based on symbolic dynamics tracking, and tracking can be performed in time sequence, which can be effectively applied to simple or complex operation scenes.

[0148] In the embodiment of the present application, a plurality of site pictures corresponding to the power operation site are acquired, and an initial background picture and an initial work picture corresponding to the power operation site are constructed according to the site pictures. Then, a target background picture and a target work picture corresponding to the power operation site are determined based on the matching degree of the initial background picture and the initial work picture. The initial safety monitoring object corresponding to the power operation site is determined by performing differential fractal dimension calculation on the target background picture and the target work picture. The parameter data dimension and the sampling data of the initial safety monitoring object in the picture data are acquired. The symbolic dynamics expression corresponding to the initial safety monitoring object is constructed by using the parameter data dimension and the sampling data. The symbolic dynamics expression is taken as the target safety monitoring object corresponding to the power operation site. The safety monitoring element object is segmented by using the fractal algorithm, and the fractal dimension feature is extracted, which takes into account the general requirements and individual special feature expression of the safety monitoring object. The safety monitoring element object is labeled by using the fractal algorithm and the feature, which can meet the continuous tracking under different monitoring angles and performance modes.

[0149] Please refer to Figure 3 , Figure 3 FIG. 3 is a structural block diagram of a power operation site safety monitoring object tracking system provided in the third embodiment of the present application.

[0150] The embodiment of the present application provides a power operation site safety monitoring object tracking system, which comprises:

[0151] The initial background picture and the initial work picture construction module 301 is configured to acquire a plurality of site pictures corresponding to the power operation site, and construct an initial background picture and an initial work picture corresponding to the power operation site according to the site pictures.

[0152] The target background picture and the target work picture determination module 302 is configured to determine a target background picture and a target work picture corresponding to the power operation site according to the matching degree of the initial background picture and the initial work picture.

[0153] The initial safety monitoring object determination module 303 is configured to determine the initial safety monitoring object corresponding to the power operation site by performing differential fractal dimension calculation on the target background picture and the target work picture.

[0154] The target safety monitoring object determination module 304 is configured to determine the target safety monitoring object corresponding to the power operation site according to the picture data corresponding to the site picture and the initial safety monitoring object.

[0155] Optionally, the initial background picture and initial work picture construction module 301 comprises:

[0156] The work state acquisition module is configured to acquire the work state corresponding to the site picture respectively.

[0157] The first site picture set and the second site picture set generation module is configured to classify the site pictures according to the work state, and generate the first site picture set and the second site picture set.

[0158] The initial background picture generation module is configured to perform picture splicing on the pictures in the first site picture set, and generate the initial background picture corresponding to the power operation site.

[0159] The initial work picture generation module is configured to perform picture splicing on the pictures in the second site picture set, and generate the initial work picture corresponding to the power operation site.

[0160] Optionally, the target background picture and target work picture determination module 302 comprises:

[0161] The first initial reference line set and the second initial reference line set acquisition module is configured to acquire the first initial reference line set corresponding to the initial background picture and the second initial reference line set corresponding to the initial work picture.

[0162] The cosine value calculation module is configured to calculate the cosine value corresponding to each reference line in the first initial reference line set and the second initial reference line set respectively.

[0163] The target background picture and target work picture determination submodule is configured to, if the cosine value meets the preset threshold, take the initial background picture and the initial work picture as the target background picture and the target work picture corresponding to the power operation site.

[0164] Optionally, the first initial reference line set comprises a plurality of first initial reference lines, and the second initial reference line set comprises a plurality of second initial reference lines. The cosine value comprises a first cosine value and a second cosine value, and the cosine value calculation module can perform the following steps:

[0165] Selecting a first matching line corresponding to the first initial reference line in the initial work picture respectively;

[0166] According to the gray value of the pixel point, a first color system vector corresponding to the first initial reference line and a second color system vector corresponding to the first matching line are constructed respectively;

[0167] A first cosine value corresponding to the first color system vector and the second color system vector is calculated to generate the first cosine value;

[0168] A second matching line corresponding to the second initial reference line in the initial background picture is selected respectively;

[0169] According to the gray value of the pixel point, a third color system vector corresponding to the second initial reference line and a fourth color system vector corresponding to the second matching line are constructed respectively;

[0170] A second cosine value corresponding to the third color system vector and the fourth color system vector is calculated.

[0171] Optionally, the initial safety supervision object determination module 303 comprises:

[0172] The difference picture generation module is configured to compare the target background picture and the target work picture in terms of features, and generate a difference picture corresponding to the power operation site.

[0173] The difference point and island selection module is configured to select a plurality of difference points and a plurality of islands in the difference picture according to a preset selection standard.

[0174] The difference singular spectrum determination module is configured to calculate a difference singular coefficient corresponding to each difference point, and determine a plurality of difference singular spectra corresponding to the islands according to a preset difference singular spectrum formula.

[0175] The first detection operator determination module is configured to take the maximum value in the difference singular spectrum as a first detection operator corresponding to the island.

[0176] The second detection operator generation module is configured to calculate the sum of all difference singular spectra to generate a second detection operator corresponding to the island.

[0177] The third detection operator generation module is configured to integrate the difference singular spectra respectively to generate a third detection operator corresponding to the island.

[0178] The feature comparison result generation module is configured to compare the first detection operator, the second detection operator and the third detection operator with corresponding standard safety supervision defect feature values respectively to generate a feature comparison result corresponding to the island.

[0179] The initial safety supervision object determination sub-module is configured to determine an initial safety supervision object corresponding to the power operation site according to the feature comparison result.

[0180] Optionally, the difference singular spectrum determination module can perform the following steps:

[0181] A plurality of square regions centered on the difference point are selected according to preset sizes;

[0182] The maximum difference value in each square region is substituted into a preset difference singular coefficient formula to generate a difference singular coefficient corresponding to the square region.

[0183] The preset difference singular coefficient formula is:

[0184]

[0185] wherein x(m, n) is the difference singular coefficient; ε i is the side length of the square region; μ i (m, n) is the maximum difference value in the square region; (m, n) is the coordinate of the difference point.

[0186] The maximum value and the minimum value of the difference singular coefficient corresponding to the island are used to construct a singular coefficient interval corresponding to the island.

[0187] The singular coefficient interval is divided into a plurality of singular coefficient local intervals according to a standard spectrum segmentation number.

[0188] The singular value number corresponding to each singular coefficient local interval and the corresponding difference singular coefficient are substituted into a preset difference singular spectrum formula to generate a plurality of difference singular spectra corresponding to the island.

[0189] The preset difference singular spectrum formula is:

[0190]

[0191] wherein P(α i ) is the difference singular spectrum; is the number of singular values falling in the singular coefficient local interval; x α (m, n) is the difference singular coefficient of the corresponding difference point under the α measure; (m, n) is the coordinate of the difference point; α i represents the left boundary value of the i-th singular coefficient local interval.

[0192] Optionally, the target safety supervision object determination module 304 comprises:

[0193] A parameter data dimension and sampling data acquisition module is configured to acquire parameter data dimensions and sampling data corresponding to the initial safety supervision object in the picture data.

[0194] A symbolic dynamics expression construction module is configured to construct a symbolic dynamics expression corresponding to the initial safety supervision object by using the parameter data dimensions and the sampling data.

[0195] A target safety supervision object determination sub-module is configured to use the symbolic dynamics expression as the target safety supervision object corresponding to the electric power operation site.

[0196] The embodiment of the present application further provides an electronic device, which comprises a memory and a processor, and the memory stores a computer program; the computer program is executed by the processor to enable the processor to execute the power operation site safety monitoring object tracking method according to any one of the above embodiments.

[0197] The memory can be an electronic memory such as a flash memory, an EEPROM (Electrically Erasable Programmable Read-Only Memory), an EPROM, a hard disk or a ROM. The memory has a storage space for program codes for executing any method step in the above method. For example, the storage space for program codes can include respective program codes for respectively implementing various steps in the above method. These program codes can be read from or written into one or more computer program products. These computer program products include program code carriers such as a hard disk, a compact disc (CD), a memory card or a floppy disk. The program codes can be compressed in an appropriate form, for example. These codes, when executed by a computing processing device, cause the computing processing device to execute respective steps in the power operation site safety monitoring object tracking method described above.

[0198] The embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the power operation site safety monitoring object tracking method according to any one of the above embodiments.

[0199] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiments, which will not be described here.

[0200] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other manners. For example, the above-described device embodiments are merely schematic; the division of units is merely a logical function division; an actual implementation can be another division manner, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, devices or units, and can be electrical, mechanical or in other forms.

[0201] The units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, i.e., may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0202] In addition, each functional unit in each embodiment of the application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0203] If the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the application, the essential part or the whole or part of the prior art, or the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method of each embodiment of the application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program code storage media.

[0204] The above embodiments are only used to illustrate the technical solutions of the application, but not to limit them; although the application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that the technical solutions recorded in the foregoing embodiments can be modified, or some technical features can be replaced by equivalents; and these modifications or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the application.

Claims

1. A power operation site safety monitoring object tracking method, characterized in that, The method comprises the following steps: acquiring multiple site pictures corresponding to a power operation site, and constructing an initial background picture and an initial work picture corresponding to the power operation site according to the site pictures; determining a target background picture and a target work picture corresponding to the power operation site according to a matching degree corresponding to the initial background picture and the initial work picture; performing differential fractal dimension calculation on the target background picture and the target work picture to determine an initial safety monitoring object corresponding to the power operation site; determining a target safety monitoring object corresponding to the power operation site according to picture data corresponding to the site pictures and the initial safety monitoring object; the step of performing differential fractal dimension calculation on the target background picture and the target work picture to determine the initial safety monitoring object corresponding to the power operation site comprises the following steps: performing feature comparison on the target background picture and the target work picture to generate a difference picture corresponding to the power operation site; selecting multiple difference points and multiple islands in the difference picture according to a preset selection standard; calculating a differential singular coefficient corresponding to each difference point, and determining multiple differential singular spectra corresponding to the islands in combination with a preset differential singular spectrum formula; taking a maximum value in the differential singular spectra as a first detection operator corresponding to the islands; calculating a sum value of all the differential singular spectra to generate a second detection operator corresponding to the islands; integrating the differential singular spectra respectively to generate a third detection operator corresponding to the islands; comparing the first detection operator, the second detection operator and the third detection operator with corresponding standard safety monitoring defect feature values respectively to generate a feature comparison result corresponding to the islands; and determining the initial safety monitoring object corresponding to the power operation site according to the feature comparison result.

2. The power work site safety object tracking method of claim 1, wherein, The step of constructing the initial background picture and the initial work picture corresponding to the power operation site according to the site pictures comprises the following steps: acquiring work states corresponding to the site pictures respectively; classifying the site pictures according to the work states to generate a first site picture set and a second site picture set; performing picture splicing on pictures in the first site picture set to generate the initial background picture corresponding to the power operation site; performing picture splicing on pictures in the second site picture set to generate the initial work picture corresponding to the power operation site.

3. The power work site safety object tracking method of claim 1, wherein, The step of determining the target background picture and the target work picture corresponding to the power operation site according to a matching degree corresponding to the initial background picture and the initial work picture comprises the following steps: acquiring a first initial reference line set corresponding to the initial background picture and a second initial reference line set corresponding to the initial work picture; calculating a cosine value corresponding to each reference line in the first initial reference line set and the second initial reference line set respectively; if the cosine value meets a preset threshold value, taking the initial background picture and the initial work picture as the target background picture and the target work picture corresponding to the power operation site.

4. The power work site safety object tracking method of claim 3, wherein, The first initial reference line set includes a plurality of first initial reference lines; the second initial reference line set includes a plurality of second initial reference lines; the cosine values include first cosine values and second cosine values; The step of respectively calculating the cosine values corresponding to each reference line in the first initial reference line set and the second initial reference line set includes: Respectively selecting a first matching line corresponding to the first initial reference line in the initial working picture; According to the gray value of the pixel point, a first color system vector corresponding to the first initial reference line and a second color system vector are respectively constructed; The first color system vector and the second color system vector are calculated to generate a first cosine value; Respectively selecting a second matching line corresponding to the second initial reference line in the initial background picture; According to the gray value of the pixel point, a third color system vector corresponding to the second initial reference line and a fourth color system vector are respectively constructed; The third color system vector and the fourth color system vector are calculated to generate a second cosine value.

5. The power work site safety object tracking method of claim 1, wherein, The step of respectively calculating the difference singular coefficient corresponding to the difference point, and combining the preset difference singular spectrum formula to determine a plurality of difference singular spectra corresponding to the island includes: According to the preset size, a plurality of square regions centered on the difference point are selected; The maximum difference value in the square region is respectively substituted into the preset difference singular coefficient formula to generate the difference singular coefficient corresponding to the square region; The preset difference singular coefficient formula is: ； wherein, is the difference singular coefficient; is the side length of the square region; is the maximum difference value in the square region; is the coordinate of the difference point; The maximum value and the minimum value of the difference singular coefficient corresponding to the island are used to construct a singular coefficient interval corresponding to the island; According to the standard spectrum segmentation number, the singular coefficient interval is respectively divided to generate a plurality of singular coefficient local intervals; The singular value number corresponding to the singular coefficient local interval and the corresponding difference singular coefficient are respectively substituted into the preset difference singular spectrum formula to generate a plurality of difference singular spectra corresponding to the island; The preset difference singular spectrum formula is: ； wherein, is the difference singular spectrum; is the number of singular values falling in the local interval of the singular coefficient; is the difference singular coefficient of the corresponding difference point under the measure is the difference singular coefficient of the corresponding difference point under the measure is the coordinate of the difference point; represents the left boundary value of the i-th singular coefficient local interval.

6. The power work site safety object tracking method of claim 1, wherein, The step of determining the target safety monitoring object corresponding to the power operation site according to the picture data corresponding to the field picture and the initial safety monitoring object includes: Obtaining the parameter data dimension and the sampling data corresponding to the initial safety monitoring object in the picture data; Using the parameter data dimension and the sampling data, a symbolic dynamics expression corresponding to the initial safety monitoring object is constructed; The symbolic dynamics expression is used as the target safety monitoring object corresponding to the power operation site.

7. A power work site safety object tracking system, comprising: It includes: An initial background picture and an initial working picture construction module is configured to acquire a plurality of field pictures corresponding to a power operation site, and construct an initial background picture and an initial working picture corresponding to the power operation site according to the field pictures; A target background picture and a target working picture determination module is configured to determine a target background picture and a target working picture corresponding to the power operation site according to the matching degree of the initial background picture and the initial working picture. An initial safety supervision object determination module is configured to perform difference fractal dimension calculation on the target background picture and the target work picture to determine an initial safety supervision object corresponding to the power operation site; A target safety supervision object determination module is configured to determine a target safety supervision object corresponding to the power operation site according to picture data corresponding to the site picture and the initial safety supervision object; The initial safety supervision object determination module includes: A difference picture generation module is configured to perform feature comparison on the target background picture and the target work picture to generate a difference picture corresponding to the power operation site; A difference point and island selection module is configured to select a plurality of difference points and a plurality of islands in the difference picture according to a preset selection standard; A difference singular spectrum determination module is configured to calculate difference singular coefficients corresponding to the difference points respectively and determine a plurality of difference singular spectra corresponding to the islands in combination with a preset difference singular spectrum formula; A first detection operator determination module is configured to take a maximum value in the difference singular spectra as a first detection operator corresponding to the islands; A second detection operator generation module is configured to calculate a sum value of all the difference singular spectra to generate a second detection operator corresponding to the islands; A third detection operator generation module is configured to integrate the difference singular spectra respectively to generate a third detection operator corresponding to the islands; A feature comparison result generation module is configured to compare the first detection operator, the second detection operator and the third detection operator with corresponding standard safety supervision defect feature values respectively to generate a feature comparison result corresponding to the islands; An initial safety supervision object determination sub-module is configured to determine an initial safety supervision object corresponding to the power operation site according to the feature comparison result.

8. An electronic device, comprising: A memory and a processor are included, the memory stores a computer program, and the computer program is executed by the processor to make the processor execute steps of the power operation site safety supervision object tracking method according to any one of claims 1-6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed to implement the power operation site safety supervision object tracking method according to any one of claims 1-6.

Images