Monitoring video optimization method based on transformer substation three-dimensional model

Abstract

The invention relates to the technical field of monitoring, and discloses a monitoring video optimization method based on a transformer substation three-dimensional model, which comprises the steps of data acquisition, three-dimensional model construction, preprocessing, video analysis, abnormity judgment and information display, and is accurate in three-dimensional model construction and effective in application. A model constructed by combining laser scanning and 3DMAX software truly restores a substation scene, and an accurate space basis is provided for video analysis. The marking and displaying functions are highlighted, so that monitoring personnel can visually check abnormity and changes in the three-dimensional model, and the monitoring intuition and effectiveness are improved; abnormality judgment and analysis are scientific and reasonable, the abnormal evaluation value is calculated through comprehensive analysis of multiple video frames, abnormal events are accurately recognized, abnormal information is fed back in time, rapid positioning and processing are assisted, the response and processing efficiency of abnormal conditions of the transformer substation is improved, safe and stable operation of the transformer substation is comprehensively guaranteed, and the method has important practical application value.

Description

Technical Field

[0001] The present invention relates to the field of monitoring technologies, and particularly to a method for optimizing monitoring videos based on a three-dimensional model of a substation. Background Art

[0002] With the continuous development and intelligent upgrade of the power system, as a key node for power transmission and distribution, the safe operation of a substation is of crucial importance. Traditional substation monitoring mainly relies on two-dimensional monitoring videos, however, this method has many limitations. On the one hand, two-dimensional monitoring videos can only provide information from a planar perspective, and it is difficult to comprehensively and accurately reflect the complex spatial layout and equipment status of the substation. When observing the video, it is difficult for monitoring personnel to quickly and intuitively correspond the scene in the video to the actual substation space and equipment, and it may be difficult to detect some subtle changes or potential abnormal situations, resulting in an inaccurate grasp of the overall operation status of the substation. On the other hand, traditional monitoring videos are relatively simple in data processing and analysis, lacking effective data preprocessing and accurate analysis methods. For example, the video data collected from different monitoring points may have differences in resolution, brightness, contrast, etc., which brings difficulties to subsequent data analysis and comparison, and is prone to errors and misjudgments. At the same time, for the detection of abnormal events in the video, it often relies on manual observation or simple threshold judgment, lacking systematic and scientific analysis means, and it is difficult to accurately identify complex abnormal situations, such as slight displacement of equipment, local abnormal temperature changes, etc., which poses a potential threat to the safe operation of the substation.

[0003] Therefore, in order to improve the accuracy, reliability, and intuitiveness of substation monitoring, there is an urgent need for an innovative method for optimizing monitoring videos, which can make full use of advanced technical means, such as three-dimensional modeling, accurate data processing, and analysis algorithms, etc., to comprehensively optimize substation monitoring videos, so as to better meet the needs of substation safety monitoring and ensure the stable operation of the power system. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for optimizing monitoring videos based on a three-dimensional model of a substation, which solves the technical problems proposed in the background art.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A method for optimizing monitoring videos based on a three-dimensional model of a substation includes the following steps:

[0007] Step 1: Data collection

[0008] Collect video data of each monitoring point in the substation;

[0009] Step 2: Three-dimensional model construction

[0010] Obtain the spatial information and equipment geometric shape data of the substation, and construct a three-dimensional model of the substation containing the equipment;

[0011] Step Three: Preprocessing

[0012] Normalize the video data collected at each monitoring point in the substation, and obtain the corresponding normalized data;

[0013] Step Four: Video Analysis

[0014] Analyze the change of frame content for each video frame in combination with the background image. Through the analysis results, determine the prominent pixel points of each video frame, and make prominent marks in combination with the three-dimensional model of the substation;

[0015] Step Five: Abnormality Judgment

[0016] Based on the analysis of abnormal events for multiple consecutive adjacent video frames corresponding to each monitoring point, generate abnormal data packets with abnormal events and their corresponding time intervals;

[0017] Step Six: Information Display

[0018] At each monitoring point, display the prominent pixel points marked prominently in the three-dimensional model of the substation corresponding to the current video frame on the preset main interface corresponding to the three-dimensional model of the substation. At the same time, in the three-dimensional model of the substation, extract the abnormal data packets separately and display them separately through the preset sub-interface.

[0019] As a further solution of the present invention: Among them, the video data is collected by a high-definition camera at a specified frame rate.

[0020] As a further solution of the present invention: The three-dimensional model construction method is as follows:

[0021] Step1.1: Obtain the spatial information and equipment geometric shape data of the substation through laser scanning technology;

[0022] Collect the spatial information and equipment geometric shape data of the substation;

[0023] At the same time, collect the position of the equipment in the substation space in the form of coordinates;

[0024] Step1.2: Utilize the spatial information and equipment geometric shape data of the substation, in combination with the use of three-dimensional modeling software, to construct the three-dimensional space model and three-dimensional equipment model of the substation;

[0025] At the same time, according to the actual sizes of the substation and its corresponding equipment, scale and adjust the three-dimensional space model and three-dimensional equipment model;

[0026] Among them, the three-dimensional modeling software uses 3DMAX;

[0027] Step1.3. Import the 3D device model into the 3D space model;

[0028] And adjust the position of the 3D device model in the 3D space model according to the corresponding position coordinates of the device in the substation space, and generate a 3D substation model containing the device.

[0029] As a further solution of the present invention: The normalization processing method is as follows:

[0030] StepK.1. Select a monitoring point;

[0031] Obtain the video data corresponding to the monitoring point, and extract all video frames from the video data;

[0032] At the same time, obtain the background image corresponding to the monitoring point;

[0033] StepK.2. Convert all video frames and the background image into grayscale images with the same resolution;

[0034] StepK.3. Select the grayscale image corresponding to the background image

[0035] Obtain the grayscale values of each pixel point on the grayscale image corresponding to the background image, and mark them as B i , i = 1, 2,..., n, n represents the number of pixel points on the grayscale image, and Bi is the grayscale value of the i-th pixel point in the background image;

[0036] StepK.4. Among the grayscale values of each pixel point on the grayscale image corresponding to the background image, extract the largest grayscale value and mark it as B max , and at the same time extract the smallest grayscale value and mark it as B min ;

[0037] Then through Calculate the normalization value GBi of each pixel point on the grayscale image corresponding to the background image;

[0038] StepK.5. Calculate the normalization values of each pixel point on the grayscale image corresponding to each video frame in the manner of StepK.3 to StepK.4.

[0039] As a further solution of the present invention: The frame content change analysis method is as follows:

[0040] StepL.1. Select a video frame;

[0041] Extract the normalization values of each pixel point on the grayscale image corresponding to the video frame, and mark them as GSi;

[0042] StepL.2. Compare the GBi at each pixel of the background image with the GSi at each pixel of this video frame one by one:

[0043] If |GBi - GSi| > Gy, it means that the same pixel point of the video frame and the background image has changed. Then mark this pixel point on the video frame as a prominent pixel point;

[0044] If |GBi - GSi| ≤ Gy, it means that the same pixel point of the video frame and the background image has not changed. Then do not mark this pixel point on the video frame as a prominent pixel point;

[0045] And so on, to obtain all the prominent pixel points on this video frame;

[0046] StepL.3. Then, according to the shooting angle of this monitoring point, in the 3D model of the substation, obtain the model area corresponding to the angle;

[0047] Among them, the model area refers to the area range corresponding to the shooting angle of the monitoring point in the 3D model of the substation;

[0048] Import all the prominent pixel points on this video frame into the 3D model of the substation, and make prominent marks on the prominent pixel points in the 3D model of the substation through prominent color marks;

[0049] StepL.4. In the way of StepL.1 to StepL.3, according to the normalized values of each pixel point on the grayscale images corresponding to all video frames, determine the prominent pixel points on each video frame;

[0050] At the same time, following the time trend, import all the prominent pixel points on each video frame into the 3D model of the substation in sequence, and update the prominent pixel points marked as prominent in the 3D model of the substation.

[0051] As a further solution of the present invention: The abnormal event analysis method is as follows:

[0052] StepU.1. Select r consecutive adjacent video frames corresponding to a monitoring point;

[0053] At the same time, according to the timestamps corresponding to the r consecutive adjacent video frames, establish the monitoring time analysis interval of this monitoring point;

[0054] StepU.2. Select a video frame from the r video frames;

[0055] In this video frame, mark each prominent pixel point with coordinates, and record the coordinates of each prominent pixel point as (X t , Y t ), where t = 1, 2,..., v, and v represents the number of all prominent pixel points in this video frame;

[0056] Subsequently, through calculate the center coordinates (X z , Y z ) of all prominent pixel points in this video frame;

[0057] StepU.3: Calculate the center coordinates of all prominent pixel points in r video frames in the same way as in StepU.2;

[0058] StepU.4: Select two adjacent video frames, and calculate the distance difference between the corresponding center coordinates in the two adjacent video frames through the Euclidean distance formula;

[0059] And so on, obtain the distance differences between the corresponding center coordinates of all adjacent video frames, then calculate the average value of these multiple distance differences and label it as the anomaly evaluation value;

[0060] StepU.5: Subsequently, compare the anomaly evaluation value with a preset anomaly evaluation threshold:

[0061] If the anomaly evaluation value is greater than the anomaly evaluation threshold, record the monitoring time analysis interval of this monitoring point as an abnormal time interval, and at the same time bundle the prominent pixel points prominently marked in the substation three-dimensional model corresponding to r consecutive adjacent video frames into an abnormal data packet;

[0062] Otherwise, do not record the monitoring time analysis interval of this monitoring point as an abnormal time interval, and do not generate an abnormal data packet.

[0063] Advantages of the present invention:

[0064] Accurate data acquisition: By using a high-definition camera to collect videos at a specified frame rate, high-quality and high-definition video data of each monitoring point in the substation can be obtained, providing rich and accurate original information for subsequent analysis and processing, helping to observe the situation in the substation more carefully and improving the ability to capture subtle changes.

[0065] Effective data normalization processing: Normalize the collected video data, convert all video frames and background images into grayscale images of the same resolution, and calculate the normalization values of each pixel point. This processing method makes the video data collected at different monitoring points and different times have a unified standard and scale, facilitating subsequent operations such as frame content change analysis, improving the comparability of data and the accuracy of processing, and reducing errors and interferences caused by factors such as data format differences.

[0066] Precise 3D model construction: Using laser scanning technology to obtain the spatial information and equipment geometric shape data of the substation, combining with 3DMAX 3D modeling software to construct the 3D spatial model and 3D equipment model of the substation, scaling and adjusting according to the actual size, and finally accurately importing the 3D equipment model into the 3D spatial model and adjusting the position according to the position coordinates to generate the 3D model of the substation containing equipment. This precise 3D model construction can truly restore the actual scene and equipment layout of the substation, providing an accurate spatial basis for the integration of surveillance videos and the actual scene, enabling more accurate combination of actual spatial position information when analyzing video data, and improving the ability to grasp and analyze the overall situation of the surveillance area.

[0067] Prominent marking and intuitive display: During the video analysis process, analyze the video frames in combination with the background image, determine the prominent pixel points and make prominent marks in the 3D model of the substation. At the same time, in the information display section, display the prominent pixel points corresponding to the current video frame in the 3D model of the substation through the preset main interface, and display the abnormal data packets separately through the preset sub-interface. This method enables the monitoring personnel to intuitively see the abnormal situations and prominent changes in the video in the 3D model, more clearly understand the real-time status of different positions in the substation, improves the intuitiveness and effectiveness of monitoring, and facilitates timely discovery of problems and taking corresponding measures.

[0068] Scientific abnormal event analysis: For the analysis of abnormal events, by selecting multiple consecutive adjacent video frames of the monitoring points, establishing a monitoring time analysis interval, marking the coordinates of the prominent pixel points in the video frames and calculating the central coordinates, and then calculating the distance difference between the central coordinates of adjacent video frames through the Euclidean distance formula, and taking the average value as the abnormal evaluation value to compare with the preset threshold to determine the abnormality. This analysis method comprehensively considers the time series and spatial position change information of multiple video frames, can more accurately identify abnormal events, avoids the limitations and misjudgments of single video frame analysis, and improves the accuracy and reliability of abnormal detection.

[0069] Timely abnormal feedback and processing: When an abnormal event is determined, bundle the prominent pixel points of the relevant video frames into an abnormal data packet and record the abnormal time interval. This method enables the monitoring personnel to quickly locate the time and location of the abnormality, obtain relevant information in a timely manner for processing, helps to improve the response speed and processing efficiency of substation abnormal situations, and ensures the safe and stable operation of the substation.

[0070] In summary, the monitoring video optimization method of the present invention improves the quality of the monitoring video, the accuracy of analysis and the reliability of abnormal detection through various optimization measures, providing strong support for the safety monitoring of substations. Brief Description of the Drawings

[0071] The present invention will be further described below with reference to the accompanying drawings.

[0072] Figure 1 It is a system block diagram of a method for optimizing monitoring videos based on a three-dimensional substation model according to the present invention.

[0073] Figure 2 It is a schematic flowchart of the normalization process in a method for optimizing monitoring videos based on a three-dimensional substation model according to the present invention.

[0074] Figure 3 It is a schematic flowchart of the frame content change analysis in a method for optimizing monitoring videos based on a three-dimensional substation model according to the present invention.

[0075] Figure 4 It is a schematic flowchart of the abnormal event analysis in a method for optimizing monitoring videos based on a three-dimensional substation model according to the present invention. Specific Embodiments

[0076] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0077] Embodiment 1

[0078] Please refer to Figure 1 As shown, the present invention is a method for optimizing monitoring videos based on a three-dimensional substation model, including the steps: three-dimensional model construction

[0079] Obtain the spatial information and equipment geometric shape data of the substation, and construct a three-dimensional substation model containing equipment;

[0080] The specific method is as follows:

[0081] Step1.1. Obtain the spatial information and equipment geometric shape data of the substation through laser scanning technology;

[0082] Collect the spatial information and equipment geometric shape data of the substation;

[0083] At the same time, collect the positions of the equipment in the substation space in the form of coordinates;

[0084] In this embodiment, the laser scanning technology is a prior art, so it will not be elaborated here;

[0085] Step1.2. Utilize the spatial information and equipment geometric shape data of the substation, combined with the use of three-dimensional modeling software, to construct a three-dimensional spatial model and a three-dimensional equipment model of the substation;

[0086] Meanwhile, according to the actual sizes of the substation and its corresponding equipment, scale adjustments are made to the three-dimensional space model and the three-dimensional equipment model;

[0087] In this embodiment, the 3D modeling software used is 3DMAX;

[0088] Step1.3: Import the three-dimensional equipment model into the three-dimensional space model;

[0089] And according to the position coordinates corresponding to the equipment in the substation space, adjust the position of the three-dimensional equipment model in the three-dimensional space model, and generate a three-dimensional substation model containing the equipment;

[0090] In this embodiment, by obtaining the spatial information of the substation and the geometric shape data of the equipment, a three-dimensional substation model containing the equipment is constructed using laser scanning technology and 3DMAX three-dimensional modeling software, which can intuitively and accurately present the spatial layout of the substation and the equipment positions. This helps the operation and maintenance personnel better understand the overall structure of the substation, provides a basic three-dimensional visualization platform for subsequent monitoring video optimization and equipment management, and improves the efficiency and accuracy of grasping the overall situation of the substation.

[0091] Embodiment Two

[0092] Please refer to Figure 1 、 Figure 2 and Figure 3 As shown in, as Embodiment Two of the present invention, when this application is specifically implemented, compared with Embodiment One, the difference in the technical solution of this embodiment from that of Embodiment One is only that in this embodiment, the following steps are further included:

[0093] The first step: Data collection

[0094] Collect video data of each monitoring point in the substation;

[0095] In this embodiment, the video data is collected by high-definition cameras at a specified frame rate, with appropriate lenses and angles configured to ensure that the equipment, scenes, and personnel activities in the substation can be clearly captured, and the arrangement of the cameras should be planned according to the layout of the substation and the key monitoring areas to ensure no monitoring blind spots;

[0096] The second step: Pretreatment

[0097] Perform normalization processing on the video data collected at each monitoring point in the substation, and obtain the corresponding normalized data;

[0098] The normalization processing method is as follows:

[0099] StepK.1: Data preparation

[0100] Select a monitoring point;

[0101] Obtain the video data corresponding to the monitoring point, and extract all video frames from the video data;

[0102] Meanwhile, obtain the background image corresponding to the monitoring point;

[0103] StepK.2: Image conversion

[0104] Convert all video frames and the background image into grayscale images of the same resolution;

[0105] StepK.3: Background grayscale value marking

[0106] Select the grayscale image corresponding to the background image

[0107] Obtain the grayscale values of each pixel point on the grayscale image corresponding to the background image, and mark them as B i , i = 1, 2,... n, where n represents the number of pixel points on the grayscale image, and Bi is the grayscale value of the i-th pixel point in the background image;

[0108] StepK.4: Background normalization processing

[0109] Among the grayscale values of each pixel point on the grayscale image corresponding to the background image, extract the largest grayscale value and mark it as B max , and at the same time extract the smallest grayscale value and mark it as B min ;

[0110] Then, through calculate the normalized value GBi of each pixel point on the grayscale image corresponding to the background image;

[0111] StepK.5: Video frame normalization processing

[0112] Calculate the normalized values of each pixel point on the grayscale image corresponding to each video frame in the same way as in StepK.3 to StepK.4;

[0113] Third step: Video analysis

[0114] Analyze the change of frame content for each video frame in combination with the background image. Through the analysis results, determine the prominent pixel points of each video frame, and make prominent marks in combination with the 3D model of the substation;

[0115] The specific method is as follows:

[0116] StepL.1: Extraction of video frame normalized value

[0117] Select a video frame;

[0118] Extract the normalized values of each pixel on the grayscale image corresponding to the video frame, and mark them as GSi;

[0119] StepL.2: Prominent pixel judgment

[0120] Compare each GBi on the background image pixel by pixel with each GSi on the video frame:

[0121] If |GBi - GSi| > Gy, it means that the same pixel point on the video frame and the background image has changed. Then mark the pixel point on the video frame as a prominent pixel point;

[0122] If |GBi - GSi| ≤ Gy, it means that the same pixel point on the video frame and the background image has not changed. Then do not mark the pixel point on the video frame as a prominent pixel point;

[0123] And so on, to obtain all the prominent pixel points on the video frame;

[0124] StepL.3: Prominent marking of the 3D model

[0125] Then, according to the shooting angle of the monitoring point, in the 3D model of the substation, obtain the model area corresponding to the angle;

[0126] Among them, the model area refers to the area range in the 3D model of the substation corresponding to the shooting angle of the monitoring point;

[0127] Import all the prominent pixel points on the video frame into the 3D model of the substation, and make prominent markings on the prominent pixel points in the 3D model of the substation through prominent color markings;

[0128] StepL.4: Processing and updating of multiple video frames

[0129] In the way of StepL.1 to StepL.3, determine the prominent pixel points on each video frame according to the normalized values of each pixel on the grayscale image corresponding to all video frames;

[0130] At the same time, in the order of time, import all the prominent pixel points on each video frame into the 3D model of the substation in turn, and update the prominent pixel points marked as prominent in the 3D model of the substation;

[0131] Fourth step, information display

[0132] In each monitoring point, display the prominent pixel points prominently marked in the 3D model of the substation corresponding to the current video frame through the preset main interface corresponding to the 3D model of the substation.

[0133] In this embodiment, the data acquisition part collects videos at a specified frame rate through a high-definition camera, and the cameras are reasonably arranged to ensure that the equipment, scenes, and personnel activities in the substation can be clearly captured without monitoring blind spots, improving the comprehensiveness and effectiveness of monitoring; in the preprocessing, the normalization process converts the video frames and background images into grayscale images with the same resolution and calculates the normalization value, providing a unified data standard for subsequent video analysis and facilitating the accurate analysis of frame content changes; the video analysis combines the background image to analyze the changes in the frame content of the video frames, determines the prominent pixel points, and makes prominent marks in the 3D model of the substation, realizing the combination of the monitoring video and the 3D model, enabling the operation and maintenance personnel to more intuitively understand the changes in the substation, and improving the discovery speed and accuracy of abnormal situations; the information display shows the prominent pixel points marked prominently in the 3D model of the substation corresponding to the current video frame through the main interface, further enhancing the visualization effect and facilitating the operation and maintenance personnel to monitor the status of the substation in real time.

[0134] Embodiment III

[0135] Please refer to Figure 1 and Figure 4 As shown in [relevant figures], as Embodiment III of the present invention, when the present application is specifically implemented, compared with Embodiment I and Embodiment II, the technical solution of this embodiment lies in combining and implementing the solutions of the above Embodiment I and Embodiment II. The difference between the technical solution of this embodiment and Embodiment I and Embodiment II is only that in this embodiment, it further includes the step: abnormal determination

[0136] Based on the analysis of abnormal events for multiple consecutive adjacent video frames corresponding to each monitoring point, an abnormal data packet indicating the existence of an abnormal event and its corresponding time interval are generated;

[0137] The abnormal event analysis method is as follows:

[0138] StepU.1. Establish a monitoring time analysis interval

[0139] Select r consecutive adjacent video frames corresponding to a monitoring point;

[0140] At the same time, based on the timestamps corresponding to the r consecutive adjacent video frames, establish the monitoring time analysis interval for this monitoring point;

[0141] StepU.2. Mark the prominent pixel points of the video frame and calculate the center coordinates

[0142] Select a video frame from the r video frames;

[0143] In this video frame, mark each prominent pixel point with coordinates, and record the coordinates of each prominent pixel point as (X t , Y t),where \(t = 1, 2, \ldots, v\), and \(v\) represents the number of all prominent pixel points in the video frame;

[0144] Subsequently, through calculate the central coordinates \((X z , Y z ) of all prominent pixel points in the video frame;

[0145] StepU.3. Calculate the central coordinates of prominent pixel points in multiple video frames

[0146] In the manner of StepU.2, calculate the central coordinates of all prominent pixel points in \(r\) video frames;

[0147] StepU.4. Calculate the distance differences between the central coordinates of adjacent video frames and find the average value

[0148] Select two adjacent video frames, and through the Euclidean distance formula, calculate the distance differences between the corresponding central coordinates in the two adjacent video frames;

[0149] And so on, obtain the distance differences between the corresponding central coordinates of all adjacent video frames, then calculate the average value of these multiple distance differences and mark it as the anomaly evaluation value;

[0150] StepU.5. Compare the anomaly evaluation value with the threshold and subsequent processing

[0151] Subsequently, compare the anomaly evaluation value with the preset anomaly evaluation threshold:

[0152] If the anomaly evaluation value is greater than the anomaly evaluation threshold, mark the monitoring time analysis interval of this monitoring point as the abnormal time interval, and at the same time bundle the prominent pixel points prominently marked in the substation 3D model corresponding to \(r\) consecutive adjacent video frames into an abnormal data packet;

[0153] Otherwise, do not mark the monitoring time analysis interval of this monitoring point as the abnormal time interval, and do not generate an abnormal data packet;

[0154] The information display step is also used to separately extract the abnormal data packet in the substation 3D model and display it separately through a preset sub-interface.

[0155] In this embodiment, abnormal determination is performed by analyzing abnormal events for multiple consecutive adjacent video frames corresponding to each monitoring point, generating abnormal data packets indicating the existence of abnormal events and their corresponding time intervals, which can timely detect abnormal situations in the substation and improve the reliability and security of monitoring. The central coordinates are calculated using the coordinates of prominent pixels, and then the distance difference between the central coordinates of adjacent video frames is calculated through the Euclidean distance formula, and the average value is obtained as the abnormal evaluation value for comparison with the preset threshold. This method is scientific and reasonable and can accurately judge abnormal situations. The information display step separately displays the abnormal data packets in the three-dimensional model of the substation through a preset sub-interface, enabling the operation and maintenance personnel to quickly focus on the abnormal situations and take corresponding measures in a timely manner, thus improving the efficiency of abnormal handling.

[0156] Embodiment 4

[0157] As Embodiment 4 of the present invention, in the specific implementation of this application, compared with Embodiment 1, Embodiment 2, and Embodiment 3, the technical solution of this embodiment lies in combining the solutions of the above-mentioned Embodiment 1, Embodiment 2, and Embodiment 3 for implementation.

[0158] This embodiment combines the advantages of Embodiment 1, 2, and 3, and combines and implements functions such as the three-dimensional model of the substation constructed by laser scanning technology, video acquisition and processing, video analysis and prominent marking, abnormal determination and display, etc., realizing all-round and multi-angle monitoring and management of the substation. It provides a more complete and efficient monitoring solution, greatly improving the operation safety and management efficiency of the substation.

[0159] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by software simulation of a large amount of collected data to obtain a formula closest to the actual situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0160] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claimed rights.

Claims

1. A monitoring video optimization method based on a three-dimensional model of a substation, characterized in that: The following steps are involved: Step 1: Collect video data from each monitoring point in the substation; Step 2: Obtain the spatial information and equipment geometry data of the substation, and construct a three-dimensional model of the substation containing the equipment; Step 3: normalize the video data collected at each monitoring point in the substation and obtain corresponding normalized data; Step 4: Analyze the changes in the content of each video frame in combination with the background image, determine the prominent pixel points of each video frame based on the analysis results, and make prominent marks in combination with the substation 3D model; Step 5: analyzing abnormal events based on the video frames corresponding to a plurality of consecutive adjacent video frames at each monitoring point, and generating abnormal data packets with abnormal events and their corresponding time intervals; Step 6. At each monitoring point, the current video frame corresponds to the highlighted pixel points in the substation three-dimensional model, and is displayed on the preset main interface corresponding to the substation three-dimensional model. At the same time, in the substation three-dimensional model, the abnormal data packets are extracted separately and displayed separately through a pre-set sub-interface.

2. A monitoring video optimization method based on a substation three-dimensional model according to claim 1, characterized in that: The 3D model is constructed as follows: Step 1.1, obtain the spatial information and equipment geometry data of the substation; The collected spatial information and equipment geometry data of the substation; At the same time, the location of the equipment in the substation space is collected in the form of coordinates; Step 1.2, using the spatial information and equipment geometry data of the substation, combined with the use of 3D modeling software, construct a 3D spatial model and 3D equipment model of the substation; At the same time, the 3D space model and 3D equipment model are scaled and adjusted according to the actual size of the substation and its corresponding equipment; Step 1.3, import the 3D equipment model into the 3D space model; And according to the corresponding position coordinates of the equipment in the substation space, the position of the three-dimensional equipment model in the three-dimensional space model is adjusted, and a three-dimensional substation model containing the equipment is generated.

3. The monitoring video optimization method based on the substation three-dimensional model according to claim 2 is characterized in that: The normalization process is as follows: StepK.

1. Select a monitoring point; Obtain the video data corresponding to the monitoring point, and extract all video frames from the video data; At the same time, obtain the background image corresponding to the monitoring point; Step K.2, convert all video frames and background images into grayscale images with the same resolution; StepK.3, select the grayscale image corresponding to the background image; Get the grayscale value of each pixel on the grayscale image corresponding to the background image and mark it as B i , i = 1, 2, ... n, n represents the number of pixels in the grayscale image, Bi is the grayscale value of the nth pixel in the background image; Step K.

4. Extract the grayscale value with the largest value from the grayscale values ​​of each pixel on the grayscale image corresponding to the background image and mark it as B. max , and extract the gray value with the smallest value and mark it as B min ; Then through Calculate the normalized value GBi of each pixel on the grayscale image corresponding to the background image; Step K.5, calculate the normalized value of each pixel on the grayscale image corresponding to each video frame according to the method from Step K.3 to Step K.

4.

4. The monitoring video optimization method based on the substation three-dimensional model according to claim 3 is characterized in that: The frame content change analysis method is as follows: StepL.

1. Select a video frame; Extract the normalized value of each pixel on the grayscale image corresponding to the video frame and mark it as GSi; Step L.2, compare GBi on each pixel of the background image with GSi on each pixel of the video frame one by one, and determine whether each pixel of the video frame is a prominent pixel based on the comparison result: Step L.3, then according to the shooting angle of the monitoring point, obtain the model area of ​​the corresponding angle in the three-dimensional model of the substation; The model area refers to the area corresponding to the shooting angle of the monitoring point in the three-dimensional model of the substation; Importing all prominent pixel points on the video frame into the three-dimensional model of the substation, and marking the prominent pixel points in the three-dimensional model of the substation by highlighting color marks; Step L.4, according to the method of Step L.1 to Step L.3, according to the normalized value of each pixel on the grayscale image corresponding to each video frame, determine the prominent pixel points on each video frame; At the same time, according to the time trend, all the prominent pixel points on each video frame are sequentially imported into the substation three-dimensional model, and the prominent pixel points used as prominent marks in the substation three-dimensional model are updated.

5. The monitoring video optimization method based on the substation three-dimensional model according to claim 4 is characterized in that: The comparison method in Step L.2 is as follows: Select a same pixel point corresponding to the video frame and the background image; If |GBi-GSi|>Gy, it means that the video frame and the background image have changed corresponding to the same pixel, and the pixel on the video frame is marked as a prominent pixel; If |GBi-GSi|≤Gy, it means that the video frame and the background image have not changed for the same pixel, and the pixel on the video frame is not marked as a prominent pixel; And so on, all the prominent pixels on the video frame are obtained.

6. The monitoring video optimization method based on the substation three-dimensional model according to claim 5 is characterized in that: The abnormal event analysis method is as follows: StepU.1, select a monitoring point corresponding to r consecutive adjacent video frames; At the same time, based on the timestamps corresponding to r consecutive adjacent video frames, a monitoring time analysis interval of the monitoring point is established; StepU.2, select a video frame from r video frames; In the video frame, each prominent pixel point is marked by a coordinate, and then the center coordinates of all the prominent pixel points in the video frame are calculated according to the coordinates corresponding to each prominent pixel point; Step U.3, calculate the center coordinates of all prominent pixels in r video frames according to the method of Step U.2; Step U.4, select two adjacent video frames, and calculate the distance difference between the corresponding center coordinates in the two adjacent video frames by using the Euclidean distance formula; Similarly, the distance differences between the corresponding center coordinates of all adjacent video frames are obtained, and then the average value of the multiple distance differences is calculated and marked as an abnormal evaluation value; Step U.5, then compare the abnormal assessment value with the preset abnormal assessment threshold, and determine whether the monitoring time analysis interval corresponding to the monitoring point is an abnormal time interval based on the comparison result.

7. The monitoring video optimization method based on the substation three-dimensional model according to claim 6 is characterized in that: In Step U.2, the center coordinates of all prominent pixels in the corresponding video frame are calculated as follows: The coordinates of each prominent pixel are marked as (X t , Y t ), t = 1, 2, ... v, v represents the number of all prominent pixels in the video frame; Then through Calculate the center coordinates (X z , Y z ).

8. The monitoring video optimization method based on the substation three-dimensional model according to claim 6 is characterized in that: The comparison method in Step U.5 is as follows: If the abnormal evaluation value is greater than the abnormal evaluation threshold, the monitoring time analysis interval of the monitoring point is recorded as the abnormal time interval, and the prominent pixel points corresponding to the prominent marks in the three-dimensional model of the substation in r consecutive adjacent video frames are bundled as an abnormal data packet; Otherwise, the monitoring time analysis interval of the monitoring point will not be recorded as an abnormal time interval, and no abnormal data packet will be generated.

9. The monitoring video optimization method based on the substation three-dimensional model according to claim 1 is characterized in that: in, The video data is collected by a high-definition camera at a specified frame rate.

10. The monitoring video optimization method based on the substation three-dimensional model according to claim 1 is characterized in that: in, The spatial information and equipment geometric shape data of the substation are obtained through laser scanning technology, and 3DMAX is used as the 3D modeling software.

Images