An electric power scene visual key information edge compression and transmission system

Abstract

The present application relates to the technical field of electric power engineering, in particular to a power scene visual key information edge compression and transmission system, comprising: when wireless network transmission is invalid or does not meet image data transmission requirements, loading band check variable length coding of each sampling time image into a carrier in the form of a millimeter wave reference signal for modulation to obtain a millimeter wave modulation signal; transmitting the millimeter wave modulation signal to a monitoring workstation through a millimeter wave transmission channel so that the monitoring workstation demodulates the millimeter wave modulation signal to obtain image edge information; when light intensity is weak or the power scene is blocked, processing multi-frame millimeter wave echo point cloud data to obtain intrusion or dangerous object information, and after fusion with corresponding timestamp image edge information, performing a non-maximum suppression operation to obtain target fusion information. The present application reduces the data capacity of transmission, guarantees the continuous transmission of information, enhances the adaptability of the system in a complex network environment, and maintains the safe and stable operation of the power system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric power engineering, in particular to an electric power scene visual key information edge compression and transmission system. BACKGROUND

[0002] In the electric power scene, with the wide application of visual monitoring devices, the visual management of the electric power system can provide more intuitive device operation status and environmental information; a large number of visual monitoring devices are installed in the existing electric power scene, which are usually installed on key electric power facilities to monitor the visual key information such as real-time visual data or image features of the electric power device operation status, environmental changes or fault warning, which can help monitoring personnel to discover potential problems in time and make quick response. However, although these visual monitoring devices can provide clear images under normal lighting conditions, they still face some challenges.

[0003] Firstly, the visual monitoring devices rely on public network to transmit video and image data, and in the electric power line operation environment, there is public network instability leading to public network fluctuation, resulting in data transmission interruption, delay or data packet loss in a specific area or time period, especially for image video data, the public network transmission capacity cannot meet the data volume demand, resulting in image quality degradation or information loss in the transmission process, so that the operation and maintenance personnel cannot obtain complete and accurate scene information in time, affecting fault judgment and processing, which may lead to expansion of fault range, power failure accident and operation and maintenance decision error, increase operation and maintenance cost and workload, reduce the reliability of the electric power system, and further affect the monitoring effect and decision support.

[0004] Secondly, the visual monitoring devices will fail or cause monitoring blind area in the case of insufficient light or physical obstruction, such as at night, in bad weather or in complex terrain area, the visual monitoring devices are difficult to obtain clear images or cover all key positions, resulting in missing of key information, so that the operation and maintenance personnel cannot effectively monitor the electric power device status and environmental changes, and it is difficult to prevent potential risks of the electric power device in advance.

[0005] In addition, even in the case of visible light, the transmission of image data may still be affected by transmission channel failure or limited transmission capacity, resulting in the decline of image data quality and transmission speed, so that the image data cannot be delivered to the monitoring center in time, hindering the normal operation of the automatic operation and maintenance system, causing the failure of the early warning function, and the hidden danger of the equipment is difficult to be discovered in time. SUMMARY

[0006] To this end, the technical problem to be solved by the present application is to overcome the problems in the prior art that the instability of the public network and the insufficient transmission capacity cause the image video data transmission to be interrupted, delayed or data to be lost, the insufficient light or physical obstruction causes the monitoring to be ineffective or blind areas to be generated, the transmission channel is invalid or the transmission channel capacity is limited, which causes the quality and transmission speed of the image data to be reduced, and further affects the safety and emergency response capability of the power equipment.

[0007] To solve the above technical problems, the present application provides a power scene visual key information edge-end compression and transmission system, comprising:

[0008] A terminal processor is arranged in the monitoring device and comprises:

[0009] An image processing module is configured to pre-process and extract edge information from each received sampling time image, obtain the edge information of each sampling time image, and then perform encoding and check code adding operations to obtain the check variable length coding of each sampling time image.

[0010] A state monitoring module is configured to send a first trigger signal to the millimeter wave transmitting component when the wireless network transmission between the monitoring device and the monitoring workstation is invalid or does not meet the image data transmission requirements.

[0011] A modulation module is configured to receive a millimeter wave reference signal, load the check variable length coding of each sampling time image onto the carrier wave for modulation processing by taking the millimeter wave reference signal as the carrier wave, and obtain a millimeter wave modulation signal.

[0012] A millimeter wave transmitting component is in communication connection with the terminal processor and is configured to receive the first trigger signal and transmit the millimeter wave reference signal.

[0013] A signal transmission module is configured to transmit the millimeter wave modulation signal to the monitoring workstation through the millimeter wave transmission channel, so that the monitoring workstation demodulates the millimeter wave modulation signal and obtains the edge information of each sampling time image.

[0014] Preferably, the monitoring workstation is connected to the monitoring device through the signal transmission module and comprises:

[0015] A signal receiving component is configured to receive the modulation signal and perform demodulation to obtain the check variable length coding of each sampling time image.

[0016] A check decoding module is configured to perform check and decoding operations on the check variable length coding of each sampling time image to obtain the edge information of each sampling time image.

[0017] The information processing module is configured to acquire abnormal information of power equipment and lines and environment and position information of an intruding or dangerous object in a preset power scene region according to edge information of images at adjacent sampling time points.

[0018] The early warning module is configured to generate early warning information according to the abnormal information of power equipment and lines and environment and the position information of the intruding or dangerous object in the preset power scene region.

[0019] The display module is configured to obtain non-edge information of each sampling time point image according to band-checked variable length coding of each sampling time point image by using an image recovery method, and display structural information and gradient change information of each sampling time point image.

[0020] Preferably, the acquiring of the abnormal information of power equipment and lines and environment and the position information of the intruding or dangerous object in the preset power scene region according to the edge information of images at adjacent sampling time points comprises:

[0021] The edge information of each sampling time point image comprises edge positions and edge amplitudes of each sampling time point image.

[0022] The edge positions of adjacent time point images are compared pixel by pixel to generate an edge position difference map, and an expression of the edge position difference map is:

[0023] ;

[0024] wherein, represents an edge position of a pixel point in the i-th sampling time point image; represents an edge position of a pixel point in the j-th sampling time point image; represents a time sequence difference value between the edge position of the pixel point in the i-th sampling time point image and the edge position of the pixel point in the j-th sampling time point image; If the time sequence difference value of the pixel point in the edge position difference map is greater than a first dynamic threshold value, and the pixel point

[0025] is located in a preset range of power equipment or power lines, it is determined that there is abnormal information of abnormal displacement or deformation of the power equipment or the power lines in the preset power scene region. If the time sequence difference value of the pixel point in the edge position difference map is greater than the first dynamic threshold value, and the pixel point

[0026] is located in the preset range of the power equipment or the power lines, it is determined that there is abnormal information of abnormal displacement or deformation of the power equipment or the power lines in the preset power scene region. If the time sequence difference value of the pixel point ​​If the edge position of the image at each sampling moment is 0, it is judged that there is an intrusion or dangerous object in the preset power scene area, and the pixel points in the image at the adjacent sampling moments are used to calculate the value of the image. edge position to obtain location information of intrusion or dangerous objects;

[0027] According to the edge amplitude of the image at adjacent moments, the amplitude difference of each pixel is calculated, and its expression is:

[0028] ;

[0029] in, Indicates the Pixels in the image at sampling time The edge amplitude of Indicates the Pixels in the image at sampling time The edge amplitude of Represents pixel points In the The edge amplitude of the image at the sampling moment and the The amplitude difference between the edge amplitudes of the image at each sampling moment;

[0030] If the pixel The amplitude difference of is greater than the second dynamic threshold, and , pixel If the device is within the preset range of the power equipment or power line, it is determined that there is abnormal information of corrosion or obstruction on the power equipment or power line in the preset power scene area;

[0031] If the pixel The amplitude difference of is greater than the second dynamic threshold, and , it is determined that there is an intrusion or dangerous object in the preset power scene area, and the location information of the intrusion or dangerous object is obtained;

[0032] According to the edge amplitude of the image at each sampling moment, the average edge amplitude at each sampling moment is calculated, and its expression is:

[0033] ;

[0034] in, Indicates the The average edge amplitude at each sampling moment; Indicates the The first sampling moment in the image pixels The edge amplitude of Indicates the total number of pixels;

[0035] According to the average edge amplitude of each sampling time, the amplitude decay index of each sampling time is calculated, and the expression is:

[0036] ;

[0037] Wherein, The amplitude decay index of the first sampling time is represented by the first sampling time; The amplitude decay index of the first sampling time is represented by the first sampling time; The amplitude decay index of the normal environment is represented by the normal environment;

[0038] If the amplitude decay index of different sampling times is less than the decay threshold, it is determined that there is environmental abnormal information in the preset power scene area.

[0039] Preferably, it further comprises:

[0040] The environmental monitoring module is placed in the terminal processor; when the gray mean value of each sampling time image is less than the first threshold value, or the gray variance of each sampling time image is less than the second threshold value, a second trigger signal is transmitted to the millimeter wave transmitting component, so that the millimeter wave transmitting component receives and emits millimeter wave radar signals;

[0041] The millimeter wave radar receiver is arranged in the preset range of the millimeter wave transmitting component; it is used to receive the millimeter wave echo signal reflected by the millimeter wave radar signal in each object in the power scene, and obtain multiple frames of millimeter wave echo point cloud data;

[0042] The data processing module is placed in the terminal processor; after the position, shape and structural features of the intruder or dangerous object at the end time of the millimeter wave transmitting window in the power scene are obtained by using the adjacent multi-frame superposition and clustering algorithm to process the multiple frames of millimeter wave echo point cloud data, the fusion algorithm is used to fuse the edge information of the corresponding time stamp image, and the fusion information at the end time of the millimeter wave transmitting window is obtained; the non-maximum suppression operation is performed on the fusion information to obtain the target fusion information; the target fusion information is encoded and added with check code to obtain the variable length coding with check of the target fusion information, so that the modulation module receives and modulates.

[0043] Preferably, the monitoring device is erected at the end of the power grid environment, and further comprises:

[0044] The camera is connected with the terminal processor through the image sensor interface; it is used to collect each sampling time image in the preset power scene area, and transmit each sampling time image to the terminal processor;

[0045] The terminal storage is connected with the camera through the high-speed storage interface; it is used to store and manage each sampling time image collected by the camera.

[0046] Preferably, the state monitoring module is further configured to send a processing signal to the modulation module when wireless network transmission between the monitoring device and the monitoring workstation is effective and meets the image data volume transmission requirement;

[0047] The modulation module is further configured to receive the processing signal and modulate the variable-length coding with check of each sampling-time image to obtain a network modulation signal.

[0048] The information transmission module is further configured to transmit the network modulation signal to the monitoring workstation through the wireless network between the monitoring device and the monitoring workstation, so that the monitoring workstation demodulates the network modulation signal to obtain the edge information of each sampling-time image.

[0049] Preferably, the system further comprises:

[0050] The signal relay component is disposed on the millimeter wave transmission channel and is configured to amplify and transmit the modulation signal over a long distance.

[0051] Preferably, the encoding and adding check code operation on the edge information of each sampling-time image to obtain the variable-length coding with check of each sampling-time image comprises:

[0052] The edge information of each sampling-time image comprises edge position and edge amplitude of each sampling-time image.

[0053] The edge position of each sampling-time image is encoded by using Huffman coding, and the edge amplitude of each sampling-time image is encoded by using a 3-bit symbol to obtain edge position encoding and edge amplitude encoding of each sampling-time image, and to form variable-length coding of each sampling-time image.

[0054] A check code is added on the basis of the variable-length coding of each sampling-time image to obtain the variable-length coding with check of each sampling-time image; and the check code is a Hamming code or a cyclic code.

[0055] Preferably, the preprocessing comprises denoising, filtering and direction correction of the collected image of each sampling-time; and the edge information extraction processing comprises using a sobel operator extraction method or a canny operator extraction method to extract the edge information in the power scene image of each sampling-time after preprocessing.

[0056] Preferably, the check and decoding operation comprises:

[0057] If the check has an error, the check decoding module returns an error prompt signal to the terminal processor; and if the check is qualified, the check decoding module decodes the variable-length coding with check of each sampling-time image by using a decoder to finally obtain the edge information of each sampling-time image.

[0058] The above technical solutions of the present application have the following beneficial effects compared with the prior art:

[0059] The power scene visual key information edge-end compression and transmission system provided by the present application realizes the extraction and compression of key information such as areas with obvious brightness changes and important structural attributes in the images of power equipment and its surrounding environment through image edge information extraction, thereby reducing the data capacity of transmission; the extracted edge information is encoded to obtain variable-length coding corresponding to the images, thereby reducing the occupation of network transmission resources and realizing efficient and accurate data transmission; at the same time, check codes are added in the variable-length coding transmission process, thereby effectively preventing the decrease of information accuracy caused by transmission errors; in addition, when wireless network transmission fails or fails to meet the transmission demand, millimeter wave signals are used as transmission media to realize emergency transmission of key information, thereby ensuring the continuous transmission of information and enhancing the adaptability of the system in complex network environments; at the same time, signal relay components are added on the millimeter wave transmission path for the case of large signal attenuation to realize the amplification and long-distance transmission of modulated signal information; the millimeter wave recognition function is introduced to fuse the millimeter wave radar data and image edge information through a fusion algorithm in a scene with weak light intensity or large obstruction, thereby improving the monitoring capability of the system under different environmental conditions and ensuring the all-around effective monitoring of power equipment and lines, surrounding environment, and intrusions or dangerous objects; finally, strict checking and decoding operations are performed at the receiving end to ensure the reliability of the data, thereby providing a solid foundation for subsequent accurate analysis of the state of power equipment and environmental changes, which helps the operation and maintenance personnel to discover potential problems in time and take corresponding measures to maintain the safe and stable operation of the power system. BRIEF DESCRIPTION OF DRAWINGS

[0060] In order to make the content of the present application more easily understood, the present application will be further described in detail below according to specific embodiments of the present application and in conjunction with the drawings, in which:

[0061] Figure 1 is a schematic diagram of a power scene visual key information edge-end compression and transmission system provided by the present application; wherein, Figure 1 The dashed part in indicates an additional operation, which can be selected according to the actual situation of the power scene whether to perform the operation;

[0062] Figure 2 is a structural schematic diagram of a terminal processor;

[0063] Figure 3 is a structural schematic diagram of a monitoring workstation. DETAILED DESCRIPTION

[0064] The application will be further described below in conjunction with the drawings and specific embodiments so that those skilled in the art can better understand and implement the application, but the embodiments are not intended to limit the application.

[0065] Referring to Figure 1 as shown, Figure 1 is a schematic diagram of an electric power scene visual key information edge compression and transmission system provided by the application; specifically comprising:

[0066] The camera is arranged in the monitoring device erected at the end of the power grid environment and is connected to the terminal processor through an image sensor interface; used for collecting images at each sampling time in the preset electric power scene area and transmitting each sampling time image to the terminal processor;

[0067] The terminal storage is arranged in the monitoring device erected at the end of the power grid environment and is connected to the camera through a high-speed storage interface; used for storing and managing the images at each sampling time collected by the camera, so as to realize data retention and management;

[0068] The terminal processor is arranged in the monitoring device erected at the end of the power grid environment and comprises:

[0069] The image processing module is used for pre-processing and edge information extraction processing of each received sampling time image, obtaining the edge information of each sampling time image, and then performing encoding and adding check code operation to obtain the variable length coding with check code of each sampling time image.

[0070] The pre-processing is denoising, filtering and direction correction operation on the collected image at each sampling time; the edge information extraction processing is extracting the edge information in the electric power scene image at each sampling time after pre-processing by using the sobel operator extraction method or the canny operator extraction method; wherein the edge information extraction processing mainly uses the first derivative form such as the sobel operator or the second derivative form such as the canny operator, so as to realize the extraction of the information such as the obvious brightness change area and important structure attribute in the image;

[0071] The encoding and adding check code operation on the edge information of each sampling time image to obtain the variable length coding with check code of each sampling time image comprises:

[0072] The edge information of each sampling time image includes the edge position and edge amplitude of each sampling time image;

[0073] The edge position of each sampling time image is encoded by using Huffman coding, and the edge amplitude of each sampling time image is encoded by using 3-bit symbols to obtain the edge position encoding and the edge amplitude encoding of each sampling time image, and to form the variable-length encoding of each sampling time image.

[0074] On the basis of the variable-length encoding of each sampling time image, a check code is added to obtain the variable-length encoding with check code of each sampling time power scene; wherein the check code is a Hamming code or a cyclic code.

[0075] Wherein, encoding the edge information helps to realize efficient and accurate data transmission, and occupies less network transmission resource; at the same time, in order to prevent the variable-length encoding from appearing errors in transmission and causing the information accuracy to decline, a check code is added to the variable-length encoding.

[0076] The state monitoring module is configured to send a first trigger signal to the millimeter wave transmitting component when the wireless network transmission between the monitoring device and the monitoring workstation is invalid or does not meet the image data transmission requirements; and is further configured to send a processing signal to the modulation module when the wireless network transmission between the monitoring device and the monitoring workstation is valid and meets the image data transmission requirements.

[0077] The modulation module is configured to receive the millimeter wave reference signal, and modulate the variable-length encoding with check code of each sampling time image on the carrier wave to obtain the millimeter wave modulation signal; and is further configured to receive the processing signal and modulate the variable-length encoding with check code of each sampling time image to obtain the network modulation signal.

[0078] The environmental monitoring module is configured to send a second trigger signal to the millimeter wave transmitting component when the gray mean value of each sampling time image is less than a first threshold value or the gray variance of each sampling time image is less than a second threshold value, so that the millimeter wave transmitting component receives and sends the millimeter wave radar signal; wherein, if the gray mean value of each sampling time image is less than the first threshold value, it indicates that the light intensity is weak, otherwise it indicates that the light intensity is strong; if the gray variance of each sampling time image is less than the second threshold value, it indicates that the power scene is blocked, otherwise it indicates that the power scene is not blocked.

[0079] The data processing module is used for processing multiple frames of millimeter wave echo point cloud data by using adjacent multi-frame superposition and clustering algorithm, obtaining the position, shape and structural features of the intrusion or dangerous object at the end time of the millimeter wave transmission window in the power scene, and fusing the edge information of the corresponding time stamp image by using a fusion algorithm to obtain the fusion information at the end time of the millimeter wave transmission window; performing a non-maximum suppression operation on the fusion information to obtain target fusion information; performing an encoding and check code adding operation on the target fusion information to obtain a check variable length encoding of the target fusion information, so as to be received and modulated by the modulation module; wherein the millimeter wave transmission window is a continuous collection time interval of adjacent multiple frames of millimeter wave echo point cloud data; the corresponding time stamp image is a collection image at the end time of the millimeter wave transmission window; the end time of the millimeter wave transmission window is aligned with the image sampling time; and the fusion algorithm is an improved pointnet++ algorithm model.

[0080] The structure diagram of the terminal processor is as shown in Figure 2 The terminal processor is powered by the power consumption generated by the operation of the photovoltaic panel and the battery.

[0081] The millimeter wave transmission component is in communication connection with the terminal processor; and is used for receiving a first trigger signal and transmitting a millimeter wave reference signal.

[0082] The millimeter wave radar receiver is arranged in a preset range of the millimeter wave transmission component; and is used for receiving millimeter wave echo signals reflected by the millimeter wave radar signals reflected by each object in the power scene to obtain multiple frames of millimeter wave echo point cloud data.

[0083] The signal transmission module is used for transmitting the millimeter wave modulated signal to the monitoring workstation through the millimeter wave transmission channel, so that the monitoring workstation demodulates the millimeter wave modulated signal to obtain the edge information of each sampling time image; and is also used for transmitting the network modulated signal to the monitoring workstation through the wireless network between the monitoring device and the monitoring workstation, so that the monitoring workstation demodulates the network modulated signal to obtain the edge information of each sampling time image; wherein, a signal relay component is further arranged on the millimeter wave transmission channel, which is used for amplifying and long-distance transmitting the modulated signal.

[0084] The monitoring workstation is connected with the monitoring device through the signal transmission module, and comprises:

[0085] The signal receiving component is used for receiving the modulated signal and performing a demodulation operation to obtain a check variable length encoding of each sampling time image.

[0086] The check decoding module is configured to perform a check and decoding operation on the variable-length coding with check of each image at a sampling time to obtain edge information of each image at the sampling time; wherein the check and decoding operation comprises: if there is an error in the check, the check decoding module returns an error prompt signal to the terminal processor; if the check is qualified, the check decoding module decodes the variable-length coding with check of each image at the sampling time by using a decoder, and finally obtains the edge information of each image at the sampling time;

[0087] The information processing module is configured to obtain abnormal information of power equipment, lines and environment and position information of an intruding or dangerous object in a preset power scene region according to the edge information of adjacent images at sampling times;

[0088] The early warning module is configured to generate early warning information according to the abnormal information of power equipment, lines and environment and the position information of the intruding or dangerous object in the preset power scene region, so as to prompt an operation and maintenance personnel;

[0089] The information processing module is configured to obtain abnormal information of power equipment, lines and environment and position information of an intruding or dangerous object in a preset power scene region according to the edge information of adjacent images at sampling times;

[0090] The edge information of each image at the sampling time comprises an edge position and an edge amplitude of each image at the sampling time;

[0091] The edge positions of adjacent images at times are compared pixel by pixel to generate an edge position difference image, and an expression of the edge position difference image is:

[0092] ;

[0093] wherein, denotes an edge position of a pixel point in the i-th image at the sampling time; denotes an edge position of a pixel point in the j-th image at the sampling time; denotes a time sequence difference value between the edge position of the pixel point in the i-th image at the sampling time and the edge position of the pixel point in the j-th image at the sampling time;

[0094] If the time sequence difference value of the pixel point in the edge position difference image is greater than a first dynamic threshold value, and the pixel point is located in a preset range of power equipment or a power line, it is determined that there is abnormal displacement or deformation of the power equipment or the power line in the preset power scene region;

[0095] ​​​If the pixel point in the edge position difference map The time difference value of is greater than the first dynamic threshold, and the pixel point In the If the edge position of the image at each sampling moment is 0, it is judged that there is an intrusion or dangerous object in the preset power scene area, and the pixel points in the image at the adjacent sampling moments are used to calculate the value of the image. edge position to obtain location information of intrusion or dangerous objects;

[0096] According to the edge amplitude of the image at adjacent moments, the amplitude difference of each pixel is calculated, and its expression is:

[0097] ;

[0098] in, Indicates the Pixels in the image at sampling time The edge amplitude of Indicates the Pixels in the image at sampling time The edge amplitude of Represents pixel points In the The edge amplitude of the image at the sampling moment and the The amplitude difference between the edge amplitudes of the image at each sampling moment;

[0099] If the pixel The amplitude difference of is greater than the second dynamic threshold, and , pixel If the device is within the preset range of the power equipment or power line, it is determined that there is abnormal information of corrosion or obstruction on the power equipment or power line in the preset power scene area;

[0100] If the pixel The amplitude difference of is greater than the second dynamic threshold, and , it is determined that there is an intrusion or dangerous object in the preset power scene area, and the location information of the intrusion or dangerous object is obtained;

[0101] According to the edge amplitude of the image at each sampling moment, the average edge amplitude at each sampling moment is calculated, and its expression is:

[0102] ;

[0103] in, Indicates the The average edge amplitude at each sampling moment; Indicates the The first sampling moment in the image pixels The edge amplitude of Indicates the total number of pixels;

[0104] According to the average edge amplitude at each sampling moment, the amplitude attenuation index at each sampling moment is calculated, and its expression is:

[0105] ;

[0106] in, Indicates the Amplitude attenuation index at each sampling moment; Indicates the normal environment amplitude attenuation index;

[0107] If the amplitude attenuation index at different sampling moments is less than the attenuation threshold, it is determined that there is environmental abnormality information in the preset power scenario area;

[0108] A display module is used to obtain non-edge information of the image at each sampling moment based on the variable-length coding with checksum of the image at each sampling moment using an image restoration method, and to display structural information and gradient change information of the image at each sampling moment;

[0109] Among them, the structural diagram of the monitoring workstation is as follows Figure 3 shown.

[0110] The present invention provides a power scene visual key information edge compression and transmission system. This system extracts and compresses key information by extracting image edge information. When wireless network information failure or transmission failure is detected, millimeter wave signals are used as the transmission medium to achieve emergency transmission of key information. Specifically, when wireless network transmission between a monitoring device and a monitoring workstation is invalid or does not meet image data transmission requirements, the image processing module, status monitoring module, modulation module, millimeter wave transmitting component, and signal transmission module in the terminal processor transmit the millimeter wave modulated signal to the monitoring workstation via a millimeter wave transmission channel. This solves the problems of image and video data transmission interruption, delay, or data packet loss caused by public network instability and insufficient transmission capacity, as well as image data quality and transmission speed degradation caused by transmission channel failure or limited transmission channel capacity. This prevents operation and maintenance personnel from obtaining complete and accurate scene information in a timely manner, hindering the normal operation of the automated operation and maintenance system, causing the failure of early warning functions, making it difficult to timely discover equipment hazards, increasing operation and maintenance costs and workload, reducing power system reliability, and thus affecting monitoring effectiveness and decision support. The key information represents abnormal information about power equipment, lines, and the environment within a preset power scene area, as well as location information of intrusions or dangerous objects.

[0111] In addition, the application also considers the reinforcement of millimeter wave radar data, and applies the millimeter wave radar data to scenes with weak light intensity and large occlusion, and only processes image data for good lighting conditions; when millimeter wave radar data is added for support, the collected millimeter wave echo point cloud data needs to be processed, and since the millimeter wave radar data is relatively sparse, the application adopts clustering processing after superimposing adjacent multiple frames to detect the position and structural form of key targets, that is, to detect the position, shape and structural features of the intruding or dangerous object; then an improved pointnet++ algorithm model is used to fuse the position, shape and structural features of the intruding or dangerous object with the image edge information, and a non-maximum suppression operation is performed on the fusion result to obtain target fusion information, that is, through the environment monitoring module and the data processing module in the terminal processor, and the millimeter wave transmitting part and the millimeter wave radar receiver, the multiple frame millimeter wave echo point cloud data is processed to obtain the position, shape and structural features of the intruding or dangerous object, and then the target fusion information is obtained, solving the problem that insufficient light or physical occlusion causes monitoring failure or monitoring blind area, resulting in missing of key information, so that the operation and maintenance personnel cannot effectively monitor the state of power equipment and environmental changes, and it is difficult to prevent potential risks of power equipment in advance.

[0112] Obviously, the above embodiments are only examples for clearly illustrating, and are not limited to the embodiments. For those skilled in the art, other different forms of changes or variations can be made on the basis of the above description. Here, all the embodiments need not and cannot be exhausted. The obvious changes or variations derived therefrom are still within the protection scope of the present application.

Claims

1. A power scene visual key information edge compression and transmission system, characterized by: include: The terminal processor is installed in the monitoring device and includes: The image processing module is used to pre-process and extract edge information of the received image at each sampling moment, and then encode and add a checksum to obtain a variable-length code with checksum for the image at each sampling moment after obtaining the edge information of the image at each sampling moment; A status monitoring module is configured to send a first trigger signal to the millimeter wave transmitting component when wireless network transmission between the monitoring device and the monitoring workstation is invalid or does not meet image data transmission requirements; The modulation module is used to receive the millimeter wave reference signal and use the millimeter wave reference signal as a carrier. The variable-length code with checksum of the image at each sampling moment is loaded onto the carrier for modulation processing to obtain a millimeter wave modulated signal. A millimeter wave transmitting component, communicatively connected to the terminal processor; configured to receive a first trigger signal and transmit a millimeter wave reference signal; The signal transmission module is used to transmit the millimeter wave modulated signal to the monitoring workstation through the millimeter wave transmission channel, so that the monitoring workstation can demodulate the millimeter wave modulated signal and obtain the edge information of the image at each sampling moment.

2. The edge compression and transmission system for visual key information of electric power scenes according to claim 1 is characterized in that: The monitoring workstation is connected to the monitoring device through a signal transmission module and includes: Signal receiving component; used to receive the modulated signal and perform demodulation operation to obtain the variable-length code with checksum of the image at each sampling moment; The check decoding module is used to perform check and decoding operations on the variable-length code with check of the image at each sampling moment to obtain the edge information of the image at each sampling moment; An information processing module is used to obtain abnormal information about power equipment, lines, and the environment within a preset power scene area, as well as location information of intrusions or dangerous objects based on edge information of images at adjacent sampling moments; An early warning module is used to generate early warning information based on abnormal information of power equipment, lines and environment in a preset power scenario area, as well as location information of intrusion or dangerous objects; The display module is used to obtain the non-edge information of the image at each sampling moment according to the variable-length coding with checksum of the image at each sampling moment by using the image restoration method, and to display the structural information and gradient change information of the image at each sampling moment.

3. The edge compression and transmission system for visual key information of electric power scenes according to claim 2 is characterized in that: The method of obtaining abnormal information of power equipment, lines, and environment and location information of intrusion or dangerous objects in a preset power scene area based on edge information of images at adjacent sampling moments includes: The edge information of the image at each sampling moment includes the edge position and edge amplitude of the image at each sampling moment; The edge positions of the images at adjacent moments are compared pixel by pixel to generate an edge position difference map, which is expressed as: ； in, Indicates the Pixels in the image at sampling time The edge position; Indicates the Pixels in the image at sampling time The edge position; Represents pixel points In the The edge position of the image at the sampling moment and the The time difference value between the edge positions of the image at each sampling moment; If the pixel point in the edge position difference map The time difference value of is greater than the first dynamic threshold, and the pixel point If the device is within a preset range of the power equipment or power line, it is determined that the power equipment or power line in the preset power scene area has abnormal displacement or deformation; If the pixel point in the edge position difference map The time difference value of is greater than the first dynamic threshold, and the pixel point In the If the edge position of the image at each sampling moment is 0, it is judged that there is an intrusion or dangerous object in the preset power scene area, and the pixel points in the image at the adjacent sampling moments are used to calculate the value of the image. edge position to obtain location information of intrusion or dangerous objects; According to the edge amplitude of the image at adjacent moments, the amplitude difference of each pixel is calculated, and its expression is: ； in, Indicates the Pixels in the image at sampling time The edge amplitude of Indicates the Pixels in the image at sampling time The edge amplitude of Represents pixel points In the The edge amplitude of the image at the sampling moment and the The amplitude difference between the edge amplitudes of the image at each sampling moment; If the pixel The amplitude difference of is greater than the second dynamic threshold, and , pixel If the device is within the preset range of the power equipment or power line, it is determined that there is abnormal information of corrosion or obstruction on the power equipment or power line in the preset power scene area; If the pixel The amplitude difference of is greater than the second dynamic threshold, and , it is determined that there is an intrusion or dangerous object in the preset power scene area, and the location information of the intrusion or dangerous object is obtained; According to the edge amplitude of the image at each sampling moment, the average edge amplitude at each sampling moment is calculated, and its expression is: ； in, Indicates the The average edge amplitude at each sampling moment; Indicates the The first sampling moment in the image pixels The edge amplitude of Indicates the total number of pixels; According to the average edge amplitude at each sampling moment, the amplitude attenuation index at each sampling moment is calculated, and its expression is: ； in, Indicates the Amplitude attenuation index at each sampling moment; Indicates the normal environment amplitude attenuation index; If the amplitude attenuation indexes at different sampling moments are all less than the attenuation threshold, it is determined that there is environmental abnormality information in the preset power scenario area.

4. The edge compression and transmission system for visual key information of electric power scenes according to claim 1 is characterized in that: Also includes: Environmental monitoring module, placed in the terminal processor; for transmitting a second trigger signal to the millimeter wave transmitting component when the grayscale mean of the image at each sampling moment is less than a first threshold, or when the grayscale variance of the image at each sampling moment is less than a second threshold, so that the millimeter wave transmitting component receives the second trigger signal and emits a millimeter wave radar signal; A millimeter-wave radar receiver is provided within a preset range of the millimeter-wave transmitting component; it is used to receive millimeter-wave echo signals reflected by various objects in the power scene from the millimeter-wave radar signal, and obtain multi-frame millimeter-wave echo point cloud data; Data processing module, placed in the terminal processor; It is used to process multi-frame millimeter wave echo point cloud data using adjacent multi-frame superposition and clustering algorithms to obtain the position, shape, and structural characteristics of intrusion or dangerous objects at the end of the millimeter wave transmission window in the power scene. Then, it uses a fusion algorithm to fuse this with the edge information of the corresponding time-stamp image to obtain the fused information at the end of the millimeter wave transmission window; A non-maximum suppression operation is performed on the fused information to obtain target fused information; and a check code operation is performed on the target fused information to obtain a variable-length code with checksum for the target fused information so that the modulation module can receive and perform modulation processing.

5. The edge compression and transmission system for visual key information of electric power scenes according to claim 1 is characterized in that: The monitoring device, installed at the end of the power grid environment, also includes: The camera is connected to the terminal processor through an image sensor interface; it is used to collect images at each sampling moment within the preset power scene area and transmit each sampling moment image to the terminal processor; The terminal memory is connected to the camera through a high-speed storage interface; it is used to store and manage images collected by the camera at each sampling moment.

6. The edge compression and transmission system for visual key information of electric power scenes according to claim 1 is characterized in that: The state monitoring module is further configured to send a processing signal to the modulation module when the wireless network transmission between the monitoring device and the monitoring workstation is effective and meets the image data transmission requirements; The modulation module is further used to receive and process the signal, and modulate the variable length code with checksum of the image at each sampling moment to obtain a network modulation signal; The signal transmission module is also used to transmit the network modulated signal to the monitoring workstation via the wireless network between the monitoring device and the monitoring workstation, so that the monitoring workstation demodulates the network modulated signal and obtains the edge information of the image at each sampling moment.

7. The edge compression and transmission system for visual key information of electric power scenes according to claim 1 is characterized in that: Also includes: Signal relay components, placed on the millimeter wave transmission channel; Used to amplify and transmit modulated signals over long distances.

8. The edge compression and transmission system for visual key information of electric power scenes according to claim 1 is characterized in that: The edge information of the image at each sampling moment is encoded and a check code is added to obtain the variable-length code with checksum for the image at each sampling moment, including: The edge information of the image at each sampling moment includes the edge position and edge amplitude of the image at each sampling moment; Huffman coding is used to encode the edge position of the image at each sampling moment, and 3-bit code elements are used to encode the edge amplitude of the image at each sampling moment, thereby obtaining the edge position code and edge amplitude code of the image at each sampling moment, and forming a variable-length code for the image at each sampling moment; A check code is added to the variable-length code of the image at each sampling moment to obtain a check-coded variable-length code of the power scene at each sampling moment; wherein the check code is a Hamming code or a cyclic code.

9. The edge compression and transmission system for visual key information of electric power scenes according to claim 1 is characterized in that: The preprocessing is to perform denoising, filtering, and direction correction operations on the collected image at each sampling moment; the edge information extraction processing is to use the Sobel operator extraction method or the Canny operator extraction method to extract the edge information in the power scene image at each sampling moment after preprocessing.

10. The edge compression and transmission system for visual key information of electric power scenes according to claim 1 is characterized in that: The checking and decoding operations include: If there is an error in the check, the decoding module is checked and an error prompt signal is returned to the terminal processor; if the check is qualified, the decoding module is checked and the decoder is used to decode the variable-length code with check of the image at each sampling moment, and finally the edge information of the image at each sampling moment is obtained.

Images