CN113381510B - One-key sequential control video double-confirmation system for transformer substation - Google Patents
One-key sequential control video double-confirmation system for transformer substation Download PDFInfo
- Publication number
- CN113381510B CN113381510B CN202110616608.7A CN202110616608A CN113381510B CN 113381510 B CN113381510 B CN 113381510B CN 202110616608 A CN202110616608 A CN 202110616608A CN 113381510 B CN113381510 B CN 113381510B
- Authority
- CN
- China
- Prior art keywords
- controllable end
- switch tube
- switching tube
- controllable
- tube
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00001—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by the display of information or by user interaction, e.g. supervisory control and data acquisition systems [SCADA] or graphical user interfaces [GUI]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00002—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by monitoring
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
- H02J13/00006—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by information or instructions transport means between the monitoring, controlling or managing units and monitored, controlled or operated power network element or electrical equipment
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
- Y02B90/20—Smart grids as enabling technology in buildings sector
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S40/00—Systems for electrical power generation, transmission, distribution or end-user application management characterised by the use of communication or information technologies, or communication or information technology specific aspects supporting them
- Y04S40/12—Systems for electrical power generation, transmission, distribution or end-user application management characterised by the use of communication or information technologies, or communication or information technology specific aspects supporting them characterised by data transport means between the monitoring, controlling or managing units and monitored, controlled or operated electrical equipment
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Human Computer Interaction (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention relates to a transformer substation one-key sequential control video double-confirmation system, which comprises a camera device, a control device and an alarm device, wherein the camera device is connected with the control device; the control device comprises a target position determining module, a target behavior specification confirming module and an object identifying module; the target position determining module is used for determining whether the target stays in the specified area or leaves the specified area; the target behavior specification confirming module is used for receiving the result that the target position confirming module confirms that the target is in the designated area, automatically detecting the target object through deep learning, determining whether the target behavior is normal or not, and if the target behavior is not normal, sending a signal to the alarm device to trigger alarm; the object identification module realizes the safe identification of the object based on the cascade model.
Description
Technical Field
The invention belongs to the technical field of transformer substation monitoring, and particularly relates to a transformer substation one-key sequential control video double-confirmation system.
Background
The natural gas transmission and distribution system provides emergency storage, liquefaction and gasification peak regulation functions, and has important significance for guaranteeing urban fuel gas storage and safe and stable gas supply. Because the reservoir area storage medium is mainly liquefied natural gas, its material has dangers such as inflammability, chemical explosion, physical explosion, suffocation, frostbite, and the geographical position borders on complicated areas of stream such as park, has great fire risk and the invasion risk hidden danger of external personnel perimeter. The number of flying birds and miscellaneous trees outside the defense area is large, the false alarm rate is high, and in order to prevent frequent false alarms from causing the lacked psychology of operators, the perimeter intrusion alarm accuracy rate needs to be improved, and foreign people need to be prevented from entering a factory and being damaged. In order to prevent the staff entering the factory from having a lacked mind, the safety helmet and the work clothes are not worn according to the requirements, whether the staff entering the factory wears the safety helmet, the work clothes and the work shoes according to the regulations is required to be detected, the ignition source is stopped, and accidents are prevented. In order to prevent people who enter the substation from having the spirit and luck and lose money, smoking and calling in violation of rules and lucks, ignition sources need to be stopped, accidents are prevented, and the like, so that the safety of the transformer substation is improved.
Disclosure of Invention
The invention discloses a transformer substation one-key sequential control video double-confirmation system, which comprises a camera device, a control device and an alarm device, wherein the camera device is connected with the control device; the control device comprises a target position determining module, a target behavior specification confirming module and an object identifying module; the target position determining module is used for determining whether the target stays in the specified area or leaves the specified area; the target behavior specification confirming module is used for receiving the result that the target position confirming module confirms that the target is in the designated area, automatically detecting a target object through deep learning, determining whether the target behavior is in specification or not, and if the target behavior is not in specification, sending a signal to the alarm device to trigger alarm; the object identification module realizes the safe identification of the object based on the cascade model.
In the transformer substation one-key sequential control video double-confirmation system, the target position determination module receives video data acquired by the camera device, judges the target position by adopting a sliding window smooth statistical method, and if a target appears in a video frame, the target is considered to exist; on the contrary, if no target exists in the video frame, the target is considered to leave; the method comprises the steps of recording the occurrence condition of a target object in a plurality of recent frames by a sliding window smooth statistical method, considering that the target exists only when the number of frames in which the target is detected exceeds a certain proportion, and considering that the target leaves when the number of frames in which the target is detected is lower than the certain proportion, so that the influence of the error identification result of individual frames on program judgment can be avoided.
According to the transformer substation one-key sequential control video double-confirmation system, a target behavior specification confirmation module forms a data set by acquiring images of a tank car, workers, an anti-skid pad and a warning board, and a deep learning model for detecting objects of the tank car, the workers, the anti-skid pad and the warning board is obtained by training by using a deep learning target detection method; then, designing a behavior specification detection algorithm, starting to detect the objects of the working personnel, the non-slip mat and the warning board after detecting that the tank car enters the designated area, and finishing the operation if detecting that the target object appears in the designated area; if the target object is not detected within the specified time, the operation is not finished according to the standard, and an alarm signal is sent to the alarm device to trigger the alarm.
According to the transformer substation one-key sequential control video double-confirmation system, false alarm of object recognition caused by field station internal equipment and background is reduced through a cascade model, the cascade model comprises a first-stage model and a second-stage model, the first-stage model is firstly input into a light and rapid single-stage target detection model (YOLOV 4-tiny), the model only detects whether a person appears in a picture, once the person appears in the picture, an algorithm cuts off a region of the person in the picture, and then the cut region is sent to the second-stage model; after receiving the human body cutting area, the second-level model is wholly enlarged to 416x416 pixel size, and then target detection is carried out; the isomorphic first-stage model eliminates the interference of the background, so that the attention of the model is focused on the human body, and the misinformation can be greatly reduced; in addition, the human body area is cut out and then is amplified, so that the details of the human body can be better seen.
The transformer substation one-key sequential control video double-confirmation system also comprises a rechecking device for checking alarm information, wherein the rechecking device comprises a four-wheel-drive mobile module, a camera, a controller, a position checking module and a data transceiving module; the rechecking device confirms whether the operation which is not executed according to the standard still exists within a certain time when the alarm device sends the alarm signal, and if the operation which is not executed according to the standard still exists, the rechecking device starts the rechecking device after receiving the starting signal; the four-wheel-drive moving module comprises four drivers, four detectors, four wheels and an information processing device connected with the detectors, wherein the detectors are used for detecting the position information of the movement of the wheels, and the information processing device is used for processing signals detected by the detectors so as to determine the movement deviation between two front wheels or two rear wheels of the four-wheel drive; the driver is used for driving the wheels according to the control command of the controller; the position checking module is used for receiving the image or video information with the alarm, determining the position and shooting angle of an alarm camera for shooting the image or video information with the alarm, controlling the rechecking device to move to a matched position according to the position and shooting angle, controlling the shooting angle to be the same, carrying out first shooting, comparing the video or image shot for the first time with the video or image of the alarm camera, determining whether the position and shooting angle are the same, if not, readjusting the position and shooting angle of the rechecking device, carrying out second shooting, and comparing the video or image shot for the second time with the video or image of the alarm camera until the position and shooting angle are the same; if the position and the shooting angle are the same, determining whether the video or image content in the rechecking device is not standard, if so, manually intervening to eliminate the non-standard content, and if not, overhauling the alarm camera or manually intervening to confirm whether the non-standard content exists; the data receiving and sending module is used for data transmission.
The transformer substation one-key sequential control video double-confirmation system comprises an information processing device and a control device, wherein the information processing device comprises: the first non-controllable ends of the switch tubes M4-M6 are connected with a power supply VDD, the controllable end of the switch tube M4 is connected with the controllable end of the switch tube M5, the controllable end of the switch tube M6, the controllable ends of the switch tubes M7-M8, the controllable end of the switch tube M5C, the controllable end of the switch tube M6C, the controllable end of the switch tube M7C, the controllable end of the switch tube M8C and the second non-controllable end of the switch tube M4, the second non-controllable end of the switch tube M4 is connected with the first end of a current source ISET, the second end of the current source ISET is grounded, and the substrate end of the switch tube M4 is connected with the first non-controllable end of the switch tube M4; the second non-controllable end of the switch tube M5 is respectively connected with the first non-controllable end of the switch tube M1 and the controllable ends of the switch tubes M13-M14, the first bias grid end of the switch tube M1 and the first bias grid end of the switch tube M2 are connected with a power supply VDD, the control end of the switch tube M1 is grounded, the second bias grid end of the switch tube M1 is connected with a left front wheel detection signal, the substrate end of the switch tube M1 is connected with the substrate end of the switch tube M2, the substrate end of the switch tube M3 and the second non-controllable end of the switch tube M3, a second non-controllable end of the switching tube M1 is respectively connected with a second non-controllable end of the switching tube M2 and a first non-controllable end of the switching tube M3, the second non-controllable end of the switching tube M3 is grounded, a substrate end of the switching tube M6 is connected with a first non-controllable end of the switching tube M6, a second non-controllable end of the switching tube M6 is respectively connected with a controllable end of the switching tube M3 and a first non-controllable end of the switching tube M2, a controllable end of the switching tube M2 is respectively connected with a first end of a resistor Rt, a second non-controllable end of a switching tube M9 and a first non-controllable end of a switching tube M11, and a second bias of the switching tube M2 is connected with a right front wheel detection signal;
the first non-controllable end of the switch tube M7-M8 is connected with a power supply VDD, the controllable end of the switch tube M7 is connected with the controllable end of the switch tube M8, the substrate end of the switch tube M7 is connected with the first non-controllable end of the switch tube M7 and the substrate end of the switch tube M9, the second non-controllable end of the switch tube M7 is connected with the first non-controllable end of the switch tube M9, the controllable end of the switch tube M9 is grounded, the second non-controllable end of the switch tube M9 is connected with the control end of the switch tube M2 and the first non-controllable end of the switch tube M11, the substrate end of the switch tube M11 is connected with the substrate end of the switch tube M13 and the second non-controllable end of the switch tube M13, the controllable end of the switch tube M11 is connected with the controllable end of the switch tube M12 and the power supply VDD, a second non-controllable end of the switching tube M11 is connected with a first non-controllable end of the switching tube M13, a controllable end of the switching tube M13 is connected with a controllable end of the switching tube M14 and a second non-controllable end of the switching tube M5, the second non-controllable end of the switching tube M13 is grounded, a substrate end of the switching tube M8 is respectively connected with a first non-controllable end of the switching tube M8 and a substrate end of the switching tube M10, the second non-controllable end of the switching tube M8 is connected with a first non-controllable end of the switching tube M10, the second non-controllable end of the switching tube M10 is connected with a first output end and a first non-controllable end of the switching tube M12, the substrate end of the switching tube M12 is connected with the substrate end of the switching tube M14 and the second non-controllable end of the switching tube M14, and the second non-controllable end of the switching tube M12 is connected with the first non-controllable end of the switching tube M14;
the first non-controllable end of the switch tube M5C-M6C is connected with a power supply VDD, the controllable end of the switch tube M5C is connected with the controllable end of the switch tube M6C, the second non-controllable end of the switch tube M5C is respectively connected with the first non-controllable end of the switch tube M1C and the controllable ends of the switch tubes M13C-M14C, the first bias grid end of the switch tube M1C and the first bias grid end of the switch tube M2C are connected with the power supply VDD, the control end of the switch tube M1C is grounded, the second bias grid end of the switch tube M1C is connected with the left rear wheel detection signal, the substrate end of the switch tube M1C is connected with the substrate end of the switch tube M2C, the substrate end of the switch tube M3C and the second non-controllable end of the switch tube M3C, a second non-controllable end of the switching tube M1C is connected with a second non-controllable end of the switching tube M2C and a first non-controllable end of the switching tube M3C respectively, a second non-controllable end of the switching tube M3C is grounded, a substrate end of the switching tube M6C is connected with a first non-controllable end of the switching tube M6C, a second non-controllable end of the switching tube M6C is connected with a controllable end of the switching tube M3C and a first non-controllable end of the switching tube M2C respectively, a controllable end of the switching tube M2C is connected with a second end of the resistor Rt, a second non-controllable end of the switching tube M9C and a first non-controllable end of the switching tube M11C respectively, and a second bias grid end of the switching tube M2C is connected with a right rear wheel detection signal;
the first non-controllable end of the switch tube M7C-M8C is connected with a power supply VDD, the controllable end of the switch tube M7C is connected with the controllable end of the switch tube M8C, the substrate end of the switch tube M7C is connected with the first non-controllable end of the switch tube M7C and the substrate end of the switch tube M9C, the second non-controllable end of the switch tube M7C is connected with the first non-controllable end of the switch tube M9C, the controllable end of the switch tube M9C is grounded, the second non-controllable end of the switch tube M9C is connected with the control end of the switch tube M2C and the first non-controllable end of the switch tube M11C, the substrate end of the switch tube M11C is connected with the substrate end of the switch tube M13C and the second non-controllable end of the switch tube M13C, the controllable end of the switch tube M11C is connected with the controllable end of the switch tube M12C and the power supply VDD, the second non-controllable end of the switching tube M11C is connected to the first non-controllable end of the switching tube M13C, the controllable end of the switching tube M13C is connected to the controllable end of the switching tube M14C and the second non-controllable end of the switching tube M5C, the second non-controllable end of the switching tube M13C is grounded, the substrate end of the switching tube M8C is connected to the first non-controllable end of the switching tube M8C and the substrate end of the switching tube M10C respectively, the second non-controllable end of the switching tube M8C is connected to the first non-controllable end of the switching tube M10C, the second non-controllable end of the switching tube M10C is connected to the second output end and the first non-controllable end of the switching tube M12C, the substrate end of the switching tube M12C is connected to the substrate end of the switching tube M14C and the second non-controllable end of the switching tube M14C, and the second non-controllable end of the switching tube M12C is connected to the first non-controllable end of the switching tube M14C.
The invention provides a transformer substation one-key sequential control video double-confirmation system which realizes accurate recognition of target behaviors and objects by adopting modes such as a cascading model, multi-task learning and sliding window smooth statistics through a target position determining module, a target behavior specification confirming module and an object recognition module, and improves the safety of video monitoring of a transformer substation and the like. One of the improvement points of the invention is that the object identification module adopts a 1. Cascade model, and a cascade model is designed for reducing false alarm caused by field station internal equipment, background and the like. The picture is firstly input into a light and fast single-stage object detection model (YOLOV 4-tiny), the model only detects whether a person appears in the picture, once the person appears, the algorithm cuts the area of the person in the picture, and then the cut area is sent to the second-stage model. After receiving the human body cutting area, the second-level model is wholly enlarged to 416x416 pixel size, and then target detection is carried out. The model outputs 8 labels such as whether to wear safety helmets, whether to wear work clothes, whether to wear work shoes, whether to take a phone call in smoking, and the like. The model is designed in such a way that the interference of the background is eliminated by the first-level model, so that the attention of the model is focused on the human body, and the misinformation can be greatly reduced; in addition, the human body area is cut out and then enlarged, so that the details of the human body, including the types of the working shoes, whether the hands hold cigarettes or not, the telephone and the like, can be better seen. 2. In order to detect whether people call in a smoking mode or not in a video, a behavior classification network model is designed, and the model is trained by performing combined multi-task learning with the cascade model in the step 1. Aiming at the fact that whether the cascade model in the step 1 outputs the target frames of smoking and calling, the feature maps of the target frames are cached, the results of the latest two hundred frames are cached totally, then the feature maps of the two hundred frames are respectively obtained through GAP (Global Average pooling) to obtain 1x1 feature points of each frame, at the moment, a feature vector with 1x200 dimensionality is obtained, and the feature vectors are input into a two-layer full-connection network to output the probability of whether the target is calling or not. As another improvement of the invention, the item realizes an automatic detection function based on the deep learning target detection method: firstly, collecting images of a tank car, workers, a non-slip mat and a warning board to form a data set, training by using a deep learning target detection method, and obtaining a deep learning model capable of detecting objects such as the tank car, the workers, the non-slip mat and the warning board; and then designing a behavior specification detection algorithm, starting to detect objects such as workers, anti-skidding pads, warning boards and the like after detecting that the tank car enters the designated area, and finishing the operation if detecting that the target object appears in the designated area. If the target object is not detected within the specified time, the job is considered not to be completed according to the specifications, and an alarm is triggered.
2. Sliding window smooth statistical method for judging target position
In detecting the work behavior profile, it is necessary to determine whether the vehicle stays in/leaves the designated area. Generally, an object is considered to be present if it appears within a video frame; conversely, if there is no object within the video frame, the object is considered to be away. In actual operation, due to reasons such as frame skipping reasoning and unclear pictures caused by object motion, the situation that the program cannot detect the object in the video frame occurs. Therefore, a sliding window smoothing statistical method is applied, the occurrence situation of a target object in a plurality of recent frames is recorded, the target is considered to exist only when the number of frames in which the target is detected in the frames exceeds a certain proportion, and the target is considered to leave when the number of frames in which the target is detected is lower than the certain proportion. The erroneous recognition result of the individual frame is prevented from affecting the program judgment.
The invention has the advantages that the video monitoring system is provided with the rechecking device, the rechecking device can control the information aiming at the alarm on site to carry out recheck, the stability of video monitoring is improved, the position verification module and the information processing device are controlled to carry out position movement and recheck information verification aiming at the rechecking device, the stability of video monitoring is improved, and the position of the movement of the rechecking device can be quickly controlled through the specific design of the information processing device.
Drawings
Fig. 1 is a schematic diagram of a transformer substation one-key sequential control video double-confirmation system.
FIG. 2 is a schematic view of a review device of the present invention.
FIG. 3 is a diagram of an information processing apparatus according to the present invention.
Detailed Description
The present application will now be described in further detail with reference to the drawings, it should be noted that the following detailed description is given for illustrative purposes only and is not to be construed as limiting the scope of the present application, as those skilled in the art will be able to make numerous insubstantial modifications and adaptations to the present application based on the above disclosure.
As shown in fig. 1, the schematic diagram of the substation one-key sequential control video double-confirmation system disclosed by the invention comprises a camera device, a control device and an alarm device; the control device comprises a target position determining module, a target behavior specification confirming module and an object identifying module; the target position determining module is used for determining whether the target stays in the designated area or leaves the designated area; the target behavior specification confirming module is used for receiving the result that the target position confirming module confirms that the target is in the designated area, automatically detecting a target object through deep learning, determining whether the target behavior is in specification or not, and if the target behavior is not in specification, sending a signal to the alarm device to trigger alarm; the object identification module realizes the safe identification of the object based on the cascade model.
In the transformer substation one-key sequential control video double-confirmation system, the target position determination module receives video data acquired by the camera device, judges the target position by adopting a sliding window smooth statistical method, and if a target appears in a video frame, the target is considered to exist; on the contrary, if no target exists in the video frame, the target is considered to leave; the method comprises the steps of recording the occurrence condition of a target object in a plurality of recent frames by a sliding window smooth statistical method, considering that the target exists only when the number of frames in which the target is detected exceeds a certain proportion, and considering that the target leaves when the number of frames in which the target is detected is lower than the certain proportion, so that the influence of the error identification result of individual frames on program judgment can be avoided.
According to the transformer substation one-key sequential control video double-confirmation system, a target behavior specification confirmation module forms a data set by acquiring images of a tank car, workers, an anti-skid pad and a warning board, and a deep learning model for detecting objects of the tank car, the workers, the anti-skid pad and the warning board is obtained by training by using a deep learning target detection method; then, designing a behavior specification detection algorithm, starting to detect the objects of the working personnel, the non-slip mat and the warning board after detecting that the tank car enters the designated area, and finishing the operation if detecting that the target object appears in the designated area; if the target object is not detected within the specified time, the operation is not finished according to the standard, and an alarm signal is sent to the alarm device to trigger the alarm.
According to the transformer substation one-key sequential control video double-confirmation system, false alarm of object recognition caused by field station internal equipment and background is reduced through a cascade model, the cascade model comprises a first-stage model and a second-stage model, the first-stage model is firstly input into a light and rapid single-stage target detection model (YOLOV 4-tiny), the model only detects whether a person appears in a picture, once the person appears in the picture, an algorithm cuts off a region of the person in the picture, and then the cut region is sent to the second-stage model; after receiving the human body cutting area, the second-level model is wholly amplified to 416x416 pixel size, and then target detection is carried out; the isomorphic first-stage model eliminates the interference of the background, so that the attention of the model is focused on the human body, and the misinformation can be greatly reduced; in addition, the human body area is cut out and then is amplified, so that the details of the human body can be better seen.
As shown in fig. 2, which is a schematic diagram of a rechecking device of the present invention, the substation one-key sequential control video double-confirmation system further includes a rechecking device for checking alarm information, and the rechecking device includes a four-drive mobile module, a camera, a controller, a position checking module, and a data transceiver module; the rechecking device confirms whether the operation which is not executed according to the standard still exists within a certain time when the alarm device sends the alarm signal, if so, the rechecking device is started, and the rechecking device starts the rechecking after receiving the starting signal; the four-wheel-drive moving module comprises four drivers, four detectors, four wheels and an information processing device connected with the detectors, wherein the detectors are used for detecting the position information of the movement of the wheels, and the information processing device is used for processing signals detected by the detectors so as to determine the movement deviation between two front wheels or two rear wheels of the four-wheel drive; the driver is used for driving the wheels according to the control command of the controller; the position checking module is used for receiving the image or video information with the alarm, determining the position and shooting angle of an alarm camera for shooting the image or video information with the alarm, controlling the rechecking device to move to a matched position according to the position and shooting angle, controlling the shooting angle to be the same, carrying out first shooting, comparing the video or image shot at the first time with the video or image of the alarm camera, determining whether the position and shooting angle are the same, if the position and shooting angle are different, readjusting the position and shooting angle of the rechecking device, carrying out second shooting, and comparing the video or image shot at the second time with the video or image of the alarm camera until the position and shooting angle are the same; if the position and the shooting angle are the same, determining whether the video or image content in the rechecking device is irregular, if so, carrying out manual intervention to eliminate irregular content, and if not, overhauling an alarm camera or manually interfering to confirm whether the irregular content exists; the data receiving and sending module is used for data transmission.
FIG. 3 is a schematic diagram of an information processing apparatus according to the present invention; the transformer substation one-key sequential control video double-confirmation system comprises an information processing device and a control device, wherein the information processing device comprises: the first non-controllable ends of the switch tubes M4-M6 are connected with a power supply VDD, the controllable end of the switch tube M4 is connected with the controllable end of the switch tube M5, the controllable end of the switch tube M6, the controllable ends of the switch tubes M7-M8, the controllable end of the switch tube M5C, the controllable end of the switch tube M6C, the controllable end of the switch tube M7C, the controllable end of the switch tube M8C and the second non-controllable end of the switch tube M4, the second non-controllable end of the switch tube M4 is connected with the first end of a current source ISET, the second end of the current source ISET is grounded, and the substrate end of the switch tube M4 is connected with the first non-controllable end of the switch tube M4; the second non-controllable end of the switch tube M5 is respectively connected with the first non-controllable end of the switch tube M1 and the controllable ends of the switch tubes M13-M14, the first bias grid end of the switch tube M1 and the first bias grid end of the switch tube M2 are connected with a power supply VDD, the control end of the switch tube M1 is grounded, the second bias grid end of the switch tube M1 is connected with a left front wheel detection signal, the substrate end of the switch tube M1 is connected with the substrate end of the switch tube M2, the substrate end of the switch tube M3 and the second non-controllable end of the switch tube M3, a second non-controllable end of the switching tube M1 is respectively connected with a second non-controllable end of the switching tube M2 and a first non-controllable end of the switching tube M3, the second non-controllable end of the switching tube M3 is grounded, a substrate end of the switching tube M6 is connected with a first non-controllable end of the switching tube M6, a second non-controllable end of the switching tube M6 is respectively connected with a controllable end of the switching tube M3 and a first non-controllable end of the switching tube M2, a controllable end of the switching tube M2 is respectively connected with a first end of a resistor Rt, a second non-controllable end of a switching tube M9 and a first non-controllable end of a switching tube M11, and a second bias of the switching tube M2 is connected with a right front wheel detection signal;
the first non-controllable end of the switch tube M7-M8 is connected with a power supply VDD, the controllable end of the switch tube M7 is connected with the controllable end of the switch tube M8, the substrate end of the switch tube M7 is connected with the first non-controllable end of the switch tube M7 and the substrate end of the switch tube M9, the second non-controllable end of the switch tube M7 is connected with the first non-controllable end of the switch tube M9, the controllable end of the switch tube M9 is grounded, the second non-controllable end of the switch tube M9 is connected with the control end of the switch tube M2 and the first non-controllable end of the switch tube M11, the substrate end of the switch tube M11 is connected with the substrate end of the switch tube M13 and the second non-controllable end of the switch tube M13, the controllable end of the switch tube M11 is connected with the controllable end of the switch tube M12 and the power supply VDD, a second non-controllable end of the switching tube M11 is connected with a first non-controllable end of the switching tube M13, a controllable end of the switching tube M13 is connected with a controllable end of the switching tube M14 and a second non-controllable end of the switching tube M5, the second non-controllable end of the switching tube M13 is grounded, a substrate end of the switching tube M8 is respectively connected with a first non-controllable end of the switching tube M8 and a substrate end of the switching tube M10, the second non-controllable end of the switching tube M8 is connected with a first non-controllable end of the switching tube M10, the second non-controllable end of the switching tube M10 is connected with a first output end and a first non-controllable end of the switching tube M12, the substrate end of the switching tube M12 is connected with the substrate end of the switching tube M14 and the second non-controllable end of the switching tube M14, and the second non-controllable end of the switching tube M12 is connected with the first non-controllable end of the switching tube M14;
the first non-controllable end of the switch tube M5C-M6C is connected with a power supply VDD, the controllable end of the switch tube M5C is connected with the controllable end of the switch tube M6C, the second non-controllable end of the switch tube M5C is respectively connected with the first non-controllable end of the switch tube M1C and the controllable ends of the switch tubes M13C-M14C, the first bias grid end of the switch tube M1C and the first bias grid end of the switch tube M2C are connected with the power supply VDD, the control end of the switch tube M1C is grounded, the second bias grid end of the switch tube M1C is connected with the left rear wheel detection signal, the substrate end of the switch tube M1C is connected with the substrate end of the switch tube M2C, the substrate end of the switch tube M3C and the second non-controllable end of the switch tube M3C, a second non-controllable end of the switching tube M1C is connected with a second non-controllable end of the switching tube M2C and a first non-controllable end of the switching tube M3C respectively, a second non-controllable end of the switching tube M3C is grounded, a substrate end of the switching tube M6C is connected with a first non-controllable end of the switching tube M6C, a second non-controllable end of the switching tube M6C is connected with a controllable end of the switching tube M3C and a first non-controllable end of the switching tube M2C respectively, a controllable end of the switching tube M2C is connected with a second end of the resistor Rt, a second non-controllable end of the switching tube M9C and a first non-controllable end of the switching tube M11C respectively, and a second bias grid end of the switching tube M2C is connected with a right rear wheel detection signal;
the first non-controllable end of the switch tube M7C-M8C is connected with a power supply VDD, the controllable end of the switch tube M7C is connected with the controllable end of the switch tube M8C, the substrate end of the switch tube M7C is connected with the first non-controllable end of the switch tube M7C and the substrate end of the switch tube M9C, the second non-controllable end of the switch tube M7C is connected with the first non-controllable end of the switch tube M9C, the controllable end of the switch tube M9C is grounded, the second non-controllable end of the switch tube M9C is connected with the control end of the switch tube M2C and the first non-controllable end of the switch tube M11C, the substrate end of the switch tube M11C is connected with the substrate end of the switch tube M13C and the second non-controllable end of the switch tube M13C, the controllable end of the switch tube M11C is connected with the controllable end of the switch tube M12C and the power supply VDD, the second non-controllable end of the switching tube M11C is connected to the first non-controllable end of the switching tube M13C, the controllable end of the switching tube M13C is connected to the controllable end of the switching tube M14C and the second non-controllable end of the switching tube M5C, the second non-controllable end of the switching tube M13C is grounded, the substrate end of the switching tube M8C is connected to the first non-controllable end of the switching tube M8C and the substrate end of the switching tube M10C respectively, the second non-controllable end of the switching tube M8C is connected to the first non-controllable end of the switching tube M10C, the second non-controllable end of the switching tube M10C is connected to the second output end and the first non-controllable end of the switching tube M12C, the substrate end of the switching tube M12C is connected to the substrate end of the switching tube M14C and the second non-controllable end of the switching tube M14C, and the second non-controllable end of the switching tube M12C is connected to the first non-controllable end of the switching tube M14C.
The gas transportation process of the transportation and distribution company has the following operation specifications: before the tank car drives into the station, station staff need walk a week around the tank car and check vehicle situation (2) the tank car stops after the loading district, before carrying out loading and unloading operation, station staff need place the slipmat in wheel department, place unexpected removal of vehicle (3) the tank car stops after the unloading district, before carrying out loading and unloading operation, station staff need place the warning sign in vehicle the place ahead, the suggestion irrelevant personnel must not be close to.
The invention realizes the research of automatically judging whether the worker operates according to the standard through the video intelligent analysis of the camera, and sends alarm information when detecting that the worker does not operate according to the standard.
The invention provides a transformer substation one-key sequential control video double-confirmation system, which realizes accurate identification of target behaviors and objects by adopting modes such as a cascade model, multi-task learning and sliding window smooth statistics through a target position determination module, a target behavior specification confirmation module and an object identification module, and improves the safety of video monitoring of a transformer substation and the like. One of the improvement points of the invention is that the object identification module adopts a 1. Cascade model, and a cascade model is designed for reducing false alarm caused by field station internal equipment, background and the like. The picture is firstly input into a light and fast single-stage object detection model (YOLOV 4-tiny), the model only detects whether a person appears in the picture, once the person appears, the algorithm cuts the area of the person in the picture, and then the cut area is sent to the second-stage model. After receiving the human body cutting area, the second-level model is wholly enlarged to 416x416 pixel size, and then target detection is carried out. The model outputs 8 labels such as whether to wear safety helmets, whether to wear work clothes, whether to wear work shoes, whether to take a phone call in smoking, and the like. The design of the model has the advantages that the interference of the background is eliminated by the first-level model, the attention of the model is focused on the human body, and the misinformation can be greatly reduced; in addition, the human body area is cut out and then enlarged, so that the details of the human body, including the types of the working shoes, whether the hands hold cigarettes or not, the telephone and the like, can be better seen. 2. In order to detect whether people call in a smoking mode or not in a video, a behavior classification network model is designed, and the model is trained by performing combined multi-task learning with the cascade model in the step 1. Aiming at the fact that whether the cascade model in the step 1 outputs the target frames of smoking and calling, the feature maps of the target frames are cached, the results of the latest two hundred frames are cached totally, then the feature maps of the two hundred frames are respectively obtained through GAP (Global Average pooling) to obtain 1x1 feature points of each frame, at the moment, a feature vector with 1x200 dimensionality is obtained, and the feature vectors are input into a two-layer full-connection network to output the probability of whether the target is calling or not. As another improvement of the invention, the item realizes an automatic detection function based on the deep learning target detection method: firstly, collecting images of a tank car, workers, a non-slip mat and a warning board to form a data set, training by using a deep learning target detection method, and obtaining a deep learning model capable of detecting objects such as the tank car, the workers, the non-slip mat and the warning board; and then designing a behavior specification detection algorithm, starting to detect objects such as workers, anti-skidding pads, warning boards and the like after detecting that the tank car enters the designated area, and finishing the operation if detecting that the target object appears in the designated area. If the target object is not detected within the specified time, the operation is not finished according to the standard, and an alarm is triggered.
2. Sliding window smooth statistical method for judging target position
In detecting the work behavior profile, it is necessary to determine whether the vehicle stays in/leaves the designated area. Generally, an object is considered to be present if it appears within a video frame; conversely, if there is no object within the video frame, the object is considered to be away. In actual operation, due to reasons such as frame skipping reasoning and unclear pictures caused by object motion, the situation that the program cannot detect the objects in the video frame occurs. Therefore, a sliding window smoothing statistical method is applied, the occurrence situation of a target object in a plurality of recent frames is recorded, the target is considered to exist only when the number of frames in which the target is detected in the frames exceeds a certain proportion, and the target is considered to leave when the number of frames in which the target is detected is lower than the certain proportion. The influence of the error identification result of the individual frame on the program judgment is avoided.
The invention has the further improvement that a rechecking device is arranged, the rechecking device can control the information of the on-site alarm to recheck, the stability of video monitoring is improved, the position verification module and the information processing device are controlled to move and check the rechecking information of the rechecking device, the stability of video monitoring is improved, and the position of the movement of the rechecking device can be quickly controlled through the specific design of the information processing device.
Claims (3)
1. The transformer substation one-key sequential control video double-confirmation system is characterized by comprising a camera device, a control device and an alarm device; the control device comprises a target position determining module, a target behavior specification confirming module and an object identifying module; the target position determining module is used for determining whether the target stays in the designated area or leaves the designated area; the target behavior specification confirming module is used for receiving the result that the target position confirming module confirms that the target is in the designated area, automatically detecting the target object through deep learning, determining whether the target behavior is normal or not, and if the target behavior is not normal, sending a signal to the alarm device to trigger alarm; the object identification module realizes the safe identification of the object based on a cascade model; the system also comprises a rechecking device for checking the alarm information, wherein the rechecking device comprises a four-wheel-drive mobile module, a camera, a controller, a position checking module and a data transceiving module; the rechecking device confirms whether the operation which is not executed according to the standard still exists within a certain time when the alarm device sends the alarm signal, if so, the rechecking device is started, and the rechecking device starts the rechecking after receiving the starting signal; the four-wheel-drive moving module comprises four drivers, four detectors, four wheels and an information processing device connected with the detectors, wherein the detectors are used for detecting the position information of the movement of the wheels, and the information processing device is used for processing signals detected by the detectors so as to determine the movement deviation between two front wheels or two rear wheels of the four-wheel drive; the driver is used for driving the wheels according to the control command of the controller; the position checking module is used for receiving the image or video information with the alarm, determining the position and shooting angle of an alarm camera for shooting the image or video information with the alarm, controlling the rechecking device to move to a matched position according to the position and shooting angle, controlling the shooting angle to be the same, carrying out first shooting, comparing the video or image shot for the first time with the video or image of the alarm camera, determining whether the position and shooting angle are the same, if not, readjusting the position and shooting angle of the rechecking device, carrying out second shooting, and comparing the video or image shot for the second time with the video or image of the alarm camera until the position and shooting angle are the same; if the position and the shooting angle are the same, determining whether the video or image content in the rechecking device is not standard, if so, manually intervening to eliminate the non-standard content, and if not, overhauling the alarm camera or manually intervening to confirm whether the non-standard content exists; the data receiving and transmitting module is used for carrying out data transmission;
the rechecking device can control the information of on-site alarm to be rechecked, so that the stability of video monitoring is improved, the position movement and the rechecking information correction are carried out on the rechecking device by controlling the position checking module and the information processing device, so that the stability of video monitoring is improved, and the position of the movement of the rechecking device can be quickly controlled by the specific design of the information processing device;
the information processing apparatus includes: the first non-controllable ends of the switch tubes M4-M6 are connected with a power supply VDD, the controllable end of the switch tube M4 is connected with the controllable end of the switch tube M5, the controllable end of the switch tube M6, the controllable ends of the switch tubes M7-M8, the controllable end of the switch tube M5C, the controllable end of the switch tube M6C, the controllable end of the switch tube M7C, the controllable end of the switch tube M8C and the second non-controllable end of the switch tube M4, the second non-controllable end of the switch tube M4 is connected with the first end of a current source ISET, the second end of the current source ISET is grounded, and the substrate end of the switch tube M4 is connected with the first non-controllable end of the switch tube M4; the second non-controllable end of the switch tube M5 is respectively connected with the first non-controllable end of the switch tube M1 and the controllable ends of the switch tubes M13-M14, the first bias grid end of the switch tube M1 and the first bias grid end of the switch tube M2 are connected with a power supply VDD, the control end of the switch tube M1 is grounded, the second bias grid end of the switch tube M1 is connected with a left front wheel detection signal, the substrate end of the switch tube M1 is connected with the substrate end of the switch tube M2, the substrate end of the switch tube M3 and the second non-controllable end of the switch tube M3, a second non-controllable end of the switching tube M1 is respectively connected with a second non-controllable end of the switching tube M2 and a first non-controllable end of the switching tube M3, the second non-controllable end of the switching tube M3 is grounded, a substrate end of the switching tube M6 is connected with a first non-controllable end of the switching tube M6, a second non-controllable end of the switching tube M6 is respectively connected with a controllable end of the switching tube M3 and a first non-controllable end of the switching tube M2, a controllable end of the switching tube M2 is respectively connected with a first end of a resistor Rt, a second non-controllable end of a switching tube M9 and a first non-controllable end of a switching tube M11, and a second bias of the switching tube M2 is connected with a right front wheel detection signal;
the first non-controllable end of the switch tube M7-M8 is connected with a power supply VDD, the controllable end of the switch tube M7 is connected with the controllable end of the switch tube M8, the substrate end of the switch tube M7 is connected with the first non-controllable end of the switch tube M7 and the substrate end of the switch tube M9, the second non-controllable end of the switch tube M7 is connected with the first non-controllable end of the switch tube M9, the controllable end of the switch tube M9 is grounded, the second non-controllable end of the switch tube M9 is connected with the control end of the switch tube M2 and the first non-controllable end of the switch tube M11, the substrate end of the switch tube M11 is connected with the substrate end of the switch tube M13 and the second non-controllable end of the switch tube M13, the controllable end of the switch tube M11 is connected with the controllable end of the switch tube M12 and the power supply VDD, the second non-controllable end of the switch tube M11 is connected with the first non-controllable end of the switch tube M13, the controllable end of the switching tube M13 is connected with the controllable end of the switching tube M14 and the second non-controllable end of the switching tube M5, the second non-controllable end of the switching tube M13 is grounded, the substrate end of the switching tube M8 is respectively connected with the first non-controllable end of the switching tube M8 and the substrate end of the switching tube M10, the second non-controllable end of the switching tube M8 is connected with the first non-controllable end of the switching tube M10, the second non-controllable end of the switching tube M10 is connected with the first output end and the first non-controllable end of the switching tube M12, the substrate end of the switching tube M12 is connected with the substrate end of the switching tube M14 and the second non-controllable end of the switching tube M14, the second non-controllable end of the switching tube M14 is grounded, and the second non-controllable end of the switching tube M12 is connected with the first non-controllable end of the switching tube M14;
the first non-controllable end of the switch tube M5C-M6C is connected with a power supply VDD, the controllable end of the switch tube M5C is connected with the controllable end of the switch tube M6C, the second non-controllable end of the switch tube M5C is respectively connected with the first non-controllable end of the switch tube M1C and the controllable ends of the switch tubes M13C-M14C, the first bias grid end of the switch tube M1C and the first bias grid end of the switch tube M2C are connected with the power supply VDD, the control end of the switch tube M1C is grounded, the second bias grid end of the switch tube M1C is connected with the left rear wheel detection signal, the substrate end of the switch tube M1C is connected with the substrate end of the switch tube M2C, the substrate end of the switch tube M3C and the second non-controllable end of the switch tube M3C, a second non-controllable end of the switching tube M1C is connected with a second non-controllable end of the switching tube M2C and a first non-controllable end of the switching tube M3C respectively, a second non-controllable end of the switching tube M3C is grounded, a substrate end of the switching tube M6C is connected with a first non-controllable end of the switching tube M6C, a second non-controllable end of the switching tube M6C is connected with a controllable end of the switching tube M3C and a first non-controllable end of the switching tube M2C respectively, a controllable end of the switching tube M2C is connected with a second end of the resistor Rt, a second non-controllable end of the switching tube M9C and a first non-controllable end of the switching tube M11C respectively, and a second bias grid end of the switching tube M2C is connected with a right rear wheel detection signal;
the first non-controllable end of the switch tube M7C-M8C is connected with a power supply VDD, the controllable end of the switch tube M7C is connected with the controllable end of the switch tube M8C, the substrate end of the switch tube M7C is connected with the first non-controllable end of the switch tube M7C and the substrate end of the switch tube M9C, the second non-controllable end of the switch tube M7C is connected with the first non-controllable end of the switch tube M9C, the controllable end of the switch tube M9C is grounded, the second non-controllable end of the switch tube M9C is connected with the control end of the switch tube M2C and the first non-controllable end of the switch tube M11C, the substrate end of the switch tube M11C is connected with the substrate end of the switch tube M13C and the second non-controllable end of the switch tube M13C, the controllable end of the switch tube M11C is connected with the controllable end of the switch tube M12C and the power supply VDD, the second non-controllable end of the switch tube M11C is connected with the first non-controllable end of the switch tube M13C, the controllable end of the switching tube M13C is connected with the controllable end of the switching tube M14C and the second non-controllable end of the switching tube M5C, the second non-controllable end of the switching tube M13C is grounded, the substrate end of the switching tube M8C is respectively connected with the first non-controllable end of the switching tube M8C and the substrate end of the switching tube M10C, the second non-controllable end of the switching tube M8C is connected with the first non-controllable end of the switching tube M10C, the second non-controllable end of the switching tube M10C is connected with the second output end and the first non-controllable end of the switching tube M12C, the substrate end of the switching tube M12C is connected with the substrate end of the switching tube M14C and the second non-controllable end of the switching tube M14C, the second non-controllable end of the switching tube M14C is grounded, and the second non-controllable end of the switching tube M12C is connected with the first non-controllable end of the switching tube M14C; the target position determining module receives video data acquired by the camera device, judges the position of a target by adopting a sliding window smooth statistical method, and if the target appears in a video frame, the target is considered to exist; on the contrary, if no target exists in the video frame, the target is considered to leave; the method comprises the steps of recording the occurrence condition of a target object in a plurality of recent frames by a sliding window smooth statistical method, considering that the target exists only when the number of frames in which the target is detected exceeds a certain proportion, and considering that the target leaves when the number of frames in which the target is detected is lower than the certain proportion, so that the influence of the error identification result of individual frames on program judgment can be avoided.
2. The substation one-key sequential control video double-confirmation system according to claim 1, wherein the target behavior specification confirmation module forms a data set by collecting images of a tank car, workers, a non-slip mat and a warning board, and obtains a deep learning model for detecting objects of the tank car, the workers, the non-slip mat and the warning board by training with a deep learning target detection method; then, designing a behavior specification detection algorithm, starting to detect the objects of the working personnel, the non-slip mat and the warning board after detecting that the tank car enters the designated area, and finishing the operation if detecting that the target object appears in the designated area; if the target object is not detected within the specified time, the operation is not finished according to the standard, and an alarm signal is sent to the alarm device to trigger the alarm.
3. The substation one-key sequential control video double-confirmation system as claimed in claim 1, wherein the object recognition module reduces false positive of object recognition caused by internal equipment and background of a substation through a cascade model, the cascade model comprises a first-stage model and a second-stage model, the first-stage model is firstly input into a light-weight and rapid single-stage object detection model (YOLOV 4-tiny), the model only detects whether a person appears in a picture, once the person appears, an algorithm cuts off a region of the person in the picture, and then sends the cut region to the second-stage model; after receiving the human body cutting area, the second-level model is wholly enlarged to 416x416 pixel size, and then target detection is carried out; the isomorphic first-stage model eliminates the interference of the background, so that the attention of the model is focused on the human body, and the misinformation can be greatly reduced; in addition, the human body area is cut out and then is amplified, so that the details of the human body can be better seen.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110616608.7A CN113381510B (en) | 2021-06-02 | 2021-06-02 | One-key sequential control video double-confirmation system for transformer substation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110616608.7A CN113381510B (en) | 2021-06-02 | 2021-06-02 | One-key sequential control video double-confirmation system for transformer substation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113381510A CN113381510A (en) | 2021-09-10 |
| CN113381510B true CN113381510B (en) | 2023-01-24 |
Family
ID=77575524
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110616608.7A Active CN113381510B (en) | 2021-06-02 | 2021-06-02 | One-key sequential control video double-confirmation system for transformer substation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113381510B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101917061A (en) * | 2010-07-14 | 2010-12-15 | 山东电力集团公司泰安供电公司 | Automatic inspection method and device for substation |
| CN109614906A (en) * | 2018-12-03 | 2019-04-12 | 北京工业大学 | A kind of security system and security alarm method based on deep learning |
| CN111507308A (en) * | 2020-05-07 | 2020-08-07 | 广东电网有限责任公司 | Transformer substation safety monitoring system and method based on video identification technology |
| CN111614913A (en) * | 2020-05-06 | 2020-09-01 | 许昌许继软件技术有限公司 | One-key sequence control double-confirmation system and method based on intelligent isolation switch position identification |
| CN111624917A (en) * | 2020-06-05 | 2020-09-04 | 内蒙古中孚明丰农业科技有限公司 | Vehicle-mounted transportation data analysis control system |
| CN111753780A (en) * | 2020-06-29 | 2020-10-09 | 广东电网有限责任公司清远供电局 | Transformer substation violation detection system and violation detection method |
| CN112183385A (en) * | 2020-09-29 | 2021-01-05 | 支付宝(杭州)信息技术有限公司 | Target object processing method and device |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8036425B2 (en) * | 2008-06-26 | 2011-10-11 | Billy Hou | Neural network-controlled automatic tracking and recognizing system and method |
| CN107257161B (en) * | 2017-06-20 | 2020-10-09 | 安徽南瑞继远电网技术有限公司 | Transformer substation disconnecting link remote control auxiliary checking method and system based on state recognition algorithm |
| CA3028630C (en) * | 2018-03-01 | 2023-10-17 | Beijing Didi Infinity Technology And Development Co., Ltd. | Systems and methods for identifying risky driving behavior |
| CN108280953A (en) * | 2018-03-27 | 2018-07-13 | 上海小蚁科技有限公司 | Video detecting alarm method and device, storage medium, camera |
-
2021
- 2021-06-02 CN CN202110616608.7A patent/CN113381510B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101917061A (en) * | 2010-07-14 | 2010-12-15 | 山东电力集团公司泰安供电公司 | Automatic inspection method and device for substation |
| CN109614906A (en) * | 2018-12-03 | 2019-04-12 | 北京工业大学 | A kind of security system and security alarm method based on deep learning |
| CN111614913A (en) * | 2020-05-06 | 2020-09-01 | 许昌许继软件技术有限公司 | One-key sequence control double-confirmation system and method based on intelligent isolation switch position identification |
| CN111507308A (en) * | 2020-05-07 | 2020-08-07 | 广东电网有限责任公司 | Transformer substation safety monitoring system and method based on video identification technology |
| CN111624917A (en) * | 2020-06-05 | 2020-09-04 | 内蒙古中孚明丰农业科技有限公司 | Vehicle-mounted transportation data analysis control system |
| CN111753780A (en) * | 2020-06-29 | 2020-10-09 | 广东电网有限责任公司清远供电局 | Transformer substation violation detection system and violation detection method |
| CN112183385A (en) * | 2020-09-29 | 2021-01-05 | 支付宝(杭州)信息技术有限公司 | Target object processing method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113381510A (en) | 2021-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112686090B (en) | Intelligent monitoring system for abnormal behavior in bus | |
| CN108297058A (en) | Intelligent security guard robot and its automatic detecting method | |
| CN112633057B (en) | Intelligent monitoring method for abnormal behavior in bus | |
| CN108284427A (en) | Security robot and its automatic detecting method | |
| CN110867046A (en) | Intelligent car washer video monitoring and early warning system based on cloud computing | |
| CN110166546A (en) | A kind of novel intelligent supervision control method and system for operational motor vehicles | |
| CN104658188A (en) | Channel security early-warning system and channel early-warning method | |
| CN108583592A (en) | A kind of subway service on buses or trains job information acquisition intelligent detecting method | |
| CN111126193A (en) | Artificial intelligence recognition system based on deep learning coal mine underground unsafe behavior | |
| CN107038830A (en) | A kind of gun dynamic management system and method | |
| KR20080044812A (en) | The automatic guard system to prevent the crime and accident using computer video image analysis technology | |
| CN113178047A (en) | Detection method of gas fire-extinguishing control system of building fire-fighting equipment | |
| CN106295470A (en) | A kind of bank self-help service area early-warning monitoring method, Apparatus and system | |
| CN113990018A (en) | Safety risk prediction system | |
| CN112885013A (en) | Monitoring and early warning method and device and readable storage medium | |
| CN113381510B (en) | One-key sequential control video double-confirmation system for transformer substation | |
| CN114565886A (en) | Method for detecting dangerous behavior of personnel crossing station tracks in railway industry | |
| JP3361399B2 (en) | Obstacle detection method and device | |
| CN112343657B (en) | Method for identifying mine underground air door and monitoring abnormal state of mine underground air door | |
| CN113055651A (en) | Artificial intelligence type vehicle security system and computer readable storage medium | |
| CN107272530A (en) | A kind of construction site long-distance monitoring method and system based on Internet of Things | |
| CN113807224B (en) | Method for detecting and tracking illegal behaviors of factory | |
| CN113609925A (en) | LNG loading and unloading operation safety management and control system and illegal behavior identification method | |
| CN112489346A (en) | Intelligent infrared monitoring and alarming equipment for office area | |
| CN111667663A (en) | Waste transport driver violation detection method based on artificial intelligence video analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |