Power Region Abnormality Detection Method, Device, Electronic Equipment and Medium

Abstract

Embodiments of the present disclosure disclose a method, apparatus, electronic device, and medium for detecting anomalies in a power region. A specific implementation of the method includes: performing grayscale processing on each meter image in a set of meter images to generate grayscale meter images; performing preprocessing on each grayscale meter image in the set of grayscale meter images to generate preprocessed meter images; performing meter reading recognition on each preprocessed meter image in the set of preprocessed meter images to generate meter reading recognition images; generating an anomaly detection result for power meter devices based on the set of meter reading recognition images; performing face detection on each region image in a set of region images to generate face detection results; and sending the anomaly detection result for power meter devices and the anomaly user detection result to an associated power region anomaly monitoring terminal. This implementation can promptly eliminate abnormal power equipment, reduce power safety hazards, and promptly detect whether there are unauthorized personnel entering.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computers, and more particularly, to methods, devices, electronic devices, and media for detecting anomalies in power regions. Background Art

[0002] With the continuous expansion of the power grid area, the requirements for operation safety, efficiency, and reliability are constantly increasing. Currently, when inspecting a power region, the commonly adopted method is to inspect power equipment and abnormal users in the power region through traditional manual inspections. However, this method reduces the efficiency of inspecting power equipment in the power region, cannot promptly eliminate abnormal power equipment, easily causes potential power safety hazards, and cannot promptly detect whether there are any outsiders breaking in.

[0003] The above information disclosed in this background art section is only used to enhance the understanding of the background of the inventive concept, and thus, it may include information that does not form the prior art known to those of ordinary skill in the art in this country. Summary of the Invention

[0004] This section of the present disclosure is used to briefly introduce concepts that will be described in detail in the subsequent Detailed Description section. This section of the present disclosure is not intended to identify the key features or essential features of the claimed technical solution, nor is it intended to be used to limit the scope of the claimed technical solution.

[0005] Some embodiments of the present disclosure propose methods, devices, electronic devices, and computer-readable media for detecting anomalies in power regions to solve one or more of the technical problems mentioned in the above Background Art section.

[0006] In a first aspect, some embodiments of the present disclosure provide a method for detecting anomalies in a power region. The method includes: in response to receiving a power meter device defect detection instruction for a target power region, controlling an associated power image acquisition device to collect meter images of each power meter device to be detected within the target power region, obtaining a set of meter images; performing grayscale processing on each meter image in the set of meter images to generate grayscale meter images, obtaining a set of grayscale meter images; performing preprocessing on each grayscale meter image in the set of grayscale meter images to generate preprocessed meter images, obtaining a set of preprocessed meter images; performing meter reading recognition on each preprocessed meter image in the set of preprocessed meter images to generate meter reading recognition images, obtaining a set of meter reading recognition images; generating an anomaly detection result for the power meter device based on the set of meter reading recognition images; in response to receiving a user detection instruction for the target power region, controlling an associated face image acquisition device to collect a set of regional images corresponding to the target power region; performing face detection on each regional image in the set of regional images to generate face detection results, obtaining a set of face detection results; generating an anomaly user detection result based on the set of face detection results; and sending the anomaly detection result of the power meter device and the anomaly user detection result to an associated power region anomaly monitoring terminal.

[0007] Second aspect, some embodiments of the present disclosure provide a power area anomaly detection device, which includes: a first control unit configured to control an associated power image acquisition device to acquire a meter image of each power meter device to be detected in a target power area in response to receiving a power meter device defect detection instruction for the target power area, so as to obtain a meter image group; a grayscale unit configured to perform grayscale processing on each meter image in the above-mentioned meter image group to generate a grayscale meter image, so as to obtain a grayscale meter image group; a preprocessing unit configured to perform preprocessing on each grayscale meter image in the above-mentioned grayscale meter image group to generate a preprocessed meter image, so as to obtain a preprocessed meter image group; a reading recognition unit configured to perform meter reading recognition on each preprocessed meter image in the above-mentioned preprocessed meter image group to generate a meter reading recognition image, so as to obtain a meter reading recognition image group; a generation unit configured to generate a power meter device anomaly detection result according to the above-mentioned meter reading recognition image group; a second control unit configured to control an associated face image acquisition device to acquire a set of area images corresponding to the target power area in response to receiving a user detection instruction for the target power area; a detection unit configured to perform face detection on each area image in the above-mentioned set of area images to generate a face detection result, so as to obtain a face detection result set; a sending unit configured to generate an abnormal user detection result according to the above-mentioned face detection result set, and send the above-mentioned power meter device anomaly detection result and the above-mentioned abnormal user detection result to an associated power area anomaly monitoring terminal.

[0008] Third aspect, some embodiments of the present disclosure provide an electronic device, which includes: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner of the first aspect above.

[0009] Fourth aspect, some embodiments of the present disclosure provide a computer-readable medium on which a computer program is stored, wherein when the program is executed by a processor, the method described in any implementation manner of the first aspect above is implemented.

[0010] The above-mentioned various embodiments of the present disclosure have the following beneficial effects: Through the power area anomaly detection method of some embodiments of the present disclosure, abnormal power equipment can be excluded in a timely manner, power safety hazards can be reduced, and whether there are unauthorized personnel breaking in can be detected in a timely manner. First, in response to receiving a power meter device defect detection instruction for a target power area, control the associated power image acquisition device to acquire meter images of each power meter device to be detected in the target power area, obtaining a set of meter images. Secondly, perform grayscale processing on each meter image in the above set of meter images to generate grayscale meter images, obtaining a set of grayscale meter images. Perform preprocessing on each grayscale meter image in the above set of grayscale meter images to generate preprocessed meter images, obtaining a set of preprocessed meter images. Thereby, efficient processing and analysis of the acquired set of meter images can be realized, and the processing speed of the set of meter images can be improved. Then, perform meter reading recognition on each preprocessed meter image in the above set of preprocessed meter images to generate meter reading recognition images, obtaining a set of meter reading recognition images. Generate an abnormal detection result of the power meter device according to the above set of meter reading recognition images. Thereby, the automation and accuracy of meter reading can be improved, and errors in manual meter reading can be reduced. Then, in response to receiving a user detection instruction for a target power area, control the associated face image acquisition device to acquire a set of area images corresponding to the target power area. Perform face detection on each area image in the above set of area images to generate face detection results, obtaining a set of face detection results. Thereby, it can be detected in a timely manner whether there are unauthorized personnel breaking in. Finally, generate an abnormal user detection result according to the above set of face detection results, and send the above abnormal detection result of the power meter device and the above abnormal user detection result to the associated power area anomaly monitoring terminal. Thereby, abnormal power equipment can be excluded in a timely manner, power safety hazards can be reduced, and whether there are unauthorized personnel breaking in can be detected in a timely manner. Description of the Drawings

[0011] In combination with the accompanying drawings and with reference to the following specific embodiments, the above and other features, advantages, and aspects of the various embodiments of the present disclosure will become more apparent. Throughout the drawings, the same or similar reference numerals indicate the same or similar elements. It should be understood that the drawings are schematic, and the elements and elements are not necessarily drawn to scale.

[0012] Figure 1 is a flowchart of some embodiments of the power area anomaly detection method according to the present disclosure;

[0013] Figure 2 is a flowchart of some embodiments of the power area anomaly detection device according to the present disclosure;

[0014] Figure 3 is a schematic structural diagram of an electronic device suitable for implementing some embodiments of the present disclosure. Detailed implementation manners

[0015] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present disclosure. It should be understood that the accompanying drawings and embodiments of the present disclosure are only for exemplary purposes and are not used to limit the protection scope of the present disclosure.

[0016] In addition, it should be noted that, for the sake of convenience of description, only parts related to the relevant invention are shown in the drawings. Without conflict, the embodiments in the present disclosure and the features in the embodiments can be combined with each other.

[0017] It should be noted that concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence relationship of the functions executed by these devices, modules or units.

[0018] It should be noted that the modifications of "one" and "a plurality" mentioned in the present disclosure are illustrative rather than restrictive. Those skilled in the art should understand that, unless otherwise clearly specified in the context, it should be understood as "one or more".

[0019] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0020] The present disclosure will be described in detail below with reference to the drawings and in combination with embodiments.

[0021] Figure 1 is a flowchart of some embodiments of the power area anomaly detection method according to the present disclosure. The process 100 of some embodiments of the power area anomaly detection method according to the present disclosure is shown. The power area anomaly detection method includes the following steps:

[0022] Step 101, in response to receiving a power meter device defect detection instruction for a target power area, controlling an associated power image acquisition device to acquire meter images of each power meter device to be detected in the target power area, obtaining a set of meter images.

[0023] In some embodiments, the execution subject of the power area anomaly detection method (e.g., a computing device) may, in response to receiving a power meter device defect detection instruction for a target power area, control an associated power image acquisition device to acquire meter images of each power meter device to be detected within the target power area, thereby obtaining a set of meter images. The target power area may refer to a power area where various power devices are placed and need to be monitored. The power meter device defect detection instruction may refer to an instruction for detecting defects in each power meter device within the target power area. The power meter device may be: a mechanical electric meter, an induction electric meter, a plug-in prepaid electric meter, or an electronic electric meter. The meter image may refer to an image showing the dial of the power meter device.

[0024] Step 102: Perform grayscale processing on each meter image in the above set of meter images to generate a grayscale meter image, thereby obtaining a set of grayscale meter images.

[0025] In some embodiments, the above execution subject may perform grayscale processing on each meter image in the above set of meter images to generate a grayscale meter image, thereby obtaining a set of grayscale meter images. For example, the above set of meter images may be subjected to grayscale processing through an image grayscale algorithm to generate a grayscale meter image, thereby obtaining a set of grayscale meter images.

[0026] Step 103: Perform preprocessing on each grayscale meter image in the above set of grayscale meter images to generate a preprocessed meter image, thereby obtaining a set of preprocessed meter images.

[0027] In some embodiments, the above execution subject may perform preprocessing on each grayscale meter image in the above set of grayscale meter images to generate a preprocessed meter image, thereby obtaining a set of preprocessed meter images.

[0028] In practice, the above execution subject may preprocess each grayscale meter image in the above set of grayscale meter images through the following steps to generate a preprocessed meter image:

[0029] First step: Perform image noise reduction processing on each set of grayscale meter images in the above set of grayscale meter images to generate a noise-reduced meter image, thereby obtaining a set of noise-reduced meter images. For example, the above set of grayscale meter images may be subjected to image noise reduction processing through a preset image noise reduction algorithm to generate a noise-reduced meter image. The image noise reduction algorithm may be: algorithms such as spatial domain filtering and frequency domain filtering. It is also possible to perform image noise reduction processing on each set of grayscale meter images in the above set of grayscale meter images through a filter to generate a noise-reduced meter image, thereby obtaining a set of noise-reduced meter images.

[0030] Second step, perform contrast enhancement processing on each of the above noise-reduced meter images in the noise-reduced meter image group to generate enhanced meter images, obtaining an enhanced meter image group. For example, the above noise-reduced meter images in the noise-reduced meter image group can be subjected to contrast enhancement processing through a preset histogram equalization algorithm to generate enhanced meter images, obtaining an enhanced meter image group. The preset histogram equalization algorithm can be a contrast-limited adaptive histogram equalization algorithm.

[0031] Third step, perform image binarization processing on each of the above enhanced meter images in the enhanced meter image group to generate binarized meter images, obtaining a binarized meter image group as a preprocessed meter image group. For example, first, a preset pixel threshold can be set. In response to determining that the pixel value in the enhanced meter image is greater than or equal to the preset pixel threshold, the above enhanced meter image is determined as a white area meter image. Second, in response to determining that the pixel value in the above enhanced meter image is less than the preset pixel threshold, the above enhanced meter image is determined as a black area meter image. Finally, a preset binarization algorithm is used to perform pixel value classification processing on each white area meter image and each black area meter image to generate a binarized meter image group. The preset binarization algorithm can be: the Otsu algorithm. The preset pixel threshold can be: 0 - 255.

[0032] Step 104, perform meter reading recognition on each of the above preprocessed meter images in the preprocessed meter image group to generate meter reading recognition images, obtaining a meter reading recognition image group.

[0033] In some embodiments, the above execution subject can perform meter reading recognition on each of the above preprocessed meter images in the preprocessed meter image group to generate meter reading recognition images, obtaining a meter reading recognition image group. For example, the above execution subject can perform meter reading recognition on each of the above preprocessed meter images in the preprocessed meter image group through a pre-trained meter reading recognition model to generate meter reading recognition images, obtaining a meter reading recognition image group. For example, the meter reading recognition model can be a neural network model based on OCR technology. OCR technology: Optical Character Recognition (OCR) is a technology that converts text in an image into machine-encoded text. For another example, the meter reading recognition model can be a reading recognition model based on YOLOv8 deep learning.

[0034] In practice, the above execution subject can perform meter reading recognition on each of the above preprocessed meter images in the preprocessed meter image group through the following steps:

[0035] First step, perform edge detection processing on each preprocessed meter image in the above-mentioned preprocessed meter image group to generate edge-detected meter images, obtaining an edge-detected meter image group. For example, the above-mentioned direct-reading instrument can use a preset edge detection algorithm to perform meter image boundary extraction processing on each preprocessed meter image in the above-mentioned preprocessed meter image group to generate edge-detected meter images, obtaining an edge-detected meter image group set. The preset edge detection algorithm can be: Canny edge detection algorithm.

[0036] Second step, perform contour extraction processing on each edge-detected meter image in the above-mentioned edge-detected meter image group to generate meter contour images, obtaining a meter contour image group, where there are bounding boxes in the meter contour images. For example, the above-mentioned edge-detected binary meter images can be used to perform contour extraction processing by using the OpenCV library to generate contour meter images, obtaining a contour meter image set. The steps for contour extraction using the OpenCV library are as follows: 1. Contour detection: Use the cv2.findContours() function to detect the contours in the image. 2. Draw contours: Use the cv2.drawContours() function to draw the contours on the original image.

[0037] Third step, for each meter contour image in the above-mentioned meter contour image group, perform the following meter reading recognition steps:

[0038] 1. In response to determining that the contour area of the above-mentioned meter contour image is less than the preset contour area threshold, generate a contour area meter bounding box group. The meter contour image can be subjected to contour filtering processing to generate a filtered contour area. Secondly, use the above-mentioned preset function to perform screening processing on the filtered contour area to generate a screened and filtered contour area. Finally, extract the contour area meter bounding box group from the above-mentioned screened and filtered contour area. The contour area threshold of the contour area meter image can be: 100.

[0039] 2. Based on the above-mentioned contour area meter bounding box group, perform character recognition processing on the above-mentioned area meter image to generate a character sequence. First, the above-mentioned contour area meter bounding box group can be sorted horizontally to generate a horizontally sorted meter bounding box group. According to the above-mentioned horizontally sorted meter bounding box group, perform cropping processing on the above-mentioned meter contour image to generate a character sequence.

[0040] 3. Based on the above-mentioned character sequence, output a meter reading recognition image. Each character in the above-mentioned character sequence can be subjected to image format conversion processing to generate a meter reading recognition image.

[0041] Thus, operations such as noise reduction, enhancement, and binarization are performed on the collected images to improve the accuracy of subsequent recognition.

[0042] Step 105: Identify the image group based on the above meter readings and generate the abnormal detection result of the power meter device.

[0043] In some embodiments, the above execution subject may identify the image group based on the above meter readings and generate the abnormal detection result of the power meter device.

[0044] In practice, the above execution subject may generate the abnormal detection result of the power meter device through the following steps:

[0045] First step: For each meter reading recognition image in the above meter reading recognition image group, perform the following processing steps:

[0046] 1. Determine the power meter device corresponding to the above meter reading recognition image as the target power meter device.

[0047] 2. Determine the standard meter reading image of the above target power meter device. The standard meter reading image may be an image of the meter reading that the target power meter device should display as preset.

[0048] 3. Determine whether the meter reading represented by the above meter reading recognition image is consistent with the meter reading represented by the above meter reading recognition image. That is, determine whether the meter reading represented by the above meter reading recognition image is the same as the meter reading represented by the above meter reading recognition image. 4. In response to determining inconsistency, generate an abnormal meter monitoring result corresponding to the above target power meter device. That is, the abnormal meter monitoring result may indicate that the current meter reading of the target power meter device is abnormal.

[0049] Second step: Combine each abnormal meter monitoring result into the abnormal detection result of the power meter device.

[0050] Step 106: In response to receiving a user detection instruction for the target power area, control the associated face image acquisition device to acquire a set of area images corresponding to the target power area.

[0051] In some embodiments, the above execution subject may, in response to receiving a user detection instruction for the target power area, control the associated face image acquisition device to acquire a set of area images corresponding to the target power area. The user detection instruction may be an instruction to detect whether there are abnormal users in the target power area. An abnormal user may refer to a user whose information has not been filed in advance. The face image acquisition device may refer to a camera device arranged in the target power area for acquiring face images, which may include multiple cameras. Each area image in the set of area images may completely represent the target power area.

[0052] Step 107: Perform face detection on each regional image in the above regional image set to generate face detection results, obtaining a face detection result set.

[0053] In some embodiments, the above execution entity may perform face detection on each regional image in the above regional image set to generate face detection results, obtaining a face detection result set.

[0054] In practice, the above execution entity may perform face detection on each regional image in the above regional image set to generate face detection results through the following steps:

[0055] First step: Perform image preprocessing on each regional image in the above regional image set to generate preprocessed regional images, obtaining a preprocessed regional image set. For example, first, image cleaning processing may be performed on each regional image in the above regional image set to generate cleaned regional images, obtaining a cleaned regional image set. Second, image feature selection processing may be performed on each cleaned regional image in the above cleaned regional image set to generate feature selection regional images, obtaining a feature selection regional image set. Then, image normalization processing may be performed on each feature selection regional image in the above feature selection regional image set to generate preprocessed regional images, obtaining a preprocessed regional image set.

[0056] Second step: Perform face localization processing on each preprocessed regional image in the above preprocessed regional image set to generate face localization images, obtaining a face localization image set. Face localization processing may be performed on each preprocessed regional image in the above preprocessed regional image set by using a deep learning model and a classifier to generate face localization images, obtaining a face localization image set. The deep learning model may be: Residual Network Model (ResNet), Multi-task Cascaded Convolutional Networks (MTCNN). The classifier may be: Feature Cascade Classifier (Haar).

[0057] Third step: Perform multi-scale detection processing on each face localization image in the above face localization image set to generate multi-scale detected face images, obtaining a multi-scale detected face image set. Among them, there are face bounding boxes in the multi-scale detected face images. For example, multi-scale detection processing may be performed on each face localization image in the above face localization image set by using multiple detection scales to generate multi-scale detected face images, obtaining a multi-scale detected face image set. The multiple detection scales may be: 1:1, 1:1.25, 1:1.5, 1:1.75, 1:2.

[0058] Step 4: Perform face region cropping on each multi-scale detected face image in the above multi-scale detected face image set to generate region-cropped face images, obtaining a region-cropped face image set. First, the bounding box coordinates of each face can be obtained from the above multi-scale detected face image set. Second, based on the bounding box coordinates of each face, the face region of the corresponding multi-scale detected face image is cropped from the above multi-scale detected face image set to obtain a region-cropped face image set. The bounding box coordinates can be: x, y, z.

[0059] Step 5: Perform face box drawing on each region-cropped face image in the above region-cropped face image set to generate box-drawn face images, obtaining a box-drawn face image set. For example, for each region-cropped face image in the above region-cropped face image set, through a preset face box drawing function, perform face box drawing on the above region-cropped face image to generate box-drawn face images, obtaining a box-drawn face image set. The preset face box drawing function can be: detect_faces.

[0060] Step 6: Perform face image alignment on each box-drawn face image in the above box-drawn face image set to generate aligned face images, obtaining an aligned face image set. For example, for each box-drawn face image in the above box-drawn face image set, through a preset alignment algorithm, perform face image alignment on the above box-drawn face image to generate aligned face images, obtaining an aligned face image set. The preset alignment algorithm can be: key point detection algorithm, geometric transformation algorithm.

[0061] Step 7: Perform image noise reduction on each aligned face image in the above aligned face image set to generate noise-reduced face images, obtaining a noise-reduced face image set. For example, for each aligned face image in the above aligned face image set, through a preset image noise reduction algorithm, perform image noise reduction on the above aligned face image to generate noise-reduced face images, obtaining a noise-reduced face image set. The preset image noise reduction algorithm can be: fast mean denoising algorithm.

[0062] Step 8: Perform image sharpening on each noise-reduced face image in the above noise-reduced face image set to generate sharpened face images, obtaining a sharpened face image set. For each noise-reduced face image in the above noise-reduced face image set, use a filter to perform image sharpening on the above noise-reduced face image to generate sharpened face images, obtaining a sharpened face image set. The filter can be: convolution filter.

[0063] Step 9. Input the above sharpened face image set into a pre-trained face detection model to obtain a face detection result set. Among them, one sharpened face image corresponds to one face detection result. The face detection model can be a pre-trained neural network model that takes a sharpened face image as input and outputs a face detection result. The face detection model can be an EigenFace model, a FisherFace model, a Local Binary Patterns (LBP) model, or a VGGFace model. For example, the EigenFace model can be a face recognition model based on principal component analysis (PCA), which extracts the main features of the face through dimensionality reduction for recognition. The FisherFace model can be optimized using linear discriminant analysis (LDA) based on the EigenFace model, improving the recognition performance under different illuminations and expressions. The face detection result can represent the user's identity information. For example, if the user has been locally registered, the user's identity information (face identifier) will be displayed; if the user has not been locally registered, the face detection result indicates that the user is abnormal. The face detection result can include the detection results of each face user in the corresponding regional image.

[0064] Thus, through the deep learning model, the system can perform face detection and tracking, and further improve the image clarity; after detecting a face, it automatically adjusts the light and contrast to ensure the image quality.

[0065] Step 108. Generate an abnormal user detection result based on the above face detection result set, and send the above power meter device abnormal detection result and the above abnormal user detection result to the associated power area abnormal monitoring terminal.

[0066] In some embodiments, the above execution entity can generate an abnormal user detection result based on the above face detection result set, and send the above power meter device abnormal detection result and the above abnormal user detection result to the associated power area abnormal monitoring terminal. The power area abnormal monitoring terminal can refer to a terminal that monitors the target power area for technicians and monitors to operate.

[0067] In practice, the above execution entity can generate an abnormal user detection result through the following steps:

[0068] First step. For each face identifier in each face detection result, perform the following processing steps:

[0069] 1. Determine whether there is user information corresponding to the face identifier in the preset user information set.

[0070] 2. In response to determining that there is no user information corresponding to the face identifier, generate an abnormal user detection sub-result corresponding to the above face identifier.

[0071] In the second step, merge the detection sub-results of each abnormal user into the detection result of abnormal users.

[0072] Optionally, in response to receiving a tracking instruction for a patrol target within a target power area, collect area target images of the patrol target within a preset time period to obtain a sequence of area target images.

[0073] In some embodiments, the above-mentioned execution entity may, in response to receiving a tracking instruction for a patrol target within a target power area, collect area target images of the patrol target within a preset time period to obtain a sequence of area target images. The patrol target may refer to a machine or a user moving within the target power area. The area target images of the patrol target within the preset time period may be collected by a camera device to obtain a sequence of area target images.

[0074] Optionally, perform object detection and object bounding box marking processing on each area target image in the above-mentioned sequence of area target images to generate a target-marked area image, and obtain a set of target-marked area images.

[0075] In some embodiments, the above-mentioned execution entity may perform object detection and object bounding box marking processing on each area target image in the above-mentioned sequence of area target images to generate a target-marked area image, and obtain a set of target-marked area images. Among them, the target-marked area image has a bounding box marking the patrol target. For example, for each area target image in the above-mentioned set of area target images, an initialized tracker may be used to perform object detection and object bounding box marking processing on the above-mentioned area target image to generate a target-marked area image, and obtain a set of target-marked area images. The initialized tracker may be: a Multiple Instance Learning (MIL) tracker.

[0076] Optionally, perform object tracking processing on each target-marked area image in the above-mentioned set of target-marked area images to generate the tracking position of the target-marked area image, and obtain a set of tracking positions of the target-marked area images.

[0077] In some embodiments, the above-mentioned execution entity may perform object tracking processing on each target-marked area image in the above-mentioned set of target-marked area images to generate the tracking position of the target-marked area image, and obtain a set of tracking positions of the target-marked area images. For example, for each target-marked area image in the above-mentioned set of target-marked area images, a tracker may be used to perform object tracking processing on the above-mentioned target-marked area image to generate the tracking position of the target-marked area image, and obtain a set of tracking positions of the target-marked area images. The tracker may be: a Kernelized Correlation Filter (KCF) tracker.

[0078] Optionally, perform trajectory prediction processing on the above-mentioned set of tracking positions of the target-marked area images to generate a target prediction trajectory of the marked detection area image.

[0079] In some embodiments, the above-mentioned execution entity may perform trajectory prediction processing on the above-mentioned target marker region image tracking position set to generate a target prediction trajectory of the marker detection region image. For example, a filter may be used to perform trajectory prediction processing on the above-mentioned target marker region image tracking position set to generate a target prediction trajectory of the marker detection region image. The filter may be a Kalman filter.

[0080] Optionally, target point feature extraction processing is performed on each region target image of the above-mentioned region target image sequence to generate feature extraction target points, and a feature extraction target point set is obtained.

[0081] In some embodiments, the above-mentioned execution entity may perform target point feature extraction processing on each region target image of the above-mentioned region target image sequence to generate feature extraction target points, and a feature extraction target point set is obtained. For example, for each region target image in the above-mentioned region target image sequence, a preset target point feature extraction algorithm may be used to perform target point feature extraction processing on the above-mentioned region target image to generate feature extraction target points, and a feature extraction target point set is obtained. The preset target point feature extraction algorithm may be the Oriented FAST and Rotated BRIEF (ORB) algorithm.

[0082] Optionally, target point feature matching processing is performed on each feature extraction target point in the above-mentioned feature extraction target point set to generate feature matching target points, and a feature matching target point set is obtained.

[0083] In some embodiments, the above-mentioned execution entity may perform target point feature matching processing on each feature extraction target point in the above-mentioned feature extraction target point set to generate feature matching target points, and a feature matching target point set is obtained. For example, for each feature extraction target point in the above-mentioned feature extraction target point set, a preset matcher may be used to perform target point feature matching processing on the above-mentioned feature extraction target point to generate feature matching target points, and a feature matching target point set is obtained. The preset matcher may be a Brute-Force Matcher (BFMatcher).

[0084] Optionally, based on the above-mentioned feature matching target point set, target image registration processing is performed on each region target image in the above-mentioned region target image sequence to generate registered target detection region images, and a registered target detection region image set is obtained.

[0085] In some embodiments, the above-mentioned execution entity may match the target point set based on the above-mentioned features, and perform target image registration processing on each regional target image in the above-mentioned regional target image sequence to generate a registered target detection regional image, thereby obtaining a set of registered target detection regional images. For example, for each regional target image in the above-mentioned regional target image sequence, the above-mentioned regional target image may be subjected to target image registration processing through a preset target image registration matrix to generate a registered target detection regional image, thereby obtaining a set of registered target detection regional images. The preset target image registration matrix may be: calculating a homography matrix (Homography).

[0086] Optionally, perform target image fusion processing on the above-mentioned set of registered target detection regional images to generate a fused target detection regional image.

[0087] In some embodiments, the above-mentioned execution entity may perform target image fusion processing on the above-mentioned set of registered target detection regional images to generate a fused target detection regional image. For example, a preset target image fusion technique may be used to perform target image fusion processing on the above-mentioned set of registered target detection regional images to generate a fused target detection regional image. The preset target image fusion technique may be: a perspective transformation (warpPerspective) technique.

[0088] Optionally, based on the above-mentioned target prediction trajectory of the marked detection regional image target, perform target locking processing on the above-mentioned fused target detection regional image to generate the future position of the above-mentioned inspection target.

[0089] In some embodiments, the above-mentioned execution entity may perform target locking processing on the above-mentioned fused target detection regional image based on the above-mentioned target prediction trajectory of the marked detection regional image target to generate the future position of the above-mentioned inspection target. For example, a preset target locking algorithm may be used to perform target locking processing on the above-mentioned fused target detection regional image to generate the future position of the above-mentioned inspection target. The preset target locking algorithm may be: a tracking algorithm (KCF).

[0090] Further referring to Figure 2 , as an implementation of the methods shown in the above figures, some embodiments of a power region anomaly detection device are provided in the present disclosure. These embodiments of the power region anomaly detection device correspond to Figure 1 the method embodiments shown, and the power region anomaly detection device may be specifically applied to various electronic devices.

[0091] As shown in Figure 2As shown, the power area anomaly detection device 200 of some embodiments includes: a first control unit 201, a grayscale unit 202, a preprocessing unit 203, a reading recognition unit 204, a generation unit 205, a second control unit 206, a detection unit 207, and a sending unit 208. Among them, the first control unit 201 is configured to, in response to receiving a power meter device defect detection instruction for a target power area, control an associated power image acquisition device to acquire a meter image of each power meter device to be detected in the target power area, obtaining a set of meter images; the grayscale unit 202 is configured to perform grayscale processing on each meter image in the above set of meter images to generate a grayscale meter image, obtaining a set of grayscale meter images; the preprocessing unit 203 is configured to perform preprocessing on each grayscale meter image in the above set of grayscale meter images to generate a preprocessed meter image, obtaining a set of preprocessed meter images; the reading recognition unit 204 is configured to perform meter reading recognition on each preprocessed meter image in the above set of preprocessed meter images to generate a meter reading recognition image, obtaining a set of meter reading recognition images; the generation unit 205 is configured to generate a power meter device anomaly detection result according to the above set of meter reading recognition images; the second control unit 206 is configured to, in response to receiving a user detection instruction for a target power area, control an associated face image acquisition device to acquire a set of area images corresponding to the target power area; the detection unit 207 is configured to perform face detection on each area image in the above set of area images to generate a face detection result, obtaining a set of face detection results; the sending unit 208 is configured to generate an abnormal user detection result according to the above set of face detection results, and send the above power meter device anomaly detection result and the above abnormal user detection result to an associated power area anomaly monitoring terminal.

[0092] It can be understood that the units described in the power area anomaly detection device 200 correspond to the respective steps in the method described with reference to Figure 1 Therefore, the operations, features, and beneficial effects described above for the method also apply to the power area anomaly detection device 200 and the units included therein, and will not be elaborated here.

[0093] The following refers to Figure 3 , which shows a schematic structural diagram of an electronic device (such as a computing device) 300 suitable for implementing some embodiments of the present disclosure. The electronic devices in some embodiments of the present disclosure may include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), vehicle terminals (such as vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 3The electronic device shown is merely an example and should not impose any limitations on the functions and usage scope of the embodiments of the present disclosure.

[0094] As Figure 3 shown, the electronic device 300 may include a processing device (such as a central processing unit, a graphics processing unit, etc.) 301, which may perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage device 308 into the random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other through a bus 304. The input / output (I / O) interface 305 is also connected to the bus 304.

[0095] Generally, the following devices may be connected to the I / O interface 305: an input device 306 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 308 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 309. The communication device 309 may allow the electronic device 300 to communicate with other devices wirelessly or wirelesly to exchange data. Although Figure 3 the electronic device 300 with various devices is shown, it should be understood that it is not required to implement or include all the shown devices. More or fewer devices may be implemented or included alternatively. Figure 3 Each block shown in

[0096] particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart may be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes program codes for executing the method shown in the flowchart. In such some embodiments, the computer program may be downloaded and installed from the network through the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above functions defined in the method of some embodiments of the present disclosure are executed.

[0097] It should be noted that the computer-readable media described in some embodiments of the present disclosure may be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. The computer-readable storage medium may, for example, be—but is not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of the present disclosure, the computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than the computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0098] In some embodiments, the client and the server can communicate using any currently known or future-developed network protocol such as HTTP (HyperText Transfer Protocol), and can be interconnected with digital data communication in any form or medium (e.g., a communication network). Examples of communication networks include local area networks (“LAN”), wide area networks (“WAN”), the Internet (e.g., the Internet), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed network.

[0099] The above computer-readable medium may be included in the above electronic device; or it may exist independently and not be assembled into the electronic device. The above computer-readable medium carries one or more programs. When the above one or more programs are executed by the electronic device, the electronic device is caused to: in response to receiving a power meter device defect detection instruction for a target power area, control an associated power image acquisition device to acquire meter images of each power meter device to be detected in the target power area, obtaining a set of meter images; perform grayscale processing on each meter image in the above set of meter images to generate grayscale meter images, obtaining a set of grayscale meter images; perform preprocessing on each grayscale meter image in the above set of grayscale meter images to generate preprocessed meter images, obtaining a set of preprocessed meter images; perform meter reading recognition on each preprocessed meter image in the above set of preprocessed meter images to generate meter reading recognition images, obtaining a set of meter reading recognition images; generate a power meter device anomaly detection result based on the above set of meter reading recognition images; in response to receiving a user detection instruction for a target power area, control an associated face image acquisition device to acquire a set of area images corresponding to the target power area; perform face detection on each area image in the above set of area images to generate face detection results, obtaining a set of face detection results; generate an abnormal user detection result based on the above set of face detection results, and send the above power meter device anomaly detection result and the above abnormal user detection result to an associated power area anomaly monitoring terminal.

[0100] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages or combinations thereof. The above programming languages include object-oriented programming languages - such as Java, Smalltalk, C++; and also include conventional procedural programming languages - such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partially on the user's computer, execute as a stand-alone software package, execute partially on the user's computer and partially on a remote computer, or execute entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network - including a local area network (LAN) or a wide area network (WAN) - or may be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0101] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a portion of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks may occur in a different order than noted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0102] The units described in some embodiments of the present disclosure can be implemented in software or in hardware. The described units can also be provided in a processor. For example, it can be described as: a processor includes: a first control unit, a grayscale unit, a preprocessing unit, a reading recognition unit, a generating unit, a second control unit, a detecting unit, and a sending unit. Among them, the names of these units do not constitute a limitation on the unit itself in some cases. For example, the detecting unit can also be described as "a unit that performs face detection on each region image in the region image set to generate a face detection result and obtain a face detection result set".

[0103] The functions described above can be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that can be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on a chip (SOCs), complex programmable logic devices (CPLDs), and so on.

[0104] The above description is only some preferred embodiments of the present disclosure and an explanation of the technical principles applied. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by the specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above inventive concept. For example, technical solutions formed by mutually replacing the above features with technical features (but not limited to) having similar functions disclosed in the embodiments of the present disclosure.

Claims

1. A method for detecting anomalies in a power region, comprising: In response to receiving a power meter device defect detection instruction for a target power region, controlling an associated power image acquisition device to acquire meter images of each power meter device to be detected in the target power region, obtaining a set of meter images; Performing grayscale processing on each meter image in the set of meter images to generate grayscale meter images, obtaining a set of grayscale meter images; Performing preprocessing on each grayscale meter image in the set of grayscale meter images to generate preprocessed meter images, obtaining a set of preprocessed meter images; Performing meter reading recognition on each preprocessed meter image in the set of preprocessed meter images to generate meter reading recognition images, obtaining a set of meter reading recognition images; Generating an anomaly detection result for the power meter device according to the set of meter reading recognition images; In response to receiving a user detection instruction for a target power region, controlling an associated face image acquisition device to acquire a set of region images corresponding to the target power region; Performing face detection on each region image in the set of region images to generate face detection results, obtaining a set of face detection results; Generating an abnormal user detection result according to the set of face detection results, and sending the anomaly detection result of the power meter device and the abnormal user detection result to an associated power region anomaly monitoring terminal; Wherein, the performing meter reading recognition on each preprocessed meter image in the set of preprocessed meter images to generate meter reading recognition images includes: performing edge detection processing on each preprocessed meter image in the set of preprocessed meter images to generate edge detection meter images, obtaining a set of edge detection meter images; performing contour extraction processing on each edge detection meter image in the set of edge detection meter images to generate meter contour images, obtaining a set of meter contour images, wherein there are bounding boxes in the meter contour images; for each meter contour image in the set of meter contour images, performing the following meter reading recognition steps: in response to determining that the contour area of the meter contour image is less than a preset contour area threshold, generating a set of meter bounding boxes for the contour area; based on the set of meter bounding boxes for the contour area, performing character recognition processing on the meter contour image to generate a character sequence; based on the character sequence, outputting a meter reading recognition image.

2. The method according to claim 1, wherein The performing preprocessing on each grayscale meter image in the set of grayscale meter images to generate preprocessed meter images, obtaining a set of preprocessed meter images, includes: Performing image denoising processing on each set of grayscale meter images in the set of grayscale meter images to generate denoised meter images, obtaining a set of denoised meter images; Performing contrast enhancement processing on each denoised meter image in the set of denoised meter images to generate enhanced meter images, obtaining a set of enhanced meter images; Performing image binarization processing on each enhanced meter image in the set of enhanced meter images to generate binarized meter images, obtaining a set of binarized meter images as the set of preprocessed meter images.

3. The method according to claim 1, wherein, Identifying the image group according to the meter reading and generating an abnormal detection result of the power meter device, including: For each meter reading identification image in the meter reading identification image group, perform the following processing steps: Determine the power meter device corresponding to the meter reading identification image as the target power meter device; Determine the standard meter reading image of the target power meter device; Determine whether the meter reading represented by the meter reading identification image is consistent with the meter reading represented by the meter reading identification image; In response to determining inconsistency, generate an abnormal meter monitoring result corresponding to the target power meter device; Combine each abnormal meter monitoring result into an abnormal detection result of the power meter device.

4. A power area abnormal detection device, including: A first control unit, configured to control an associated power image acquisition device to acquire a meter image of each power meter device to be detected in a target power area in response to receiving a power meter device defect detection instruction for the target power area, so as to obtain a meter image group; A grayscale unit, configured to perform grayscale processing on each meter image in the meter image group to generate a grayscale meter image, so as to obtain a grayscale meter image group; A preprocessing unit, configured to perform preprocessing on each grayscale meter image in the grayscale meter image group to generate a preprocessed meter image, so as to obtain a preprocessed meter image group; A reading identification unit, configured to perform meter reading identification on each preprocessed meter image in the preprocessed meter image group to generate a meter reading identification image, so as to obtain a meter reading identification image group; A generating unit, configured to generate an abnormal detection result of the power meter device according to the meter reading identification image group; A second control unit, configured to control an associated face image acquisition device to acquire a regional image set corresponding to the target power area in response to receiving a user detection instruction for the target power area; A detection unit, configured to perform face detection on each regional image in the regional image set to generate a face detection result, so as to obtain a face detection result set; A sending unit, configured to generate an abnormal user detection result according to the face detection result set, and send the abnormal detection result of the power meter device and the abnormal user detection result to an associated power area abnormal monitoring terminal Among them, performing meter reading recognition on each preprocessed meter image in the preprocessed meter image group to generate a meter reading recognition image includes: performing edge detection processing on each preprocessed meter image in the preprocessed meter image group to generate an edge detection meter image, obtaining an edge detection meter image group; performing contour extraction processing on each edge detection meter image in the edge detection meter image group to generate a meter contour image, obtaining a meter contour image group, where there are bounding boxes in the meter contour image; for each meter contour image in the meter contour image group, perform the following meter reading recognition steps: in response to determining that the area of the contour region of the meter contour image is less than a preset contour region area threshold, generate a contour region meter bounding box group; based on the contour region meter bounding box group, perform character recognition processing on the meter contour image to generate a character sequence; based on the character sequence, output a meter reading recognition image.

5. An electronic device, comprising: One or more processors; A storage device having one or more programs stored thereon; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1-3.

6. A computer-readable medium having a computer program stored thereon, wherein, The program, when executed by the processor, implements the method according to any one of claims 1-3.

Images