Substation safety monitoring method and system based on video image processing
Through video image processing technology, abnormal maps and personnel information in the twin space of the substation are generated, which solves the problem of low efficiency in transmission of inspection abnormal information, and realizes timely and intuitive display of substation equipment and optimizes the maintenance sequence.
Patent Information
- Application Number
- CN202410914137.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-09
- Publication Date
- 2024-11-12
Smart Images

Figure CN118944283A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to data processing technology, and in particular to a substation safety monitoring method and system based on video image processing. Background Art
[0002] Efficiently monitoring the complex equipment of the substation, understanding the abnormal operation, and taking corresponding measures in time are of great significance to reducing the occurrence of accidents and their impact. In the prior art, when inspecting various equipment, the on-duty personnel can only record the relevant abnormal information when they find a major problem with the equipment, and then report the abnormal information after the inspection is completed, resulting in low efficiency in abnormal information transmission, and the management personnel cannot intuitively view the specific abnormal problems and take maintenance measures in time, which will cause a great impact.
[0003] Therefore, how to display the monitoring data of the actual substation intuitively so that personnel can intuitively view the corresponding monitoring data images and improve processing efficiency has become an urgent problem to be solved. Summary of the invention
[0004] The embodiment of the present invention provides a substation safety monitoring method and system based on video image processing, which can be intuitively displayed according to the monitoring data of the actual substation, so that personnel can intuitively view the corresponding monitoring data images, thereby improving processing efficiency.
[0005] A first aspect of an embodiment of the present invention provides a substation safety monitoring method based on video image processing, comprising: responding to update information of a monitoring terminal, retrieving a substation twin space corresponding to the substation, wherein the substation twin space includes a plurality of inspection areas and non-inspection areas, and the monitoring terminal includes a mobile monitoring terminal and a fixed monitoring terminal; Receiving trigger information of the inspection area from the mobile monitoring terminal, retrieving a regional equipment map corresponding to the inspection area based on the trigger information, retrieving a layer update strategy to customize the regional equipment map, and generating a regional anomaly map; Acquire the workstation image collected by the fixed monitoring terminal on the non-inspection area, call up the personnel recognition strategy to identify and process the workstation image, generate abnormal personnel information, and determine the corresponding non-inspection area as the update area based on the abnormal personnel information; The substation twin space is updated according to the regional anomaly map and the updated area, and a safe display space is generated and sent to the management end.
[0006] Optionally, in a possible implementation manner of the first aspect, the receiving trigger information of the inspection area by the mobile monitoring terminal, retrieving a regional equipment map corresponding to the inspection area based on the trigger information, retrieving a layer update strategy to customize the regional equipment map, and generating a regional anomaly map, includes: Receiving trigger information of the inspection area from the mobile monitoring terminal, and retrieving a regional equipment map corresponding to the inspection area based on the trigger information; Retrieving a transparent interactive layer to be superimposed on the regional equipment map, recording the triggering trace of the transparent interactive layer by the mobile monitoring terminal, and generating an abnormal area at a corresponding position in the regional equipment map based on the triggering trace; The regional equipment map is updated according to the abnormal area to obtain a regional abnormality map.
[0007] Optionally, in a possible implementation of the first aspect, the method further includes: When it is determined that the abnormal area is a non-closed area, two end points of a line segment corresponding to the non-closed area are connected to obtain a contour screening area; Acquire pixel values in the contour screening area as screening pixel values, and extract pixel points of the corresponding regional anomaly map as screening pixel points based on the screening pixel values; Counting adjacent screening pixel points in the regional anomaly map to obtain multiple screening sets, identifying the shapes of the screening sets based on OpenCV, and obtaining screening contours corresponding to the screening sets; The screening contour where the contour screening area is located is determined as an abnormal area.
[0008] Optionally, in a possible implementation of the first aspect, the method further includes: When it is determined that the mobile monitoring terminal clicks and triggers the abnormal area, the corresponding abnormal area is used as the area to be filled, and a filling menu bar is retrieved and sent to the mobile monitoring terminal, wherein the filling menu bar includes a one-to-one correspondence between a preset pixel value and an emergency coefficient; Receive the selection information of the preset pixel value in the filling menu bar by the mobile monitoring terminal, determine the corresponding preset pixel value as the filling pixel value based on the selection information, and fill the filling pixel value into the area to be filled.
[0009] Optionally, in a possible implementation of the first aspect, acquiring the workstation image collected by the fixed monitoring terminal on the non-inspection area, calling a personnel recognition strategy to perform recognition processing on the workstation image, generating abnormal personnel information, and determining the corresponding non-inspection area as the update area based on the abnormal personnel information, includes: Retrieving a preset helmet pixel value and a preset number of personnel corresponding to the fixed monitoring terminal, and obtaining a workstation image collected by the fixed monitoring terminal for the non-inspection area; Extracting pixel points in the workstation image as helmet pixel points based on the preset helmet pixel values, and counting adjacent helmet pixel points in the workstation image to obtain a helmet pixel point set; Identify the number of people in the workstation image according to the helmet pixel point set, and generate quantity abnormality information; Based on the helmet pixel point set, personnel position recognition is performed on the workstation image to generate position abnormality information; The corresponding non-inspection area is determined as an update area based on the abnormal personnel information, and the abnormal personnel information includes abnormal quantity information and abnormal location information.
[0010] Optionally, in a possible implementation manner of the first aspect, the identifying the number of personnel in the workstation image according to the helmet pixel point set and generating quantity abnormality information includes: The number of the helmet pixel point set is obtained as the real-time number of personnel, and when it is determined that the real-time number of personnel is less than the preset number of personnel, the current moment is taken as the abnormal moment; Counting the abnormal moments within a preset time period to obtain an abnormal duration, extracting a preset duration corresponding to the preset time period, and obtaining an abnormal proportion according to a ratio of the abnormal duration to the preset duration; When it is determined that the abnormal proportion is greater than the preset proportion, quantity abnormality information is generated.
[0011] Optionally, in a possible implementation manner of the first aspect, the performing personnel position recognition on the workstation image based on the helmet pixel point set to generate position abnormality information includes: According to the ratio of the preset time length and the preset number corresponding to the preset time period, the extraction time length is obtained, and the extraction time in the preset time period is determined based on the extraction time length; Based on the extraction time, the workstation images within the preset time period are sequentially extracted to obtain the image to be verified; the image center point of the image to be verified is obtained, a coordinate system is constructed based on the image center point, and the minimum value of the horizontal coordinate, the maximum value of the horizontal coordinate, the minimum value of the vertical coordinate and the maximum value of the vertical coordinate in each helmet pixel point set in the image to be verified are obtained; the horizontal coordinate center value is obtained according to the minimum value of the horizontal coordinate and the maximum value of the horizontal coordinate, and the vertical coordinate center value is obtained according to the minimum value of the vertical coordinate and the maximum value of the vertical coordinate; Based on the central value of the horizontal coordinate and the central value of the vertical coordinate, the central coordinates corresponding to each of the helmet pixel point sets are obtained; the central coordinates are counted to obtain a coordinate point set corresponding to each of the images to be verified, and when it is determined that the intersection of the coordinate point sets is not an empty set, position anomaly information is generated.
[0012] Optionally, in a possible implementation of the first aspect, updating the substation twin space according to the regional anomaly map and the update area, generating a safe display space and sending it to the management end includes: Determine the corresponding inspection area as the image update area according to the regional anomaly map, update the regional anomaly map to the image update area, and highlight the updated area in the substation twin space to obtain an initial display space; Obtaining the number of pixels in the abnormal area in the regional abnormality map and the emergency coefficient of the abnormal area, and counting the number of abnormal areas in the regional abnormality map as the number of abnormal areas; Calculate according to the number of pixels, the emergency coefficient and the number of abnormal regions to obtain the maintenance coefficient corresponding to the abnormal map of each region; Based on the maintenance coefficient, each of the regional anomaly maps is sorted in descending order to obtain a maintenance sequence, and the regional anomaly maps in the initial display space are connected in sequence according to the order of the regional anomaly maps in the maintenance sequence to obtain a safe display space and send it to the management end.
[0013] Optionally, in a possible implementation of the first aspect, the method further includes: The management end receives the modification information of the connection lines between the regional abnormality diagrams in the safety display space, obtains the updated safety display space and sends it to the maintenance end.
[0014] A second aspect of an embodiment of the present invention provides a substation safety monitoring system based on video image processing, including: a retrieval module, used to respond to update information of a monitoring terminal, and retrieve a substation twin space corresponding to the substation, wherein the substation twin space includes multiple inspection areas and non-inspection areas, and the monitoring terminal includes a mobile monitoring terminal and a fixed monitoring terminal; A generation module, used for receiving the trigger information of the inspection area from the mobile monitoring terminal, retrieving the regional equipment map corresponding to the inspection area based on the trigger information, retrieving the layer update strategy to customize the regional equipment map, and generating a regional anomaly map; an identification module, used to obtain the workstation image collected by the fixed monitoring terminal on the non-inspection area, call the personnel identification strategy to identify and process the workstation image, generate abnormal personnel information, and determine the corresponding non-inspection area as the update area based on the abnormal personnel information; An update module is used to update the substation twin space according to the regional anomaly map and the update area, generate a safe display space and send it to the management end.
[0015] The beneficial effects of the present invention are as follows: 1. The present invention can be intuitively displayed based on the monitoring data of the actual substation, so that personnel can intuitively view the corresponding monitoring data image and improve processing efficiency. Among them, the present invention can obtain the corresponding regional abnormality map, which is convenient for subsequent updating at the corresponding position in the twin space, so that subsequent management personnel can intuitively view the corresponding abnormal equipment, so as to promptly take appropriate maintenance personnel for timely maintenance. At the same time, the present invention can also use different pixel values for differentiated display according to the emergency situation of abnormal equipment maintenance, so that personnel can more intuitively determine the equipment processing order of the corresponding area.
[0016] 2. The present invention can obtain the workstation image corresponding to the non-inspection area through a fixed monitoring terminal, and use the personnel identification strategy to identify and process the workstation image, generate abnormal personnel information, thereby determining the corresponding update area, and displaying it in the corresponding substation twin space, so that the management personnel can intuitively view the corresponding abnormal workstation personnel information, and thus promptly issue warnings to the corresponding personnel. In addition, the present invention can also determine the priority maintenance order of the equipment in the corresponding area of the regional abnormality map according to the maintenance coefficient, and intuitively display it in the safe display space through sequential connection, so that the management personnel can promptly arrange maintenance personnel to repair the relevant equipment in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 A flowchart of a substation safety monitoring method based on video image processing provided by the present invention; Figure 2 This is a structural schematic diagram of a substation safety monitoring system based on video image processing provided by the present invention. DETAILED DESCRIPTION
[0018] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0019] The terms "first", "second", "third", "fourth", etc. (if any) in the specification and claims of the present invention and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in sequences other than those illustrated or described herein.
[0020] It should be understood that in various embodiments of the present invention, the size of the sequence number of each process does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.
[0021] It should be understood that in the present invention, "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.
[0022] It should be understood that in the present invention, "plurality" refers to two or more than two. "And / or" is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "Contains A, B and C", "Contains A, B, C" means that A, B, and C are all included, "Contains A, B or C" means that one of A, B, and C is included, and "Contains A, B and / or C" means that any one, any two, or any three of A, B, and C are included.
[0023] It should be understood that in the present invention, "B corresponding to A", "B corresponding to A", "A corresponds to B" or "B corresponds to A" means that B is associated with A and B can be determined based on A. Determining B based on A does not mean determining B based only on A, but B can also be determined based on A and / or other information. A and B match when the similarity between A and B is greater than or equal to a preset threshold.
[0024] Depending on the context, "if" as used herein may be interpreted as "when" or "when" or "in response to determining" or "in response to detecting."
[0025] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.
[0026] like Figure 1 As shown, the present invention provides a substation safety monitoring method based on video image processing, comprising steps S1-S4: S1, respond to the update information of the monitoring terminal, call the substation twin space corresponding to the substation, the substation twin space includes multiple inspection areas and non-inspection areas, and the monitoring terminal includes a mobile monitoring terminal and a fixed monitoring terminal.
[0027] It should be noted that in order to ensure the safe operation of the substation, each area of the substation will be monitored in real time. Therefore, when an abnormal situation is found in the corresponding monitoring area during the inspection or monitoring process, the monitoring end will send corresponding update information so that personnel can intuitively view the corresponding abnormal data through the substation twin space.
[0028] In addition, the areas in the substation where electrical equipment is usually stored need to be inspected regularly to check for any abnormalities in the appearance of the equipment, such as whether the color has changed, whether there are any foreign objects, whether the needle indication is normal, whether the sound of the equipment is normal, and the changes in the operating parameters of the electrical equipment during operation, etc., so as to determine whether the operating status of the equipment is normal. At the same time, the on-duty personnel in the duty room or office corresponding to the substation will be monitored to determine whether the personnel are on duty, so as to ensure the normal operation of the substation and reduce the occurrence of accidents. Therefore, the safety monitoring method for different areas divides the generated twin space into different inspection areas and non-inspection areas.
[0029] Among them, the substation twin space is a virtual twin space established based on the actual substation. The inspection area is the area in the substation that needs to be inspected, such as the distribution room area. The non-inspection area is the area in the substation that does not need to be inspected, such as the office staff area or the on-duty staff area in the monitoring room of the substation. The mobile monitoring end is the terminal corresponding to the inspection personnel, which can be a mobile phone. The fixed monitoring end is a monitoring end with a fixed position, which can be a camera.
[0030] Through the above implementation mode, the present invention can update information in the twin space, so as to facilitate the subsequent intuitive display of abnormal data to personnel through the twin space.
[0031] S2, receiving the trigger information of the inspection area from the mobile monitoring terminal, retrieving the regional equipment map corresponding to the inspection area based on the trigger information, retrieving the layer update strategy to customize the regional equipment map, and generating a regional anomaly map.
[0032] It should be noted that in order to facilitate management personnel to view abnormal information in a timely manner and quickly determine the abnormal problems of corresponding equipment, so as to arrange suitable maintenance personnel to perform maintenance in a timely manner and improve maintenance efficiency, therefore, when the inspection personnel find that the corresponding equipment has an abnormality during the inspection process, due to the large number of equipment in the corresponding inspection area, in order to provide accurate abnormal equipment for display, the abnormal equipment will be circled and confirmed in the equipment map of the corresponding area, thereby extracting the image of the corresponding abnormal equipment, so that the management personnel can intuitively view the corresponding abnormal equipment information.
[0033] The trigger information is the trigger click operation information on the mobile monitoring terminal, and the regional equipment map is the equipment image in the corresponding area.
[0034] Through the above implementation, the present invention obtains the corresponding regional anomaly map, which is convenient for subsequent updating at the corresponding position in the twin space, so that subsequent management personnel can intuitively view the corresponding abnormal equipment.
[0035] In some embodiments, step S2 (receiving the trigger information of the inspection area from the mobile monitoring terminal, retrieving the regional equipment map corresponding to the inspection area based on the trigger information, retrieving the layer update strategy to customize the regional equipment map, and generating a regional abnormality map) includes steps S21-S23: S21, receiving trigger information of the inspection area from the mobile monitoring terminal, and retrieving a regional equipment map corresponding to the inspection area based on the trigger information.
[0036] It should be noted that in an actual substation, different equipment has different installation areas. For example, a substation has a high-voltage distribution room, a high-voltage capacitor room, etc., and correspondingly there are different inspection areas in the twin space.
[0037] It is not difficult to understand that when the inspection personnel find an abnormal situation when inspecting equipment in the high-voltage distribution room, the corresponding inspection area will be triggered on the corresponding mobile monitoring terminal, so as to retrieve the regional equipment map of the corresponding area, so as to facilitate the subsequent identification of the corresponding abnormal equipment.
[0038] Through the above implementation, the present invention obtains the regional equipment map corresponding to the inspection area, which is convenient for subsequent determination of specific abnormal equipment images.
[0039] S22, calling up a transparent interactive layer to be superimposed on the area equipment map, recording the triggering trace of the transparent interactive layer by the mobile monitoring terminal, and generating an abnormal area at a corresponding position in the area equipment map based on the triggering trace.
[0040] It can be understood that by superimposing the retrieved transparent interactive layer on top of the regional equipment map, personnel can circle the corresponding location area of the abnormal equipment on the transparent interactive layer, thereby recording the triggering traces of the transparent interactive layer by the mobile monitoring terminal. Therefore, an abnormal area is generated at the corresponding position in the regional equipment map based on the triggering traces.
[0041] Among them, the transparent interactive layer is a layer for realizing the interaction between personnel and images, which can be a transparent layer. The trigger trace is the movement trace triggered by circling the transparent interactive layer, and the abnormal area is the area corresponding to the equipment where the abnormality occurs in the regional equipment map.
[0042] Through the above implementation, the present invention generates a corresponding abnormal area, which is convenient for subsequent updating of the position corresponding to the abnormal area, thereby intuitively displaying the corresponding abnormal equipment to personnel.
[0043] S23, updating the regional equipment map according to the abnormal area to obtain a regional abnormality map.
[0044] Among them, the regional anomaly map is an updated image of the regional equipment map.
[0045] For example, the position corresponding to the abnormal area in the regional equipment map is highlighted and updated, thereby obtaining a regional abnormality map with updated information.
[0046] It should be noted that when the inspection personnel trigger and circle the transparent interactive layer covered on the regional equipment map, when the circled area is an abnormal part of the corresponding equipment, and because the circle is not standardized, the recorded trigger traces do not form a closed area, so further processing will be carried out to update and display the area where the corresponding equipment is located as an abnormal area. Therefore, the specific implementation steps also include A1-A4: A1, when it is determined that the abnormal area is a non-closed area, two end points of a line segment corresponding to the non-closed area are connected to obtain a contour screening area.
[0047] It can be understood that when the trigger trace is not a closed circle or box, and the corresponding area is judged to be a non-closed area, the two endpoints of the corresponding area segment in the non-closed area are connected to form a complete closed line to complete the circling of the area, and then the corresponding contour screening area is obtained.
[0048] Among them, the non-closed area is the area surrounded by the unclosed line segment, the area line segment is the circle line corresponding to the circled non-closed area, and the contour screening area is the area circled by connecting the two ends of the area line segment.
[0049] Through the above implementation, the present invention obtains the corresponding contour screening area, which is convenient for subsequently determining the corresponding abnormal area.
[0050] A2, obtaining pixel values in the contour screening area as screening pixel values, and extracting pixel points of the corresponding regional anomaly map as screening pixel points based on the screening pixel values.
[0051] For example: when the corresponding abnormal part in the contour screening area is a transformer coil, and the pixel value of the transformer coil is red, red is used as the screening pixel value, and the pixel point corresponding to the red pixel value in the regional abnormality map is used as the screening pixel point.
[0052] A3, counting adjacent screening pixel points in the regional anomaly map to obtain multiple screening sets, identifying the shapes of the screening sets based on OpenCV, and obtaining screening contours corresponding to the screening sets.
[0053] It should be noted that there may be multiple identical devices in the same inspection area, and therefore, there may also be multiple sets of pixel points determined based on the pixel values corresponding to the devices.
[0054] The screening set is a set of adjacent screening pixels, and the screening contour is a contour corresponding to the shape formed by the screening set.
[0055] Through the above implementation, the present invention obtains the corresponding screening contour, which is convenient for subsequently determining the corresponding abnormal area, A4, determining the screening contour where the contour screening area is located as the abnormal area.
[0056] It can be understood that the screening contour where the contour screening area is located is determined as the abnormal area.
[0057] It is not difficult to understand that among the numerous screening contours, only one corresponding screening contour contains the contour screening area, and therefore, the area corresponding to the screening contour containing the contour screening area is regarded as the abnormal area.
[0058] Through the above implementation, the present invention determines the corresponding abnormal contour, thereby facilitating subsequent display updates in the corresponding abnormal area, and facilitating personnel to intuitively view the corresponding abnormal equipment.
[0059] It should be noted that in order to enable the management personnel to visually view the corresponding abnormal equipment through the image, the abnormal area corresponding to the abnormal equipment will be highlighted according to the maintenance emergency of the abnormal equipment. The specific implementation steps include B1-B2: B1, when it is determined that the mobile monitoring terminal clicks and triggers the abnormal area, the corresponding abnormal area is used as the area to be filled, and a filling menu bar is retrieved and sent to the mobile monitoring terminal, wherein the filling menu bar includes a one-to-one correspondence between preset pixel values and emergency coefficients.
[0060] Among them, the filling menu bar includes preset pixel values and emergency coefficients corresponding to the preset pixel values, which are preset manually. The preset pixel values are preset pixel values, which can be red, orange, yellow, etc. The emergency coefficient is a value for emergency maintenance of abnormal equipment, which can be 1, 2, 3. The larger the value, the more urgent it is, and priority maintenance is required.
[0061] B2, receiving the selection information of the preset pixel value in the filling menu bar from the mobile monitoring terminal, determining the corresponding preset pixel value as the filling pixel value based on the selection information, and filling the filling pixel value into the area to be filled.
[0062] The selected information is the result information of the selected click operation.
[0063] For example, when a person selects the preset pixel value of red in the fill menu bar on the mobile monitoring terminal, red is used as the fill pixel value, and the red is filled in the corresponding area to be filled.
[0064] Through the above implementation, the present invention completes the filling of preset pixel values of the area to be filled, which is convenient for subsequent intuitive display to personnel through the pixel values in the abnormal area, so that management personnel can quickly take maintenance measures for abnormal equipment.
[0065] S3, obtaining the workstation image collected by the fixed monitoring terminal in the non-inspection area, calling the personnel recognition strategy to identify and process the workstation image, generating abnormal personnel information, and determining the corresponding non-inspection area as the update area based on the abnormal personnel information.
[0066] It should be noted that when monitoring the on-duty personnel on duty, it is necessary to obtain the workstation image through the camera in the office area, and process the workstation image to determine whether the on-duty personnel have any abnormal conditions on duty, such as being away from the workstation for a long time. Therefore, the corresponding update area can be determined based on the abnormal personnel information and displayed intuitively to the management personnel.
[0067] The workstation image is an image of the workstation corresponding to the non-inspection area.
[0068] Through the above implementation, the present invention determines the corresponding update area so that the management personnel can view it intuitively.
[0069] In some embodiments, step S3 (obtaining the workstation image collected by the fixed monitoring terminal for the non-inspection area, calling the personnel identification strategy to identify and process the workstation image, generating abnormal personnel information, and determining the corresponding non-inspection area as the update area based on the abnormal personnel information) includes steps S31-S35: S31, retrieve the preset helmet pixel value and the preset number of personnel corresponding to the fixed monitoring terminal, and obtain the workstation image collected by the fixed monitoring terminal for the non-inspection area.
[0070] Among them, the preset helmet pixel value is the pixel value of the personnel helmet, which can be white, yellow, etc., and is pre-set manually based on the actual color of the personnel helmet. The preset number of personnel is the preset number of personnel in the corresponding non-inspection area.
[0071] Through the above implementation, the present invention obtains the workstation image, and then determines the personnel information by subsequently recognizing the workstation image.
[0072] S32, extracting pixel points in the workstation image as helmet pixel points based on the preset helmet pixel values, and counting adjacent helmet pixel points in the workstation image to obtain a helmet pixel point set.
[0073] The helmet pixel point set is a set of adjacent helmet pixel points.
[0074] For example, when the preset helmet pixel value is yellow, the pixel points corresponding to the yellow pixel value in the workstation image are extracted, and the extracted pixel points are helmet pixel points.
[0075] Through the above implementation, the present invention obtains the corresponding helmet pixel point set, which is convenient for subsequently determining the corresponding number of personnel according to the number of the helmet pixel point set.
[0076] S33, identifying the number of personnel in the workstation image according to the helmet pixel point set, and generating quantity abnormality information.
[0077] It can be understood that the number of personnel in the workstation image is identified based on the helmet pixel point set to generate quantity abnormality information. The specific operation steps are as follows.
[0078] In some embodiments, step S33 (recognizing the number of personnel in the workstation image according to the helmet pixel point set and generating abnormal number information) includes steps S331-S333: S331, obtaining the number of the helmet pixel point set as the real-time number of personnel, and when it is determined that the real-time number of personnel is less than the preset number of personnel, taking the current moment as an abnormal moment.
[0079] It should be noted that it is usually required that personnel wear safety helmets while on duty, so the number of personnel on duty can be determined based on the number of helmet pixel point sets.
[0080] The real-time number of personnel refers to the number of personnel at the workstation at the current moment.
[0081] It is not difficult to understand that when the real-time number of people is less than the preset number of people, it means that someone has left his or her workstation in the office area. Therefore, the current moment is taken as the abnormal moment, which makes it easier to determine the length of time the person at the corresponding location is away. When the departure time is long, the corresponding quantity abnormality information is generated.
[0082] S332, counting the abnormal moments within the preset time period to obtain an abnormal duration, extracting a preset duration corresponding to the preset time period, and obtaining an abnormal proportion according to a ratio of the abnormal duration to the preset duration.
[0083] It should be noted that in order to avoid the generation of abnormal information of personnel due to the personnel leaving the workstation for a short time for fetching water, going to the toilet, etc., the present invention sets a preset duration in advance. And the abnormal proportion is obtained according to the ratio of the abnormal duration to the preset duration, and the abnormal proportion is used to facilitate the subsequent judgment of whether the corresponding abnormal information needs to be generated.
[0084] Among them, the preset time period is a time period set manually in advance, for example, 14:00-15:00, the abnormal duration is the duration composed of abnormal moments, for example, 1 hour, the preset duration is a time value set manually in advance, for example, 30 minutes, and the abnormal proportion is the ratio of the abnormal duration to the preset duration, for example, the abnormal proportion is 2 / 1.
[0085] S333: When it is determined that the abnormal proportion is greater than the preset proportion, quantity abnormality information is generated.
[0086] It should be noted that when the ratio of the abnormal time when the corresponding personnel is absent from the workstation to the preset time is large, it means that the corresponding personnel is absent from the workstation for a long time. Therefore, corresponding abnormal information needs to be generated to remind the relevant personnel of the abnormality.
[0087] Among them, the preset proportion is a pre-set proportion, which may be 1 / 2, and the abnormal quantity information is information about the abnormal number of personnel.
[0088] Through the above implementation mode, the present invention obtains quantity abnormality information, which is convenient for subsequent updating of the substation twin space, intuitively displaying it to personnel, and reminding relevant personnel of abnormalities.
[0089] S34, identifying the position of a person in the workstation image based on the helmet pixel point set, and generating position abnormality information.
[0090] It should be noted that when personnel are working at their workstations, they usually move to a certain extent at their workstations and cannot remain still for a long time. Therefore, when the helmet pixel point set is processed and judged, if it is determined that the personnel at the corresponding position is still, corresponding position abnormality information will be generated to prompt the management personnel.
[0091] In some embodiments, step S34 (recognizing the number of personnel in the workstation image according to the helmet pixel point set and generating abnormal number information) includes steps S341-S346: S341, obtaining an extraction duration according to a ratio of a preset duration corresponding to the preset time period to a preset number, and determining an extraction time in the preset time period based on the extraction duration.
[0092] It should be noted that, since there are a lot of image data within the preset time period, several image data corresponding to different moments will be randomly selected from the numerous image data for processing, so as to complete the judgment of abnormal information while reducing the amount of data processing.
[0093] Among them, the preset number is a number set in advance by humans, which may be 3, and the extraction duration is the time duration for extracting images. For example, when the preset duration is 30 minutes and the preset number is 3, the extraction duration is 10 minutes, which means that a workstation image is extracted every 10 minutes in the preset time period, and the extraction time is the time value corresponding to the extracted image, for example, 14:10, 14:20, etc.
[0094] S342: extract the workstation images within the preset time period in sequence based on the extraction time to obtain images to be verified.
[0095] The image to be verified is a workstation image that is extracted and waiting to be processed.
[0096] S343, obtaining the image center point of the image to be verified, constructing a coordinate system based on the image center point, and obtaining the minimum value of the horizontal coordinate, the maximum value of the horizontal coordinate, the minimum value of the vertical coordinate, and the maximum value of the vertical coordinate in each set of helmet pixel points in the image to be verified.
[0097] Among them, the image center point is the center point of the corresponding image.
[0098] It is not difficult to understand that by constructing a coordinate system, the minimum value of the horizontal coordinate, the maximum value of the horizontal coordinate, the minimum value of the vertical coordinate and the maximum value of the vertical coordinate of each helmet pixel point set are obtained, so that it is convenient to obtain the coordinates of the center point in the corresponding helmet pixel point set later, and then compare and judge with the coordinates of the center point in the helmet pixel point set in the image corresponding to other moments. When the coordinates are consistent, it means that there is no person at the position, but the corresponding helmet is placed on the workstation, or the person has an abnormal condition, so it is necessary to generate abnormal information.
[0099] S344, obtaining a central value of the horizontal coordinate according to the minimum value of the horizontal coordinate and the maximum value of the horizontal coordinate, and obtaining a central value of the vertical coordinate according to the minimum value of the vertical coordinate and the maximum value of the vertical coordinate.
[0100] The central value of the abscissa is the abscissa of the central point of each helmet pixel point set, and the central value of the ordinate is the ordinate of the central point of each helmet pixel point set.
[0101] S345: Obtain center coordinates corresponding to each of the helmet pixel point sets based on the abscissa center value and the ordinate center value.
[0102] The center coordinates are the coordinates of the center point of each helmet pixel set.
[0103] S346, counting the center coordinates to obtain a set of coordinate points corresponding to each of the images to be verified, and generating position anomaly information when it is determined that the intersection of the set of coordinate points is not an empty set.
[0104] It can be understood that the central coordinates corresponding to the multiple helmet pixel point sets in each image to be verified are counted to obtain multiple coordinate point sets, and then the intersection judgment is performed on the coordinate point sets corresponding to the multiple images to be verified. When it is judged that the intersection is not an empty set, the position abnormality information is generated.
[0105] It is not difficult to understand that when the intersection is not an empty set, it indicates that the position of the employee remains consistent in the workstation images corresponding to multiple time periods, which means that the employee remains stationary at different time periods corresponding to multiple images with consistent coordinates. Therefore, it can be determined that an abnormal situation has occurred, and position abnormality information is generated based on the corresponding workstation position.
[0106] S35, determining the corresponding non-inspection area as the update area based on the abnormal personnel information, wherein the abnormal personnel information includes abnormal quantity information and abnormal location information.
[0107] Through the above implementation mode, the present invention obtains the corresponding update area, which is convenient for the subsequent update of the substation twin space, thereby generating a safe display space for intuitive display to personnel.
[0108] S4, updating the substation twin space according to the regional anomaly map and the updated area, generating a safe display space and sending it to the management end.
[0109] Among them, the safe display space is a substation twin space that can intuitively display updated information, and the management end is the terminal corresponding to the management personnel, which can be a computer.
[0110] Through the above implementation, the present invention sends the safe display space to the management end, so that the management personnel can intuitively view the corresponding updated abnormal information on the management end, and thus take corresponding measures in time.
[0111] In some embodiments, step S4 (updating the substation twin space according to the regional anomaly map and the updated area, generating a safe display space and sending it to the management end) includes steps S41-S44: S41, determining the corresponding inspection area as the image update area according to the regional anomaly map, updating the regional anomaly map to the image update area, and highlighting the updated area in the substation twin space to obtain an initial display space.
[0112] S42, obtaining the number of pixels in the abnormal area in the regional abnormality map and the emergency coefficient of the abnormal area, and counting the number of abnormal areas in the regional abnormality map as the number of abnormal areas.
[0113] It should be noted that when multiple devices have abnormalities, the maintenance of the devices will have a certain priority, so the maintenance will be prioritized according to the emergency coefficient in the abnormal area and the number of abnormalities.
[0114] It is not difficult to understand that the more pixels there are in the abnormal area, the greater the abnormal problem of the corresponding device. Moreover, the more abnormal areas there are, the more abnormal devices there are in the corresponding device area. In order to improve maintenance efficiency, the abnormal devices in the corresponding area will be repaired first.
[0115] S43, calculating according to the number of pixel points, the emergency coefficient and the number of abnormal regions to obtain a maintenance coefficient corresponding to each of the abnormal regions.
[0116] It is not difficult to understand that the priority of maintenance is affected by many factors. Therefore, the corresponding maintenance coefficient will be calculated according to the weights of various factors to obtain a more accurate priority maintenance area. For example, when the number of pixels is X, the emergency coefficient is Y, and the number of abnormal areas is Z, K1 is 20%, K2 is 50%, and K3 is 30%. Therefore, K1×X+K2×Y+K3×Z=A, where A is the maintenance coefficient.
[0117] S44, sorting each of the regional anomaly maps in descending order based on the maintenance coefficient to obtain a maintenance sequence, and connecting the regional anomaly maps in the initial display space in sequence according to the order of the regional anomaly maps in the maintenance sequence to obtain a safe display space and send it to the management end.
[0118] It is not difficult to understand that the regional anomaly maps in the initial display space are connected in sequence according to the maintenance sequence, so that when personnel view the safe display space, they can intuitively view the priority maintenance order of the corresponding areas, so as to make corresponding arrangements and measures in time, carry out timely maintenance, and reduce the impact of equipment anomalies.
[0119] It should be noted that when the equipment in the area that is prioritized for maintenance requires more maintenance personnel to coordinate maintenance, but the current number of maintenance personnel is insufficient, the management personnel can actively modify the connection order of the regional anomaly diagram so that the maintenance order can be more consistent with the current maintenance schedule and improve the equipment maintenance efficiency. Therefore, the implementation steps also include C1: C1, receiving the change information of the connection lines between the regional abnormality diagrams in the safety display space from the management end, obtaining the updated safety display space and sending it to the maintenance end.
[0120] Among them, the change information is the sequential connection between the regional abnormality maps, and the maintenance end is the terminal corresponding to the maintenance personnel, which can be a mobile phone.
[0121] Through the above implementation, maintenance personnel can promptly and quickly repair equipment problems that can be solved.
[0122] like Figure 2 FIG. 1 is a schematic diagram of a structure of a substation safety monitoring system based on video image processing provided by an embodiment of the present invention, the system comprising: A calling module, used to respond to the update information of the monitoring terminal, call the substation twin space corresponding to the substation, the substation twin space includes multiple inspection areas and non-inspection areas, and the monitoring terminal includes a mobile monitoring terminal and a fixed monitoring terminal; A generation module, used for receiving the trigger information of the inspection area from the mobile monitoring terminal, retrieving the regional equipment map corresponding to the inspection area based on the trigger information, retrieving the layer update strategy to customize the regional equipment map, and generating a regional anomaly map; an identification module, used to obtain the workstation image collected by the fixed monitoring terminal on the non-inspection area, call the personnel identification strategy to identify and process the workstation image, generate abnormal personnel information, and determine the corresponding non-inspection area as the update area based on the abnormal personnel information; An update module is used to update the substation twin space according to the regional anomaly map and the update area, generate a safe display space and send it to the management end.
[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A substation safety monitoring method based on video image processing, characterized in that: include: In response to the update information of the monitoring terminal, a substation twin space corresponding to the substation is retrieved, wherein the substation twin space includes a plurality of inspection areas and non-inspection areas, and the monitoring terminal includes a mobile monitoring terminal and a fixed monitoring terminal; Receiving trigger information of the inspection area from the mobile monitoring terminal, retrieving a regional equipment map corresponding to the inspection area based on the trigger information, retrieving a layer update strategy to customize the regional equipment map, and generating a regional anomaly map; Acquire the workstation image collected by the fixed monitoring terminal on the non-inspection area, call up the personnel recognition strategy to identify and process the workstation image, generate abnormal personnel information, and determine the corresponding non-inspection area as the update area based on the abnormal personnel information; The substation twin space is updated according to the regional anomaly map and the updated area, and a safe display space is generated and sent to the management end.
2. The method according to claim 1, characterized in that: The receiving trigger information of the inspection area by the mobile monitoring terminal, retrieving the regional equipment map corresponding to the inspection area based on the trigger information, retrieving the layer update strategy to customize the regional equipment map, and generating a regional abnormality map, including: Receiving trigger information of the inspection area from the mobile monitoring terminal, and retrieving a regional equipment map corresponding to the inspection area based on the trigger information; Retrieving a transparent interactive layer to be superimposed on the regional equipment map, recording the triggering trace of the transparent interactive layer by the mobile monitoring terminal, and generating an abnormal area at a corresponding position in the regional equipment map based on the triggering trace; The regional equipment map is updated according to the abnormal area to obtain a regional abnormality map.
3. The method according to claim 2, characterized in that Also includes: When it is determined that the abnormal area is a non-closed area, two end points of a line segment corresponding to the non-closed area are connected to obtain a contour screening area; Acquire pixel values in the contour screening area as screening pixel values, and extract pixel points of the corresponding regional anomaly map as screening pixel points based on the screening pixel values; Counting adjacent screening pixel points in the regional anomaly map to obtain multiple screening sets, identifying the shapes of the screening sets based on OpenCV, and obtaining screening contours corresponding to the screening sets; The screening contour where the contour screening area is located is determined as an abnormal area.
4. The method according to claim 2 or 3, characterized in that: Also includes: When it is determined that the mobile monitoring terminal clicks and triggers the abnormal area, the corresponding abnormal area is used as the area to be filled, and a filling menu bar is retrieved and sent to the mobile monitoring terminal, wherein the filling menu bar includes a one-to-one correspondence between a preset pixel value and an emergency coefficient; Receive the selection information of the preset pixel value in the filling menu bar by the mobile monitoring terminal, determine the corresponding preset pixel value as the filling pixel value based on the selection information, and fill the filling pixel value into the area to be filled.
5. The method according to claim 4, characterized in that The step of acquiring the workstation image collected by the fixed monitoring terminal on the non-inspection area, calling a personnel identification strategy to identify and process the workstation image, generating abnormal personnel information, and determining the corresponding non-inspection area as an update area based on the abnormal personnel information includes: Retrieving a preset helmet pixel value and a preset number of personnel corresponding to the fixed monitoring terminal, and obtaining a workstation image collected by the fixed monitoring terminal for the non-inspection area; Extracting pixel points in the workstation image as helmet pixel points based on the preset helmet pixel values, and counting adjacent helmet pixel points in the workstation image to obtain a helmet pixel point set; Identify the number of people in the workstation image according to the helmet pixel point set, and generate quantity abnormality information; Based on the helmet pixel point set, personnel position recognition is performed on the workstation image to generate position abnormality information; The corresponding non-inspection area is determined as an update area based on the abnormal personnel information, and the abnormal personnel information includes abnormal quantity information and abnormal location information.
6. The method according to claim 5, characterized in that The identifying the number of personnel in the workstation image according to the helmet pixel point set and generating the number abnormality information includes: obtaining the number of the helmet pixel point set as the real-time number of personnel, and when it is determined that the real-time number of personnel is less than the preset number of personnel, taking the current moment as the abnormal moment; Counting the abnormal moments within a preset time period to obtain an abnormal duration, extracting a preset duration corresponding to the preset time period, and obtaining an abnormal proportion according to a ratio of the abnormal duration to the preset duration; When it is determined that the abnormal proportion is greater than the preset proportion, quantity abnormality information is generated.
7. The method according to claim 5, characterized in that The method of identifying the position of a person on the workstation image based on the set of helmet pixels to generate position abnormality information includes: obtaining an extraction time length according to a ratio of a preset time length and a preset number corresponding to a preset time period, and determining an extraction time in the preset time period based on the extraction time length; Sequentially extracting the workstation images within the preset time period based on the extraction time to obtain an image to be verified; Obtain the image center point of the image to be verified, construct a coordinate system based on the image center point, obtain the minimum value of the horizontal coordinate, the maximum value of the horizontal coordinate, the minimum value of the vertical coordinate and the maximum value of the vertical coordinate in each set of helmet pixel points in the image to be verified; obtain the central value of the horizontal coordinate according to the minimum value of the horizontal coordinate and the maximum value of the horizontal coordinate, and obtain the central value of the vertical coordinate according to the minimum value of the vertical coordinate and the maximum value of the vertical coordinate; Based on the central value of the horizontal coordinate and the central value of the vertical coordinate, the central coordinates corresponding to each of the helmet pixel point sets are obtained; the central coordinates are counted to obtain a coordinate point set corresponding to each of the images to be verified, and when it is determined that the intersection of the coordinate point sets is not an empty set, position anomaly information is generated.
8. The method according to claim 7, characterized in that The updating of the substation twin space according to the regional anomaly map and the update area, generating a safe display space and sending it to the management end, includes: Determine the corresponding inspection area as the image update area according to the regional anomaly map, update the regional anomaly map to the image update area, and highlight the updated area in the substation twin space to obtain an initial display space; Obtaining the number of pixels in the abnormal area in the regional abnormality map and the emergency coefficient of the abnormal area, and counting the number of abnormal areas in the regional abnormality map as the number of abnormal areas; Calculate according to the number of pixels, the emergency coefficient and the number of abnormal regions to obtain the maintenance coefficient corresponding to the abnormal map of each region; Based on the maintenance coefficient, each of the regional anomaly maps is sorted in descending order to obtain a maintenance sequence, and the regional anomaly maps in the initial display space are connected in sequence according to the order of the regional anomaly maps in the maintenance sequence to obtain a safe display space and send it to the management end.
9. The method according to claim 8, characterized in that Also includes: The management end receives the modification information of the connection lines between the regional abnormality diagrams in the safety display space, obtains the updated safety display space and sends it to the maintenance end.
10. The substation safety monitoring system based on video image processing is characterized by: include: A calling module, used to respond to the update information of the monitoring terminal, call the substation twin space corresponding to the substation, the substation twin space includes multiple inspection areas and non-inspection areas, and the monitoring terminal includes a mobile monitoring terminal and a fixed monitoring terminal; A generation module, used for receiving the trigger information of the inspection area from the mobile monitoring terminal, retrieving the regional equipment map corresponding to the inspection area based on the trigger information, retrieving the layer update strategy to customize the regional equipment map, and generating a regional anomaly map; an identification module, used to obtain the workstation image collected by the fixed monitoring terminal on the non-inspection area, call the personnel identification strategy to identify and process the workstation image, generate abnormal personnel information, and determine the corresponding non-inspection area as the update area based on the abnormal personnel information; An update module is used to update the substation twin space according to the regional anomaly map and the update area, generate a safe display space and send it to the management end.
