Power transformation non-inductive access method and system based on communication mutual recognition mechanism

Abstract

The embodiment of the invention provides a power transformation non-inductive access method and system based on a communication mutual recognition mechanism, and belongs to the technical field of transformer substation detection. The method comprises the following steps: when a worker enters a set range before the worker enters a transformer substation, obtaining an action video about the worker; performing frame extraction on the action video to obtain an image containing the face of the worker; preprocessing the image containing the face of the worker to obtain a face region in the image; identifying a face region in the image to obtain information about a worker; under the condition that the identified information of the worker is determined to be the worker belonging to the transformer substation, allowing the worker to enter the transformer substation, and continuing to obtain a video about the worker in the transformer substation; and analyzing the video about the worker in the transformer substation, and identifying the violation condition of the worker in the transformer substation. According to the method, the access personnel of the transformer substation can be accurately identified, and violation identification can be carried out.

Description

Technical Field

[0001] The present invention relates to the technical field of substation detection, and in particular to a method and system for non-sensing access to substations based on a mutual recognition mechanism. Background Art

[0002] State Grid's common safety management and control scenarios for worksites face the following major challenges: First, traditional verification methods are outdated. Safety access verification for worksite personnel relies on traditional methods and technical means, such as manual verification or fixed gates. This results in cumbersome operations, lengthy verification times, and missed personnel. Second, there is a lack of a dynamic joint control mechanism. Existing security management methods, such as gates, struggle to aggregate multi-dimensional, full-scale access business data, leading to prominent issues such as delayed information synchronization and missing data dimensions. Therefore, a seamless access method is needed to accurately identify substation access personnel and identify violations. Summary of the Invention

[0003] The purpose of the embodiments of the present invention is to provide a method and system for non-sensing access to substations based on a mutual recognition mechanism, which can accurately identify access personnel to substations and identify violations.

[0004] To achieve the above objectives, an embodiment of the present invention provides a method for non-sensing access to substations based on a mutual recognition mechanism, the method comprising: When a worker enters a set range before entering the substation, obtaining a video of the worker's actions; Extracting frames from the action video to obtain an image containing the staff member's face; Preprocessing the image containing the staff member's face to obtain a face region in the image; Recognizing a facial region in the image to obtain information about the staff member; If it is determined that the identified worker is a worker of the substation, the worker is allowed to enter the substation and the video of the worker in the substation is continuously obtained; Analyze videos of workers in the substation and identify violations of regulations by the workers in the substation.

[0005] Optionally, preprocessing the image containing the staff member's face to obtain a face region in the image includes: Acquiring an image containing the staff member's face and performing brightness compensation on the image; Converting the brightness-compensated image from RGB format to HSI format; Performing skin color segmentation on the image converted into the HSI format according to the HSI face skin color segmentation standard to obtain a binary image; Performing smoothing filtering on the binary image to obtain a smoothed skin area; Filling in the missing parts of the skin area, and performing shape screening on the complete skin area to remove areas with irregular shapes or unsatisfactory aspect ratios; Perform template matching on the filtered skin area to obtain the initial face area; The initial face region is confirmed according to geometric rules to obtain a confirmed face region in the image.

[0006] Optionally, obtaining an image containing the staff member's face and performing brightness compensation on the image includes: Acquire an image containing a staff member's face, obtain brightness values ​​in the image, and obtain a maximum value among the brightness values; The image is brightness compensated according to formula (1): Formula (1), in, Indicates the pixel after brightness compensation The brightness value, Indicates the pixel before brightness compensation The brightness value, Indicates the maximum value of the brightness value before brightness compensation. Indicates the logarithmic mean of the brightness values ​​before brightness compensation.

[0007] Optionally, the logarithmic mean of the brightness value before brightness compensation is obtained according to formula (2): Formula (2), in, Indicates the size of the image, Indicates compensation parameters.

[0008] Optionally, the image converted into the HSI format is subjected to skin color segmentation according to an HSI face skin color segmentation standard to obtain a binary image, including: Obtain a face image under existing conditions and randomly crop the facial skin color area that does not contain facial features to establish a face skin color library; Performing statistics on the chromaticity information components in the face skin color database to obtain a value range of the chromaticity information components; Expanding the value range by a preset range to obtain a final value range; It is determined whether the pixels of the image converted into the HSI format are within a final value range, and a binary image of the image is obtained according to the determination result.

[0009] Optionally, performing smoothing filtering on the binary image to obtain a smoothed skin area includes: Obtain a binary image of the image, wherein pixels in the skin color area are marked as 1, and pixels that do not belong to the skin color area are marked as 0; For all pixels in the image , in pixels As the center, extract the number of pixels marked as 1 in the 5*5 neighborhood; Determine whether the number of pixels marked as 1 is greater than 0; When the number of pixels marked as 1 is greater than 0, the RGB mean is taken based on the RGB channels of all pixels in the neighborhood; Replace the RGB value of the central pixel with the RGB mean, and perform binary segmentation on the central pixel based on the RGB value of the replaced pixel; When the number of pixels marked as 1 is 0, the pixel value in the center remains unchanged; Traverse all pixels to obtain the smoothed skin area.

[0010] Optionally, the filtered skin area is subjected to template matching to obtain an initial face area, including: Detecting the screened skin area using a binocular detection template to screen the skin area including both eyes; Obtain existing face templates with different aspect ratios; Gradually shrinking the skin area at a ratio of 1:1.1 and matching it with the template to obtain a corresponding matching degree until the skin area is reduced to a preset threshold; The template with the largest matching degree is used as the matching template of the skin area, and the overlapping area of ​​the matching template and the skin area is used as the initial face area.

[0011] Optionally, performing geometric rule confirmation on the initial face region to obtain a confirmed face region in the image includes: Performing grayscale transformation on the initial face region, and binarizing the image after the grayscale transformation by setting a preset grayscale value to determine the eye region; Make vertical and horizontal integral projections corresponding to the horizontal and vertical parts of the binarized image to detect the positions of the eyes and the nose tip; Calculating the distance from both eyes to the nose tip and the ratio between the two eyes, and determining whether the distance from both eyes to the nose tip and the ratio between the two eyes are within a preset range; When the distance from both eyes to the nose tip and the ratio between both eyes are within a preset range, the detected initial face region is determined to be the confirmed face region in the image.

[0012] Optionally, after the information of the staff is entered into one unit, it will be mutually recognized in other units of the same level or subordinate units.

[0013] On the other hand, the present invention also provides a substation non-sensing access system based on the China Unicom mutual recognition mechanism, characterized in that the system includes: A video acquisition module, used to acquire a video of the worker's actions when the worker enters a set range before the substation; A video frame extraction module is used to extract frames from the action video to obtain an image containing the staff member's face; An image processing module, configured to pre-process the image containing the staff member's face to obtain a face region in the image; A face recognition module, configured to recognize a face region in the image to obtain information about the staff member; The personnel access module is used to determine that the identified personnel information is a substation staff member, allow the personnel to enter the substation, and continue to obtain the video of the personnel in the substation; The violation identification module is used to analyze the video of the staff in the substation and identify the violation of the staff in the substation.

[0014] In yet another aspect, the present invention further provides a computer device comprising: one or more processors; The processor is configured to store one or more programs; When the one or more programs are executed by the one or more processors, a method for seamless access to substations based on a China Unicom mutual recognition mechanism as described above is implemented.

[0015] A computer-readable storage medium stores a computer program, which, when executed, implements the above-mentioned method for non-sensing access to substations based on the interconnection mutual recognition mechanism.

[0016] Through the above technical solution, the present invention provides a method for non-sensing access to substations based on a mutual recognition mechanism. When a staff member is within a set range before entering the substation, the action video of the staff member can be obtained. After obtaining the action video, the action video can be framed to obtain an image containing the staff member's face. After obtaining the image containing the staff member's face, the image can be preprocessed to obtain the face area in the image. After obtaining the face area, the face area in the image can be identified to obtain information about the staff member. When it is determined that the information of the identified staff member is a staff member belonging to the substation, the staff member can be allowed to enter the substation, and the video of the staff member in the substation can continue to be obtained. Based on the analysis of the video of the staff member in the substation obtained, the staff member's violation in the substation can be identified. This method can accurately identify the substation access personnel and identify violations.

[0017] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following detailed description, they are used to explain the embodiments of the present invention, but do not constitute a limitation of the embodiments of the present invention. In the accompanying drawings: Figure 1 This is a flow chart of a method for non-sensing access to substations based on a China Unicom mutual recognition mechanism according to an embodiment of the present invention; Figure 2 This is a flowchart of obtaining a facial region in a method for non-sensing access to a power substation based on a China Unicom mutual recognition mechanism according to an embodiment of the present invention; Figure 3 This is a flowchart of brightness compensation of a transformer substation non-sensing access method based on a China Unicom mutual recognition mechanism according to an embodiment of the present invention; Figure 4 This is a flowchart of obtaining a binary graph of a method for non-sensing access to substations based on a China Unicom mutual recognition mechanism according to an embodiment of the present invention; Figure 5 This is a flowchart of a skin area smoothing process of a method for non-sensing access to substations based on a China Unicom mutual recognition mechanism according to an embodiment of the present invention; Figure 6 This is a flowchart of obtaining an initial face region in a method for non-sensing access to a power substation based on a China Unicom mutual recognition mechanism according to an embodiment of the present invention; Figure 7The present invention is a flowchart of determining the final face area of ​​a method for non-sensing access to substations based on a mutual recognition mechanism according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The following describes the specific implementation of the embodiment of the present invention in detail with reference to the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the embodiment of the present invention and is not used to limit the embodiment of the present invention.

[0020] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application are in compliance with the relevant provisions of laws and regulations. In the embodiments of this application, certain software, components, models, and other existing solutions in the industry may be mentioned. These should be considered as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of this application, but it does not mean that the applicant has or will necessarily use such solutions.

[0021] Figure 1 This is a flow chart of a method for providing seamless access to substations based on a China Unicom mutual recognition mechanism according to an embodiment of the present invention. In the present invention, the process of the method for providing seamless access to substations may include: In step S1, when a worker enters a set range before entering the substation, a motion video of the worker is obtained.

[0022] In step S2, the action video is frame-extracted to obtain an image containing the staff member's face.

[0023] In step S3, the image containing the staff member's face is preprocessed to obtain the face area in the image.

[0024] In step S4, the face area in the image is recognized to obtain information about the staff member.

[0025] In step S5 , when it is determined that the identified staff member is a staff member of the substation, the staff member is allowed to enter the substation, and the video of the staff member in the substation is continuously obtained.

[0026] In step S6, the video of the workers in the substation is analyzed, and violations of the workers in the substation are identified.

[0027] In the present invention, when a staff member is allowed to enter the substation without any sense of contact, an action video of the staff member can be obtained when the staff member is within the set range before entering the substation. After obtaining the action video, the action video can be framed, so that an image containing the staff member's face can be obtained. After obtaining the image containing the staff member's face, the image can be preprocessed, so that the face area in the image can be obtained. After obtaining the face area, the face area in the image can be identified, so that information about the staff member can be obtained. When it is determined that the information of the identified staff member is a staff member belonging to the substation, the staff member can be allowed to enter the substation, and the video of the staff member in the substation can continue to be obtained. Based on the analysis of the video of the staff member in the substation obtained, the violation of the staff member in the substation can be identified. This method can accurately identify the substation access personnel and identify violations.

[0028] In one embodiment of the present invention, Figure 2 As shown, the process of obtaining the face area may include: In step S7, an image containing the staff member's face is acquired, and brightness compensation is performed on the image.

[0029] In step S8, the brightness-compensated image is converted from RGB format to HSI format.

[0030] In step S9, the image converted into the HSI format is subjected to skin color segmentation according to the HSI face skin color segmentation standard to obtain a binary image.

[0031] In step S10, the binary image is subjected to smoothing filtering to obtain a smoothed skin area.

[0032] In step S11, the missing parts in the skin area are filled, and the shape of the complete skin area is screened to remove areas with irregular shapes or unsatisfactory aspect ratios.

[0033] In step S12, the screened skin area is subjected to template matching to obtain an initial face area.

[0034] In step S13, the initial face region is confirmed according to geometric rules to obtain a confirmed face region in the image.

[0035] In the present invention, when acquiring the face area in the image, the image containing the staff member's face after the frame extraction can be first acquired, and then the image can be brightness compensated so that the brightness of the image meets the requirements. After the image is brightness compensated, the brightness compensated image can be converted from RGB format to HSI format. By converting to HSI format, the influence of brightness on skin color segmentation can be reduced. After the format conversion, the image converted to HSI format can be subjected to skin color segmentation according to the HSI face skin color segmentation standard, so that a binary image can be obtained. After obtaining the binary image, the binary image can be smoothed and filtered to facilitate subsequent shape analysis. The image after filtering and smoothing may still contain a large number of non-skin areas. Since the human face is a connected area with skin color, the outer contour of the skin area can be regionally connected. Due to the presence of facial organs, the face area is not completely filled and there are some "black holes". Therefore, the "black holes" in the skin area, that is, the missing parts, can be filled. After filling, the target area may not meet the required shape and proportions due to the presence of face-like regions. Therefore, geometric shape analysis can be used to exclude areas with skin-like color but irregular shapes or substandard aspect ratios. This means that shape screening can be performed on the complete skin area to remove these substandard areas. After screening, the screened areas may contain facial regions. Template matching can then be performed on these screened skin areas to obtain the initial facial region. Once the initial facial region is obtained, geometric rule verification can be performed on this initial facial region to confirm the facial region in the image.

[0036] In one embodiment of the present invention, Figure 3 As shown, the brightness compensation process may include: In step S14, an image containing the staff member's face is obtained, the brightness value in the image is obtained, and the maximum value of the brightness value is obtained.

[0037] In step S15, the image is brightness compensated according to formula (1): Formula (1), in, Indicates the pixel after brightness compensation The brightness value, Indicates the pixel before brightness compensation The brightness value, Indicates the maximum value of the brightness value before brightness compensation. Indicates the logarithmic mean of the brightness values ​​before brightness compensation.

[0038] In the present invention, when performing brightness compensation, an image containing the staff member's face can be first obtained, so that the brightness value in the image can be obtained. After obtaining the brightness value in the image, the maximum value of the brightness value can be obtained. After obtaining the maximum value of the brightness value and all the brightness values ​​in the image, the image can be brightness compensated according to the formula (1). The formula (1) It can be the brightness value of the pixel after brightness compensation.

[0039] In one embodiment of the present invention, the logarithmic mean of the brightness value before brightness compensation can be obtained by using formula (2): Formula (2), in, Indicates the size of the image, Indicates compensation parameters.

[0040] The logarithmic mean obtained by formula (2) can be used in the subsequent brightness compensation of the image.

[0041] In one embodiment of the present invention, Figure 4 As shown, the process of obtaining a binary image may include: In step S16, a face image under existing conditions is acquired, and a facial skin color area excluding facial features is randomly cropped to establish a face skin color library.

[0042] In step S17, statistics are collected on the chromaticity information components in the facial skin color database to obtain a value range of the chromaticity information components.

[0043] In step S18, the value range is expanded by the preset range to obtain a final value range.

[0044] In step S19, it is determined whether the pixels of the image converted into the HSI format are within the final value range, and a binary image of the image is obtained according to the determination result.

[0045] In the present invention, facial regions within an image can be obtained based on the distribution of a binary image. To obtain the binary image, a pre-defined facial image can be acquired, and then the facial skin color region excluding the facial features can be randomly cropped to establish a facial skin color library. After obtaining the facial skin color library, the chromaticity information components within the library can be statistically analyzed to determine the value range of the chromaticity information components. This value range can be the chromaticity range of most faces. To accommodate a certain degree of color noise, a slightly wider range than the actual standard can be used. Therefore, the skin color segmentation standard should conform to a certain range. Specifically, the value range can be expanded by a preset range to obtain a final value range. After obtaining the final value range, it is determined whether pixels in the image converted to HSI format are within the value range. If within the final value range, the pixel can be marked as 1; if not, the pixel can be marked as 0. By arranging the corresponding 0 and 1 labels for the pixels, a binary image of the corresponding image can be obtained. Based on this binary image, the facial region within the image can be obtained.

[0046] In one embodiment of the present invention, Figure 5 As shown, the process of smoothing a skin area may include: In step S20 , a binary image of the image is obtained. In the binary image, pixels belonging to the skin color area are marked as 1, and pixels not belonging to the skin color area are marked as 0.

[0047] In step S21, all pixels in the image are , in pixels As the center, extract the number of pixels marked as 1 in the 5*5 neighborhood.

[0048] In step S22 , it is determined whether the number of pixels marked as 1 is greater than 0.

[0049] In step S23 , when the number of pixels marked as 1 is greater than 0, the RGB mean is taken based on the RGB channels of all pixels in the area.

[0050] In step S24, the RGB value of the central pixel is replaced with the RGB mean value, and the central pixel is binary segmented according to the RGB value of the replaced pixel.

[0051] In step S25 , when the number of pixels marked as 1 is 0, the pixel value at the center is kept unchanged.

[0052] In step S26, all pixels are traversed to obtain the smoothed skin area.

[0053] In the present invention, after obtaining the binary image, the face image will inevitably have noise, so the binary image can be smoothed. In the binary image, the pixels belonging to the skin color area can be marked as 1, and the pixels not belonging to the skin color area can be marked as 0. All the pixels in the image can be marked as pixel Take the center as the center and extract the number of pixels marked as 1 in the 5*5 neighborhood. Then determine whether the number of pixels marked as 1 is greater than 0. When the number of pixels marked as 1 is greater than 0, the RGB mean of the pixels in the neighborhood can be calculated based on the values ​​of the RGB channels of all the pixels in the neighborhood. After obtaining the RGB mean, the RGB value of the pixel in the center can be replaced with the RGB mean, so that the RGB value of the pixel in the center can be updated, and binary segmentation can be performed based on the RGB value of the updated pixel. The binary segmentation procedure is the same as the above-mentioned binary segmentation method. When the number of pixels marked as 1 is 0, the pixel value in the center can be kept unchanged. By traversing all pixels through the above method, a smoothed skin area can be obtained.

[0054] In one embodiment of the present invention, Figure 6 As shown, the process of obtaining the initial face region may include: In step S27, the filtered skin area is detected using a binocular detection template to filter out the skin area containing both eyes.

[0055] In step S28, existing face templates with different aspect ratios are obtained.

[0056] In step S29, the skin area is gradually reduced in size according to a ratio of 1:1.1 and matched with the template to obtain a corresponding matching degree until the skin area is reduced to a preset threshold.

[0057] In step S30, the template with the greatest matching degree is used as the matching template of the skin area, and the overlapping area between the matching template and the skin area is used as the initial face area.

[0058] In the present invention, when obtaining the initial face area, the filtered skin area can be detected using a binocular detection template, thereby filtering the skin area containing both eyes. Then, templates of existing faces with different aspect ratios can be obtained. After obtaining the template, the skin area can be gradually reduced at a ratio of 1:1.1 and matched with the template after each reduction until the skin area is reduced to a preset threshold, thereby detecting faces whose actual size is larger than the standard template. When matching with the template, a corresponding matching degree can be obtained. After obtaining the matching degree, the template with the highest matching degree can be used as the matching template for the skin area, and the overlapping area of ​​the matching template and the skin area can be used as the initial face area.

[0059] In one embodiment of the present invention, Figure 7 As shown, the process of determining the final face area may include: In step S31, grayscale transformation is performed on the initial face region, and the image after grayscale transformation is binarized by setting a preset grayscale value to determine the eye region.

[0060] In step S32, vertical and horizontal integral projections corresponding to the horizontal and vertical parts are made in the binarized image to detect the positions of the eyes and the nose tip.

[0061] In step S33, the distance from both eyes to the nose tip and the ratio between the two eyes are calculated, and it is determined whether the distance from both eyes to the nose tip and the ratio between the two eyes are within a preset range.

[0062] In step S34 , when the distance from both eyes to the nose tip and the ratio between both eyes are within a preset range, the detected initial face region is determined to be the confirmed face region in the image.

[0063] In the present invention, when determining the final facial region, the initial facial region can first be grayscale transformed. The grayscale transformed image can then be binarized by setting a preset grayscale value. The eye region can then be determined based on the binarized image. The grayscale values ​​and binarized image can disperse the grayscale levels of low-value areas, such as the iris, and concentrate the grayscale levels of high-value areas, such as the skin. This can widen the grayscale difference between the eyeball and its surroundings, enhancing the contrast between the features and other parts. Vertical and horizontal integral projections corresponding to the horizontal and vertical components of the binarized image can be created to detect the positions of the eyes and the tip of the nose. The distance from the eyes to the tip of the nose and the ratio between the eyes can then be calculated. After the ratios are determined, it can be determined whether these distances and ratios are within a preset range. If these distances and ratios are within the preset range, the initial facial region detected can be determined to be the facial region in the confirmed image. The acquired facial region can then be compared with a database to obtain information corresponding to the face.

[0064] In one embodiment of the present invention, after the staff member's information is entered into one unit, it can be retained and mutually recognized in other units of the same level or subordinate units.

[0065] On the other hand, the present invention can also provide a substation non-sensing access system based on the China Unicom mutual recognition mechanism, the system comprising: The video acquisition module is used to acquire action videos of the staff when the staff enters the set range before the substation.

[0066] The video frame extraction module is used to extract frames from the action video to obtain an image containing the staff member's face.

[0067] The image processing module is used to pre-process the image containing the staff member's face to obtain the face area in the image.

[0068] The face recognition module is used to recognize the face area in the image to obtain information about the staff member.

[0069] The personnel admission module is used to determine that the identified personnel information is a personnel belonging to the substation, allow the personnel to enter the substation, and continue to obtain the video of the personnel in the substation.

[0070] The violation identification module is used to analyze the video of the staff in the substation and identify the violation of the staff in the substation.

[0071] In yet another aspect, the present invention further provides a computer device comprising: one or more processors; The processor is configured to store one or more programs; When the one or more programs are executed by the one or more processors, a method for seamless access to substations based on a China Unicom mutual recognition mechanism as described above is implemented.

[0072] A computer-readable storage medium stores a computer program, which, when executed, implements the above-mentioned method for non-sensing access to substations based on the interconnection mutual recognition mechanism.

[0073] Through the above technical solution, the present invention provides a method for non-sensing access to substations based on a mutual recognition mechanism. When a staff member is within a set range before entering the substation, the action video of the staff member can be obtained. After obtaining the action video, the action video can be framed to obtain an image containing the staff member's face. After obtaining the image containing the staff member's face, the image can be preprocessed to obtain the face area in the image. After obtaining the face area, the face area in the image can be identified to obtain information about the staff member. When it is determined that the information of the identified staff member is a staff member belonging to the substation, the staff member can be allowed to enter the substation, and the video of the staff member in the substation can continue to be obtained. Based on the analysis of the video of the staff member in the substation obtained, the staff member's violation in the substation can be identified. This method can accurately identify the substation access personnel and identify violations.

[0074] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0075] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0076] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0077] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0078] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0079] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0080] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can be implemented using any method or technology for information storage. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change RAM (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media, such as modulated data signals and carrier waves.

[0081] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus that includes a series of elements includes not only those elements but also other elements not explicitly listed, or includes elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity, or apparatus that includes the element.

[0082] The above are merely embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

Claims

1. A method for non-sensing access to substations based on the mutual recognition mechanism of China Unicom, characterized in that: The method comprises: When a worker enters a set range before entering the substation, obtaining a video of the worker's actions; Extracting frames from the action video to obtain an image containing the staff member's face; Preprocessing the image containing the staff member's face to obtain a face region in the image; Recognizing a facial region in the image to obtain information about the staff member; If it is determined that the identified worker is a worker of the substation, the worker is allowed to enter the substation and the video of the worker in the substation is continuously obtained; Analyze videos of workers in the substation and identify violations of regulations by the workers in the substation.

2. The method according to claim 1, characterized in that Preprocessing the image containing the staff member's face to obtain the face area in the image includes: Acquiring an image containing the staff member's face and performing brightness compensation on the image; Converting the brightness-compensated image from RGB format to HSI format; Performing skin color segmentation on the image converted into the HSI format according to the HSI face skin color segmentation standard to obtain a binary image; Performing smoothing filtering on the binary image to obtain a smoothed skin area; Filling in the missing parts of the skin area, and performing shape screening on the complete skin area to remove areas with irregular shapes or unsatisfactory aspect ratios; Perform template matching on the filtered skin area to obtain the initial face area; The initial face region is confirmed according to geometric rules to obtain a confirmed face region in the image.

3. The method according to claim 2, characterized in that Acquiring an image containing the staff member's face and performing brightness compensation on the image, including: Acquire an image containing a staff member's face, obtain brightness values ​​in the image, and obtain a maximum value among the brightness values; The image is brightness compensated according to formula (1): Formula (1), in, Indicates the pixel after brightness compensation The brightness value, Indicates the pixel before brightness compensation The brightness value, Indicates the maximum value of the brightness value before brightness compensation. Indicates the logarithmic mean of the brightness values ​​before brightness compensation.

4. The method according to claim 3, characterized in that The logarithmic mean of the brightness value before brightness compensation is obtained according to formula (2): Formula (2), in, Indicates the size of the image, Indicates compensation parameters.

5. The method according to claim 2, characterized in that The image converted into the HSI format is subjected to skin color segmentation according to the HSI face skin color segmentation standard to obtain a binary image, including: Obtain a face image under existing conditions and randomly crop the facial skin color area that does not contain facial features to establish a face skin color library; Performing statistics on the chromaticity information components in the face skin color database to obtain a value range of the chromaticity information components; Expanding the value range by a preset range to obtain a final value range; It is determined whether the pixels of the image converted into the HSI format are within a final value range, and a binary image of the image is obtained according to the determination result.

6. The method according to claim 2, characterized in that The binary image is subjected to smoothing filtering to obtain a smoothed skin area, including: Obtain a binary image of the image, wherein pixels in the skin color area are marked as 1, and pixels that do not belong to the skin color area are marked as 0; For all pixels in the image , in pixels As the center, extract the number of pixels marked as 1 in the 5*5 neighborhood; Determine whether the number of pixels marked as 1 is greater than 0; When the number of pixels marked as 1 is greater than 0, the RGB mean is taken based on the RGB channels of all pixels in the neighborhood; Replace the RGB value of the central pixel with the RGB mean, and perform binary segmentation on the central pixel based on the RGB value of the replaced pixel; When the number of pixels marked as 1 is 0, the pixel value in the center remains unchanged; Traverse all pixels to obtain the smoothed skin area.

7. The method according to claim 2, characterized in that The filtered skin area is template matched to obtain the initial face area, including: Detecting the screened skin area using a binocular detection template to screen the skin area including both eyes; Obtain existing face templates with different aspect ratios; Gradually shrinking the skin area at a ratio of 1:1.1 and matching it with the template to obtain a corresponding matching degree until the skin area is reduced to a preset threshold; The template with the largest matching degree is used as the matching template of the skin area, and the overlapping area of ​​the matching template and the skin area is used as the initial face area.

8. The method according to claim 2, characterized in that The initial face region is subjected to geometric rule confirmation to obtain a confirmed face region in the image, including: Performing grayscale transformation on the initial face region, and binarizing the image after the grayscale transformation by setting a preset grayscale value to determine the eye region; Make vertical and horizontal integral projections corresponding to the horizontal and vertical parts of the binarized image to detect the positions of the eyes and the nose tip; Calculating the distance from both eyes to the nose tip and the ratio between the two eyes, and determining whether the distance from both eyes to the nose tip and the ratio between the two eyes are within a preset range; When the distance from both eyes to the nose tip and the ratio between both eyes are within a preset range, the detected initial face region is determined to be the confirmed face region in the image.

9. The method according to claim 1, characterized in that Once the information of the staff members is entered into one unit, it will be mutually recognized in other units of the same level or subordinate units.

10. A substation non-sensing access system based on the China Unicom mutual recognition mechanism, characterized in that: The system comprises: A video acquisition module, used to acquire a video of the worker's actions when the worker enters a set range before the substation; A video frame extraction module is used to extract frames from the action video to obtain an image containing the staff member's face; An image processing module, configured to pre-process the image containing the staff member's face to obtain a face region in the image; A face recognition module, configured to recognize a face region in the image to obtain information about the staff member; The personnel access module is used to determine that the identified personnel information is a substation staff member, allow the personnel to enter the substation, and continue to obtain the video of the personnel in the substation; The violation identification module is used to analyze the video of the staff in the substation and identify the violation of the staff in the substation.

Images