CN113283496A - Method applied to intelligent identification of dangerous actions of distribution network uninterrupted operation - Google Patents
Method applied to intelligent identification of dangerous actions of distribution network uninterrupted operation Download PDFInfo
- Publication number
- CN113283496A CN113283496A CN202110560448.9A CN202110560448A CN113283496A CN 113283496 A CN113283496 A CN 113283496A CN 202110560448 A CN202110560448 A CN 202110560448A CN 113283496 A CN113283496 A CN 113283496A
- Authority
- CN
- China
- Prior art keywords
- limb
- charged body
- image
- initial
- monitoring image
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
Abstract
The invention relates to the technical field of artificial intelligence and discloses a method for intelligently identifying dangerous actions of distribution network uninterrupted operation. The invention provides a security scheme capable of automatically identifying dangerous actions of operators in the distribution network uninterrupted operation process, which comprises the steps of firstly determining initial coordinates of a charged body and limbs of the operators in a camera coordinate system based on a field operation monitoring image acquired by a binocular camera in real time, then determining current coordinates of the charged body and the limbs in the camera coordinate system based on the initial coordinates and attitude data acquired by at least two attitude sensors in real time, and finally identifying the current dangerous actions when the distances between the limbs and the charged body are too close based on the geometric relationship of coordinates, and sending alarm prompt information so that the operators can correct errors in time, further personnel electric shock safety accidents are avoided, the safety of the operators and equipment is guaranteed, and the security scheme is convenient to be practically applied and popularized.
Description
Technical Field
The invention belongs to the technical field of artificial intelligence, and particularly relates to a method for intelligently identifying dangerous actions of distribution network uninterrupted operation.
Background
Distribution network uninterrupted operation is a service for guaranteeing normal work of a power grid in the power industry (namely, power failure is eliminated under the condition of uninterrupted power supply, and normal power utilization in production and life is guaranteed). The business process of the distribution network uninterrupted operation specifically comprises the steps of personnel reporting requirements, site actual survey, maintenance plan arrangement, operation scheduling, site operation, operation evaluation and the like in sequence.
When distribution lines live working, because the voltage of electric wire netting is low, electromagnetic strength is low, and the spatial distance between the wire is little, and distribution equipment is also more simultaneously, and the staff touches electric power facility easily in home range, is unfavorable for the staff operation. In order to ensure the safety of operators and equipment, at present, a main insulating tool and an auxiliary insulating tool are mainly used to form multiple safety protection, and an insulating rod operation method, an insulating glove operation method and the like are adopted to carry out field operation.
However, in the actual field operation process, due to factors such as different quality of personnel, irregular operation steps and/or thin self-protection consciousness, the operating personnel may unconsciously make dangerous motions, so that the distance between the body and the charged body is too short, and further, a safety accident of electric shock is caused to the personnel.
Disclosure of Invention
The invention aims to provide a method, a device, computer equipment and a storage medium for intelligently identifying dangerous actions of distribution network uninterrupted operation, which can automatically sense the current coordinates of a limb and an electrified body based on a field operation monitoring image and attitude data collected in real time, identify the current dangerous actions when the distance between the limb and the electrified body is judged to be too close based on the geometric relation of the coordinates, and send alarm prompt information so that an operator can correct errors in time, further personnel electric shock safety accidents are avoided, the safety of the operator and the equipment is guaranteed, and the method is convenient for practical application and popularization.
In a first aspect, the invention provides a method for intelligently identifying dangerous actions in distribution network uninterrupted operation, which comprises the following steps:
acquiring a field operation monitoring image acquired by a binocular camera in real time, wherein the binocular camera is installed on a safety helmet of an operator;
identifying whether the live working monitoring image contains a charged body image and a limb image of the operator;
when the live working monitoring image is identified to contain the charged body image, determining an initial coordinate of a charged body in a camera coordinate system of the binocular camera, and recording a collection time stamp of the live working monitoring image as an initial time of the charged body, wherein the charged body corresponds to the charged body image one to one;
when the limb image is identified to be contained in the on-site operation monitoring image, determining an initial coordinate of a limb in the camera coordinate system, and recording an acquisition timestamp of the on-site operation monitoring image as an initial moment of the limb, wherein the limb corresponds to the limb image one by one;
determining the current coordinate of the charged body in the camera coordinate system according to the initial coordinate of the charged body and first posture data which is recorded from the initial moment of the charged body and acquired by a first posture sensor in real time, wherein the first posture sensor is bound with the binocular camera;
determining the current coordinate of the limb in the camera coordinate system according to the initial coordinate of the limb, first posture data which is recorded from the initial moment of the limb and acquired by the first posture sensor in real time, and second posture data which is recorded from the initial moment of the limb and acquired by the second posture sensor in real time, wherein the second posture sensor is bound with the limb;
calculating to obtain the current distance from the limb to the charged body according to the current coordinate of the charged body and the current coordinate of the limb;
judging whether the current distance is smaller than or equal to a preset safety distance threshold value or not;
if so, judging that dangerous actions exist at present, and sending first alarm prompt information.
Based on the content of the invention, a security scheme capable of automatically identifying dangerous actions of operators in the distribution network uninterrupted operation process can be provided, namely, the initial coordinates of a charged body and limbs of the operators in a camera coordinate system are determined based on a field operation monitoring image acquired by a binocular camera in real time, then the current coordinates of the charged body and the limbs in the camera coordinate system are determined based on the initial coordinates and posture data acquired by at least two posture sensors in real time, and finally when the distance between the limbs and the charged body is judged to be too close based on the coordinate geometric relationship, the current dangerous actions are identified, and alarm prompt information is sent out, so that the operators can correct errors in time, further personnel electric shock safety accidents are avoided, the safety of the operators and equipment is guaranteed, and the security scheme is convenient for practical application and popularization.
In one possible design, after acquiring the live-work monitoring image acquired by the binocular camera in real time, the method further includes:
extracting limb action characteristics of the operator from the on-site operation monitoring image;
judging whether the limb action characteristics are matched with prestored wrong limb action characteristics;
if so, judging that dangerous actions exist at present, and sending out second alarm prompt information.
In one possible design, identifying whether the live-work monitoring image includes an image of a live body and an image of an extremity of the worker includes:
importing the on-site operation monitoring image into a pre-trained first detection model, and then judging whether the on-site operation monitoring image contains a charged body image according to a detection result of the first detection model, wherein the first detection model adopts a Faster R-CNN detection model, an SSD detection model or a YOLO detection model;
and importing the on-site operation monitoring image into a pre-trained second detection model, and then judging whether the on-site operation monitoring image contains a limb image according to a detection result of the second detection model, wherein the second detection model adopts a Faster R-CNN detection model, an SSD detection model or a YOLO detection model.
In one possible design, determining initial coordinates of a charged body/appendage in a camera coordinate system of the binocular camera includes:
determining the distance from the charged body/the limb to the origin of the camera coordinate system according to the binocular distance measuring principle of the binocular camera;
and acquiring initial coordinates of the charged body/the limb in the camera coordinate system according to the distance and the plane coordinates of the charged body image/the limb image in the on-site operation monitoring image.
In one possible design, after determining the initial/current coordinates of the charged body in the camera coordinate system, the method further comprises:
judging whether the distance from the charged body to the binocular camera is smaller than or equal to the safety distance threshold or not according to the initial coordinate/current coordinate of the charged body;
if so, judging that dangerous actions exist at present, and sending out third alarm prompt information.
In one possible design, after determining initial coordinates of a charged body/appendage in a camera coordinate system of the binocular camera and recording a time stamp of acquisition of the live action surveillance image as an initial time of the charged body/appendage, the method further comprises:
acquiring a new on-site operation monitoring image acquired by the binocular camera in real time;
identifying whether the live working monitoring image contains an image corresponding to the charged body/the limb;
if so, updating the initial coordinates of the charged body/the limb according to the newly determined coordinates of the charged body/the limb in the camera coordinate system, and updating the acquisition timestamp of the new field operation monitoring image to the initial time of the charged body/the limb.
In one possible design, updating the initial coordinates of the charged body according to the newly determined coordinates of the charged body in the camera coordinate system includes:
determining new coordinates of the charged body in the camera coordinate system and at the time of the acquisition timestamp of the new live working monitoring image according to the initial coordinates of the charged body and first attitude data which is recorded from the initial moment of the charged body to the acquisition timestamp of the new live working monitoring image and acquired by the first attitude sensor in real time;
calculating an initial coordinate updating result (x, y, z) of the charged body according to the following formula:
in the formula (x)1,y1,z1) Representing the newly determined coordinates of the charged body in the camera coordinate system, (x)2,y2,z2) And the system also comprises a new coordinate of the charged body in the camera coordinate system and at the time of the acquisition time stamp of the new on-site operation monitoring image, wherein alpha represents a preset weight coefficient corresponding to the newly determined coordinate, beta represents a preset weight coefficient corresponding to the new coordinate, and alpha + beta is 1.
The invention provides a device for intelligently identifying dangerous actions in distribution network uninterrupted operation, which comprises an image acquisition module, an image identification module, an initial data determination module, a current coordinate determination module, a current distance calculation module, a comparison and judgment module and an alarm prompt module, wherein the image acquisition module, the image identification module, the initial data determination module, the current coordinate determination module, the current distance calculation module, the comparison and judgment module and the alarm prompt module are sequentially in communication connection;
the image acquisition module is used for acquiring a field operation monitoring image acquired by a binocular camera in real time, wherein the binocular camera is installed on a safety helmet of an operator;
the image identification module is used for identifying whether the on-site operation monitoring image contains a charged body image and a limb image of the operator;
the initial data determining module is configured to determine an initial coordinate of a charged body in a camera coordinate system of the binocular camera when the live working monitoring image is identified to include the image of the charged body, record a collection timestamp of the live working monitoring image as an initial time of the charged body, determine an initial coordinate of a limb in the camera coordinate system when the live working monitoring image is identified to include the image of the limb, and record a collection timestamp of the live working monitoring image as an initial time of the limb, where the charged body corresponds to the image of the charged body one to one, and the limb corresponds to the image of the limb one to one;
the current coordinate determination module is used for determining the current coordinate of the electrified body in the camera coordinate system according to the initial coordinate of the electrified body and first posture data which is recorded from the initial moment of the electrified body and acquired in real time by a first posture sensor, and determining the current coordinate of the limb in the camera coordinate system according to the initial coordinate of the limb, first posture data which is recorded from the initial moment of the limb and acquired in real time by the first posture sensor and second posture data which is recorded from the initial moment of the limb and acquired in real time by a second posture sensor, wherein the first posture sensor is bound with the binocular camera, and the second posture sensor is bound with the limb;
the current distance calculation module is used for calculating the current distance from the limb to the charged body according to the current coordinate of the charged body and the current coordinate of the limb;
the comparison and judgment module is used for judging whether the current distance is smaller than or equal to a preset safety distance threshold value;
and the alarm prompt module is used for judging that dangerous actions exist at present and sending first alarm prompt information when judging that the current distance is smaller than or equal to the safety distance threshold value.
In a third aspect, the present invention provides a computer device, comprising a memory and a processor, wherein the memory is used for storing a computer program, and the processor is used for reading the computer program and executing the method according to the first aspect or any one of the possible designs of the first aspect.
In a fourth aspect, the present invention provides a storage medium having stored thereon instructions for carrying out the method according to the first aspect or any one of the possible designs of the first aspect, when the instructions are run on a computer.
In a fifth aspect, the present invention provides a computer program product comprising instructions which, when run on a computer, cause the computer to perform the method as described in the first aspect or any one of the possible designs of the first aspect.
The invention has the technical effects that:
(1) the invention provides a security scheme capable of automatically identifying dangerous actions of operators in the distribution network uninterrupted operation process, namely, firstly, based on a field operation monitoring image acquired by a binocular camera in real time, determining initial coordinates of a charged body and limbs of the operators in a camera coordinate system, then based on the initial coordinates and attitude data acquired by at least two attitude sensors in real time, determining current coordinates of the charged body and the limbs in the camera coordinate system, and finally, when the distance between the limbs and the charged body is judged to be too close based on the geometric relationship of coordinates, identifying the current dangerous actions, and sending alarm prompt information so that the operators correct errors in time, further avoiding the electric shock safety accidents of the operators, ensuring the safety of the operators and equipment, and facilitating the practical application and popularization;
(2) the method can also identify whether the operator makes dangerous actions from the action characteristic matching angle in the uninterrupted operation process of the distribution network, and send alarm prompt information when the operator makes dangerous actions, so as to further ensure the safety of the operator and equipment;
(3) whether the distance between the charged body and the head of the operator is too close can be judged in real time, and when the distance is found to be too close, alarm prompt information is also triggered, so that the safety of the operator and equipment is further guaranteed;
(4) the initial coordinate and the initial moment of the charged body/limb can be dynamically corrected based on the image recognition result of the new field operation monitoring image, the subsequent determination error corresponding to the current coordinate is reduced, the correctness of the comparison result of the subsequent distance and the threshold value is ensured, and the safety of the operating personnel and equipment is further ensured.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flow chart of a method for intelligently identifying dangerous actions in distribution network uninterrupted operation, which is provided by the invention.
Fig. 2 is a schematic structural diagram of the device for intelligently identifying dangerous actions in distribution network uninterrupted operation, provided by the invention.
Fig. 3 is a schematic structural diagram of a computer device provided by the present invention.
Detailed Description
The invention is further described with reference to the following figures and specific embodiments. It should be noted that the description of the embodiments is provided to help understanding of the present invention, but the present invention is not limited thereto. Specific structural and functional details disclosed herein are merely representative of exemplary embodiments of the invention. This invention may, however, be embodied in many alternate forms and should not be construed as limited to the embodiments set forth herein.
It should be understood that, for the term "and/or" as may appear herein, it is merely an associative relationship that describes an associated object, meaning that three relationships may exist, e.g., a and/or B may mean: a exists alone, B exists alone, and A and B exist at the same time; for the term "/and" as may appear herein, which describes another associative object relationship, it means that two relationships may exist, e.g., a/and B, may mean: a exists independently, and A and B exist independently; in addition, for the character "/" that may appear herein, it generally means that the former and latter associated objects are in an "or" relationship.
As shown in fig. 1, the method for intelligently identifying dangerous actions in uninterruptible power distribution network according to the first aspect of this embodiment may be implemented, but not limited to, by a computer device capable of acquiring, in real time, in an uninterruptible power distribution network work site, a work surveillance image from a binocular camera (installed on a helmet of a worker) and posture data from a posture sensor (the number of which is at least two, one of which is bound to the binocular camera and the other of which is bound to a limb of the worker), so as to automatically sense current coordinates of the limb and a charged body based on the work surveillance image and posture data acquired in real time, identify that dangerous actions exist currently and send out alarm prompt information when it is determined that the limb and the charged body are too close to each other based on a geometric relationship of the coordinates. The computer device is preferably mounted on the safety helmet of the operator together with the binocular camera so as to give an alarm prompt in time when a dangerous action is found. The method for intelligently identifying dangerous actions in distribution network uninterrupted operation can be but is not limited to the following steps S1-S9.
S1, acquiring a field operation monitoring image acquired by a binocular camera in real time, wherein the binocular camera is installed on a safety helmet of an operator.
In the step S1, the binocular camera is an existing camera device in which the binocular lenses are symmetrically arranged right and left/up and down, and can acquire a monitoring image of a site operation located around an operator in real time by being mounted on the helmet of the operator. The binocular camera is preferably fixed to a position right in front of the helmet so as to collect a site work monitoring image in front of an operator. Since the computer device is preferably mounted on a helmet together with the binocular camera, the computer device can acquire the on-site work monitoring image through wired communication.
And S2, identifying whether the field operation monitoring image contains a charged body image and a limb image of the operator.
In step S2, the specific identification method may be implemented by using an existing image identification technology. The charged body image can be but is not limited to a wired section image, a transformer image, a switch image or a capacitor image and the like; the limb image may be, but is not limited to, a left hand image, a right hand image, a left foot image, a right foot image, or the like. Preferably, the identification of whether the live working monitoring image includes an image of a charged body and an image of an extremity of the operator includes, but is not limited to: importing the on-site operation monitoring image into a pre-trained first detection model, and then judging whether the on-site operation monitoring image contains a charged body image according to a detection result of the first detection model, wherein the first detection model can be but is not limited to a Faster R-CNN detection model, an SSD detection model or a YOLO detection model and the like; and importing the on-site operation monitoring image into a pre-trained second detection model, and then judging whether the on-site operation monitoring image contains a limb image according to a detection result of the second detection model, wherein the second detection model can be but is not limited to a Faster R-CNN detection model, an SSD detection model or a YOLO detection model. The fast R-CNN (fast Region conditional Neural Networks, a target detection algorithm proposed by hoeming et al in 2015), the SSD (Single Shot multi box Detector, a target detection algorithm proposed by Wei Liu et al in 2016) and the YOLO (young Only Look one, a target detection algorithm proposed by Redmon et al in 2016) are commonly used image target detection models, respectively, so that the first detection model and the second detection model, which have been pre-trained, can be obtained based on a conventional training mode, so that the first detection model has the capability of accurately identifying a charged body image, and further can identify whether the charged body image is included in the operation site monitoring image, and the second detection model has the capability of accurately identifying a charged body image, and then whether the limb image is contained in the operation site monitoring image or not can be identified. In addition, in order to improve the recognition accuracy of different limb images, different patterns or color marks and the like can be attached to the outer surfaces of the insulating sleeves worn on different limbs, for example, different colors can be attached to the outer surfaces of the insulating left glove, the insulating right glove, the insulating left foot boot and the insulating right foot boot.
And S3, when the live working monitoring image is identified to contain the charged body image, determining the initial coordinate of the charged body in the camera coordinate system of the binocular camera, and recording the acquisition time stamp of the live working monitoring image as the initial time of the charged body, wherein the charged body corresponds to the charged body image one by one.
In step S3, the number of the charged bodies may be plural for different charged body images, for example, when the on-site operation monitoring image simultaneously includes a line segment image, a transformer image, a switch image, a capacitor image, or the like, it is necessary to determine the corresponding initial coordinates and initial time for each charged body. Preferably, the determining the initial coordinates of the charged object in the camera coordinate system of the binocular camera includes, but is not limited to, the following steps S31 to S32.
And S31, determining the distance from the charged body to the origin of the camera coordinate system according to the binocular range finding principle of the binocular lens camera.
In the step S31, the binocular range finding principle is that the difference between the horizontal/vertical coordinates of the target point imaged on the left/right/up/down two views (which are respectively acquired by the two lenses symmetrically arranged left/right/up/down one to one) is directly proportional to the distance between the target point and the imaging plane, so that the distance between the charged body and the origin of the camera coordinate system, that is, the Z-axis coordinate of the charged body in the camera coordinate system, can be calculated based on the existing binocular range finding technology.
And S32, acquiring initial coordinates of the charged body in the camera coordinate system according to the distance and the plane coordinates of the charged body image in the field operation monitoring image.
In step S32, since the live working monitor image is perpendicular to the optical axis of the binocular camera (i.e., Z axis in the camera coordinate system), the X axis coordinate and the Y axis coordinate of the charged body in the camera coordinate system can be directly obtained based on the coordinate position of the charged body image in the live working monitor image, and the coordinate position of the charged body in the camera coordinate system can be further obtained.
And S4, when the limb image is identified to be contained in the field operation monitoring image, determining the initial coordinate of the limb in the camera coordinate system, and recording the acquisition time stamp of the field operation monitoring image as the initial time of the limb, wherein the limb corresponds to the limb image one by one.
In step S4, the number of limbs may be a plurality of different limb images, for example, when the on-site operation monitoring image includes both a left-hand image and a right-hand image, the corresponding initial coordinates and initial time must be determined for each limb. Preferably, the initial coordinates of the limb in the camera coordinate system are determined similarly to the charged body, including but not limited to the following steps S41 to S42.
S41, determining the distance from the limb to the origin of the camera coordinate system according to a binocular range finding principle of the binocular camera.
And S42, acquiring initial coordinates of the limb in the camera coordinate system according to the distance and the plane coordinates of the limb image in the field operation monitoring image.
And S5, determining the current coordinate of the charged body in the camera coordinate system according to the initial coordinate of the charged body and first posture data which is recorded from the initial moment of the charged body and acquired by a first posture sensor in real time, wherein the first posture sensor is bound with the binocular camera.
In step S5, a motion trajectory of the first attitude sensor from the initial time of the charged body may be derived based on the first attitude data recorded from the initial time of the charged body and collected by the first attitude sensor in real time, specifically based on the existing conventional attitude data processing technology. Meanwhile, the first posture sensor and the binocular camera are bound together, so that the position relation between the first posture sensor and the binocular camera is fixed and known, and the motion track of the binocular camera from the initial moment of the charged body can be further deduced. Considering that the charged body is generally in a static state from the initial time, the motion track of the charged body in the camera coordinate system from the initial time of the charged body can be derived through conventional geometric coordinate system transformation by combining the initial coordinates of the charged body, and the current coordinates of the charged body in the camera coordinate system can be determined. In addition, if there are a plurality of charged bodies, the corresponding current coordinates in the camera coordinate system can be determined for each charged body.
S6, determining the current coordinate of the limb in the camera coordinate system according to the initial coordinate of the limb, first posture data which are recorded from the initial moment of the limb and acquired in real time by the first posture sensor, and second posture data which are recorded from the initial moment of the limb and acquired in real time by the second posture sensor, wherein the second posture sensor is bound with the limb.
In step S6, the motion trajectory of the first posture sensor from the initial time of the limb is derived based on the first posture data recorded from the initial time of the limb and collected by the first posture sensor in real time, and the motion trajectory of the second posture sensor from the initial time of the limb is derived based on the second posture data recorded from the initial time of the limb and collected by the second posture sensor in real time. Meanwhile, as the first posture sensor and the binocular camera are bound together, the position relationship between the first posture sensor and the binocular camera is fixed and known, so that the movement locus of the binocular camera from the initial moment of the limbs can be further deduced. Meanwhile, as the second posture sensor is bound with the limb, the position relation between the second posture sensor and the limb is fixed and known, so that the motion track of the limb from the initial moment of the limb can be further deduced. Finally, combining the initial coordinates of the limb, the motion track of the limb in the camera coordinate system from the initial moment of the limb can be derived through conventional geometric coordinate system transformation, and the current coordinates of the limb in the camera coordinate system are determined. Furthermore, if there are multiple limbs, a corresponding current coordinate in the camera coordinate system may be determined for each limb.
And S7, calculating to obtain the current distance from the limb to the charged body according to the current coordinate of the charged body and the current coordinate of the limb.
In step S7, since the current coordinates of the charged body and the current coordinates of the limb are both coordinates in the camera coordinate system, the current distance may be calculated based on conventional geometric knowledge. In addition, if there are a plurality of charged bodies and/or a plurality of limbs, the corresponding current distance may be calculated for each pair of charged body and limb.
And S8, judging whether the current distance is smaller than or equal to a preset safety distance threshold value.
In step S8, if the current distances between the pairs of charged bodies and the limbs are calculated, it is necessary to separately determine each pair of charged bodies and limbs. In addition, the safe distance threshold value may be uniformly preset for each pair of charged bodies and limbs, or may be differently preset for different charged bodies, for example, different safe distance threshold values may be preset for a wire segment, a transformer, a switch, a capacitor, or the like, so as to refine and match specific recognition conditions.
And S9, if yes, judging that dangerous actions exist at present, and sending first alarm prompt information.
In step S9, the first alarm prompt message is preferably a voice prompt message, such as "caution! Left hand and wire section with electric shock risk! Please avoid "in order to remind the operator to correct the error in time, avoid further electric shock safety accident of the operator, and ensure the safety of the operator and the equipment.
Therefore, by the method detailed in the steps S1-S9, a security scheme capable of automatically identifying dangerous actions of operators in the uninterrupted operation process of the distribution network can be provided, firstly, based on the field operation monitoring image collected by the binocular camera in real time, the initial coordinates of the charged body and the limbs of the operator in the camera coordinate system are determined, then determining the current coordinates of the charged body and the limb in a camera coordinate system based on the initial coordinates and the posture data acquired by at least two posture sensors in real time, finally identifying the current dangerous action when the distance between the limb and the charged body is judged to be too close based on the coordinate geometric relationship, sending out alarm prompt information, so that the operators can correct the errors in time, further electric shock safety accidents of the operators are avoided, the safety of the operators and the equipment is guaranteed, and the practical application and the popularization are facilitated.
On the basis of the technical solution of the first aspect, the present embodiment further provides another possible design for assisting in identifying a dangerous action, that is, after acquiring a live working monitoring image acquired by the binocular camera in real time, the method further includes, but is not limited to, the following steps S101 to S103.
And S101, extracting the limb action characteristics of the operator from the field operation monitoring image.
In the step S101, a specific extraction process may be implemented based on an existing feature extraction technology, for example, the field work monitoring image is introduced into a feature extraction model based on a convolutional neural network algorithm, so as to obtain the limb movement features of the operator.
And S102, judging whether the limb action characteristics are matched with pre-stored error limb action characteristics.
And S103, if yes, judging that dangerous actions exist at present, and sending out second alarm prompt information.
In step S103, the second warning message is preferably also a voice message, for example "attention! Risk action currently exists! Please avoid "to remind the operator to correct the error in time, further ensuring the safety of the operator and the equipment.
Therefore, based on the possible design one described in detail in the foregoing steps S101 to S103, in the distribution network uninterrupted operation process, whether the operator makes a dangerous motion or not can be identified from the motion characteristic matching angle, and when the operator makes a dangerous motion, an alarm prompt message is also sent out, so that the safety of the operator and the equipment is further ensured.
In this embodiment, on the basis of the technical solution of the first aspect, a second possible design for identifying whether the head of the operator is close to the charged body is provided, that is, after determining the initial coordinates/current coordinates of the charged body in the camera coordinate system, the method further includes, but is not limited to, the following steps S201 to S202.
S201, judging whether the distance from the charged body to the binocular camera is smaller than or equal to the safety distance threshold value or not according to the initial coordinate/current coordinate of the charged body.
S202, if yes, judging that dangerous actions exist at present, and sending out third alarm prompt information.
In the foregoing steps S201 to S202, since the binocular camera is attached to a helmet of the worker and the helmet is worn on the head of the worker during the work, the distance between the charged body and the binocular camera reflects the distance between the charged body and the head of the worker, and when the distance is found to be too short, it is necessary to trigger an alarm message. Furthermore, the third warning message is preferably also a voice message, for example "attention! The head is at risk of electric shock! Please avoid "to remind the operator to correct the error in time, further ensuring the safety of the operator and the equipment.
Therefore, based on the second possible design detailed in the foregoing steps S201 to S202, it can be determined in real time whether the charged body is too close to the head of the operator, and when the charged body is found to be too close, an alarm prompt message is also triggered, so as to further ensure the safety of the operator and the equipment.
On the basis of the technical solution of the foregoing first aspect, this embodiment further provides a third possible design for dynamically correcting the initial data of the charged body/limb, that is, after determining the initial coordinates of the charged body/limb in the camera coordinate system of the binocular camera and recording the acquisition timestamp of the live working monitoring image as the initial time of the charged body/limb, the method further includes, but is not limited to, the following steps S301 to S303.
S301, acquiring a new field operation monitoring image acquired by the binocular camera in real time.
S302, identifying whether the on-site operation monitoring image contains an image corresponding to the charged body/the limb.
And S303, if so, updating the initial coordinates of the charged body/the limb according to the newly determined coordinates of the charged body/the limb in the camera coordinate system, and updating the acquisition timestamp of the new field operation monitoring image to the initial time of the charged body/the limb.
In the foregoing steps S301 to S303, the scheme for acquiring the new on-site monitoring image, identifying the image, and acquiring the new determined coordinates may be derived by referring to the foregoing steps S1 to S4, and will not be described herein again. Considering that the motion track derived based on the attitude data gradually deviates from the real situation along with the extension of time, so that a large error may exist in the current coordinate determined after a long time, the initial coordinate and the initial moment of the charged body/limb are dynamically corrected based on the image recognition result of the new field operation monitoring image, the determination error of the subsequent corresponding current coordinate can be reduced, the correctness of the comparison result of the subsequent distance and the threshold value is ensured, and the safety of the operating personnel and equipment is further ensured.
In the step S303, it is preferable that the initial coordinates of the charged body are updated according to the newly determined coordinates of the charged body in the camera coordinate system, including but not limited to the following steps S3031 to S3032.
S3031, determining new coordinates of the charged body in the camera coordinate system and in the acquisition time stamp of the new field operation monitoring image according to the initial coordinates of the charged body and first attitude data which is recorded from the initial time of the charged body to the acquisition time stamp of the new field operation monitoring image and acquired by the first attitude sensor in real time.
In the step S3031, the scheme for determining the new coordinate may refer to the aforementioned step S5, and is not described herein again.
S3032, calculating an initial coordinate updating result (x, y, z) of the charged body according to the following formula:
in the formula (x)1,y1,z1) Representing the newly determined coordinates of the charged body in the camera coordinate system, (x)2,y2,z2) And the system also comprises a new coordinate of the charged body in the camera coordinate system and at the time of the acquisition time stamp of the new on-site operation monitoring image, wherein alpha represents a preset weight coefficient corresponding to the newly determined coordinate, beta represents a preset weight coefficient corresponding to the new coordinate, and alpha + beta is 1.
In step S3032, the initial coordinate of the charged body is updated based on the newly determined coordinate of the charged body and the weighted calculation result of the new coordinate, so that the updated correction accuracy of the initial coordinate can be improved by integrating the positioning results determined in two ways, the determination error of the subsequent current coordinate is further reduced, the correctness of the comparison result between the subsequent distance and the threshold value is ensured, and the safety of the operator and the equipment is further ensured. Further, for example, the preset weight coefficients α and β may be 0.5, respectively.
In step S303, similarly, the new coordinates of the limb in the camera coordinate system and at the time of the acquisition time stamp of the new work in place monitoring image may be determined according to the initial coordinates of the limb, the first posture data recorded from the initial time of the limb to the acquisition time stamp of the new work in place monitoring image and acquired in real time by the first posture sensor, and the second posture data recorded from the initial time of the limb to the acquisition time stamp of the new work in place monitoring image and acquired in real time by the second posture sensor, and then the initial coordinates of the limb may be updated based on the weighted calculation results of the newly determined coordinates and the new coordinates of the limb.
Therefore, based on the third possible design detailed in the foregoing steps S301 to S303, the initial coordinates and the initial time of the charged body/limb can be dynamically corrected based on the image recognition result of the new field operation monitoring image, the determination error of the subsequent corresponding current coordinate is reduced, the accuracy of the comparison result between the subsequent distance and the threshold value is ensured, and the safety of the operator and the equipment is further ensured.
As shown in fig. 2, a second aspect of this embodiment provides a virtual device for implementing the method according to any one of the first aspect or the possible designs of the first aspect, including an image acquisition module, an image recognition module, an initial data determination module, a current coordinate determination module, a current distance calculation module, a comparison and judgment module, and an alarm prompt module, which are sequentially connected in a communication manner;
the image acquisition module is used for acquiring a field operation monitoring image acquired by a binocular camera in real time, wherein the binocular camera is installed on a safety helmet of an operator;
the image identification module is used for identifying whether the on-site operation monitoring image contains a charged body image and a limb image of the operator;
the initial data determining module is configured to determine an initial coordinate of a charged body in a camera coordinate system of the binocular camera when the live working monitoring image is identified to include the image of the charged body, record a collection timestamp of the live working monitoring image as an initial time of the charged body, determine an initial coordinate of a limb in the camera coordinate system when the live working monitoring image is identified to include the image of the limb, and record a collection timestamp of the live working monitoring image as an initial time of the limb, where the charged body corresponds to the image of the charged body one to one, and the limb corresponds to the image of the limb one to one;
the current coordinate determination module is used for determining the current coordinate of the electrified body in the camera coordinate system according to the initial coordinate of the electrified body and first posture data which is recorded from the initial moment of the electrified body and acquired in real time by a first posture sensor, and determining the current coordinate of the limb in the camera coordinate system according to the initial coordinate of the limb, first posture data which is recorded from the initial moment of the limb and acquired in real time by the first posture sensor and second posture data which is recorded from the initial moment of the limb and acquired in real time by a second posture sensor, wherein the first posture sensor is bound with the binocular camera, and the second posture sensor is bound with the limb;
the current distance calculation module is used for calculating the current distance from the limb to the charged body according to the current coordinate of the charged body and the current coordinate of the limb;
the comparison and judgment module is used for judging whether the current distance is smaller than or equal to a preset safety distance threshold value;
and the alarm prompt module is used for judging that dangerous actions exist at present and sending first alarm prompt information when judging that the current distance is smaller than or equal to the safety distance threshold value.
For the working process, working details and technical effects of the foregoing apparatus provided in the second aspect of this embodiment, reference may be made to the method described in the first aspect or any one of the possible designs of the first aspect, which is not described herein again.
As shown in fig. 3, a third aspect of the present embodiment provides a computer device for executing the method according to any one of the first aspect or any one of the possible designs of the first aspect, and the computer device includes a memory and a processor, which are communicatively connected, where the memory is used to store a computer program, and the processor is used to read the computer program and execute the method according to any one of the possible designs of the first aspect or the first aspect. For example, the Memory may include, but is not limited to, a Random-Access Memory (RAM), a Read-Only Memory (ROM), a Flash Memory (Flash Memory), a First-in First-out (FIFO), and/or a First-in Last-out (FILO), and the like; the processor may not be limited to the use of a microprocessor of the model number STM32F105 family. In addition, the computer device may also include, but is not limited to, a power module, a display screen, and other necessary components.
For the working process, working details, and technical effects of the foregoing computer device provided in the third aspect of this embodiment, reference may be made to the method in the first aspect or any one of the possible designs in the first aspect, which is not described herein again.
A fourth aspect of the present embodiment provides a storage medium storing instructions of the method according to any one of the possible designs of the first aspect or the first aspect, that is, the storage medium has instructions stored thereon, which when executed on a computer, perform the method according to any one of the possible designs of the first aspect or the first aspect. The storage medium refers to a carrier for storing data, and may include, but is not limited to, a computer-readable storage medium such as a floppy disk, an optical disk, a hard disk, a flash Memory, a flash disk and/or a Memory Stick (Memory Stick), and the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.
For the working process, the working details, and the technical effects of the foregoing storage medium provided in the fourth aspect of this embodiment, reference may be made to the method in the first aspect or any one of the possible designs in the first aspect, which is not described herein again.
A fifth aspect of the present embodiments provides a computer program product comprising instructions which, when run on a computer, cause the computer to perform the method as set forth in the first aspect or any one of the possible designs of the first aspect. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable devices.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: modifications may be made to the embodiments described above, or equivalents may be substituted for some of the features described. And such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Finally, it should be noted that the present invention is not limited to the above alternative embodiments, and that various other forms of products can be obtained by anyone in light of the present invention. The above detailed description should not be taken as limiting the scope of the invention, which is defined in the claims, and which the description is intended to be interpreted accordingly.
Claims (10)
1. The utility model provides a be applied to distribution network not power off operation dangerous action intelligent recognition's method which characterized in that includes:
acquiring a field operation monitoring image acquired by a binocular camera in real time, wherein the binocular camera is installed on a safety helmet of an operator;
identifying whether the live working monitoring image contains a charged body image and a limb image of the operator;
when the live working monitoring image is identified to contain the charged body image, determining an initial coordinate of a charged body in a camera coordinate system of the binocular camera, and recording a collection time stamp of the live working monitoring image as an initial time of the charged body, wherein the charged body corresponds to the charged body image one to one;
when the limb image is identified to be contained in the on-site operation monitoring image, determining an initial coordinate of a limb in the camera coordinate system, and recording an acquisition timestamp of the on-site operation monitoring image as an initial moment of the limb, wherein the limb corresponds to the limb image one by one;
determining the current coordinate of the charged body in the camera coordinate system according to the initial coordinate of the charged body and first posture data which is recorded from the initial moment of the charged body and acquired by a first posture sensor in real time, wherein the first posture sensor is bound with the binocular camera;
determining the current coordinate of the limb in the camera coordinate system according to the initial coordinate of the limb, first posture data which is recorded from the initial moment of the limb and acquired by the first posture sensor in real time, and second posture data which is recorded from the initial moment of the limb and acquired by the second posture sensor in real time, wherein the second posture sensor is bound with the limb;
calculating to obtain the current distance from the limb to the charged body according to the current coordinate of the charged body and the current coordinate of the limb;
judging whether the current distance is smaller than or equal to a preset safety distance threshold value or not;
if so, judging that dangerous actions exist at present, and sending first alarm prompt information.
2. The method of claim 1, wherein after acquiring the live action surveillance image acquired by the binocular camera in real time, the method further comprises:
extracting limb action characteristics of the operator from the on-site operation monitoring image;
judging whether the limb action characteristics are matched with prestored wrong limb action characteristics;
if so, judging that dangerous actions exist at present, and sending out second alarm prompt information.
3. The method of claim 1, wherein identifying whether the live action surveillance image includes an image of a charged body and an image of an extremity of the operator comprises:
importing the on-site operation monitoring image into a pre-trained first detection model, and then judging whether the on-site operation monitoring image contains a charged body image according to a detection result of the first detection model, wherein the first detection model adopts a Faster R-CNN detection model, an SSD detection model or a YOLO detection model;
and importing the on-site operation monitoring image into a pre-trained second detection model, and then judging whether the on-site operation monitoring image contains a limb image according to a detection result of the second detection model, wherein the second detection model adopts a Faster R-CNN detection model, an SSD detection model or a YOLO detection model.
4. The method of claim 1, wherein determining initial coordinates of a charged body/appendage in a camera coordinate system of the binocular camera comprises:
determining the distance from the charged body/the limb to the origin of the camera coordinate system according to the binocular distance measuring principle of the binocular camera;
and acquiring initial coordinates of the charged body/the limb in the camera coordinate system according to the distance and the plane coordinates of the charged body image/the limb image in the on-site operation monitoring image.
5. The method of claim 1, wherein after determining the initial/current coordinates of the charged body in the camera coordinate system, the method further comprises:
judging whether the distance from the charged body to the binocular camera is smaller than or equal to the safety distance threshold or not according to the initial coordinate/current coordinate of the charged body;
if so, judging that dangerous actions exist at present, and sending out third alarm prompt information.
6. The method of claim 1, wherein after determining initial coordinates of a charged body/appendage in a camera coordinate system of the binocular camera and recording a time stamp of acquisition of the live action surveillance image as an initial time of the charged body/appendage, the method further comprises:
acquiring a new on-site operation monitoring image acquired by the binocular camera in real time;
identifying whether the live working monitoring image contains an image corresponding to the charged body/the limb;
if so, updating the initial coordinates of the charged body/the limb according to the newly determined coordinates of the charged body/the limb in the camera coordinate system, and updating the acquisition timestamp of the new field operation monitoring image to the initial time of the charged body/the limb.
7. The method of claim 6, wherein updating the initial coordinates of the charged body according to the newly determined coordinates of the charged body in the camera coordinate system comprises:
determining new coordinates of the charged body in the camera coordinate system and at the time of the acquisition timestamp of the new live working monitoring image according to the initial coordinates of the charged body and first attitude data which is recorded from the initial moment of the charged body to the acquisition timestamp of the new live working monitoring image and acquired by the first attitude sensor in real time;
calculating an initial coordinate updating result (x, y, z) of the charged body according to the following formula:
in the formula (x)1,y1,z1) Representing the newly determined coordinates of the charged body in the camera coordinate system, (x)2,y2,z2) And the system also comprises a new coordinate of the charged body in the camera coordinate system and at the time of the acquisition time stamp of the new on-site operation monitoring image, wherein alpha represents a preset weight coefficient corresponding to the newly determined coordinate, beta represents a preset weight coefficient corresponding to the new coordinate, and alpha + beta is 1.
8. A device applied to intelligent identification of dangerous actions of distribution network uninterrupted operation is characterized by comprising an image acquisition module, an image identification module, an initial data determination module, a current coordinate determination module, a current distance calculation module, a comparison judgment module and an alarm prompt module which are sequentially in communication connection;
the image acquisition module is used for acquiring a field operation monitoring image acquired by a binocular camera in real time, wherein the binocular camera is installed on a safety helmet of an operator;
the image identification module is used for identifying whether the on-site operation monitoring image contains a charged body image and a limb image of the operator;
the initial data determining module is configured to determine an initial coordinate of a charged body in a camera coordinate system of the binocular camera when the live working monitoring image is identified to include the image of the charged body, record a collection timestamp of the live working monitoring image as an initial time of the charged body, determine an initial coordinate of a limb in the camera coordinate system when the live working monitoring image is identified to include the image of the limb, and record a collection timestamp of the live working monitoring image as an initial time of the limb, where the charged body corresponds to the image of the charged body one to one, and the limb corresponds to the image of the limb one to one;
the current coordinate determination module is used for determining the current coordinate of the electrified body in the camera coordinate system according to the initial coordinate of the electrified body and first posture data which is recorded from the initial moment of the electrified body and acquired in real time by a first posture sensor, and determining the current coordinate of the limb in the camera coordinate system according to the initial coordinate of the limb, first posture data which is recorded from the initial moment of the limb and acquired in real time by the first posture sensor and second posture data which is recorded from the initial moment of the limb and acquired in real time by a second posture sensor, wherein the first posture sensor is bound with the binocular camera, and the second posture sensor is bound with the limb;
the current distance calculation module is used for calculating the current distance from the limb to the charged body according to the current coordinate of the charged body and the current coordinate of the limb;
the comparison and judgment module is used for judging whether the current distance is smaller than or equal to a preset safety distance threshold value;
and the alarm prompt module is used for judging that dangerous actions exist at present and sending first alarm prompt information when judging that the current distance is smaller than or equal to the safety distance threshold value.
9. A computer device comprising a memory and a processor communicatively coupled, wherein the memory is configured to store a computer program and the processor is configured to read the computer program and perform the method of any of claims 1 to 7.
10. A storage medium having stored thereon instructions for performing a method according to any one of claims 1-7 when the instructions are run on a computer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110560448.9A CN113283496B (en) | 2021-05-21 | 2021-05-21 | Method applied to intelligent identification of dangerous actions of distribution network uninterrupted operation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110560448.9A CN113283496B (en) | 2021-05-21 | 2021-05-21 | Method applied to intelligent identification of dangerous actions of distribution network uninterrupted operation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113283496A true CN113283496A (en) | 2021-08-20 |
| CN113283496B CN113283496B (en) | 2022-07-29 |
Family
ID=77280888
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110560448.9A Active CN113283496B (en) | 2021-05-21 | 2021-05-21 | Method applied to intelligent identification of dangerous actions of distribution network uninterrupted operation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113283496B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115471977A (en) * | 2022-07-31 | 2022-12-13 | 贵州电网有限责任公司 | Safe distance early warning system based on depth camera |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108846848A (en) * | 2018-06-25 | 2018-11-20 | 广东电网有限责任公司电力科学研究院 | A kind of the operation field method for early warning and device of fusion UWB positioning and video identification |
| CN110060299A (en) * | 2019-04-18 | 2019-07-26 | 中国测绘科学研究院 | Danger source identifies and positions method in passway for transmitting electricity based on binocular vision technology |
| CN110111016A (en) * | 2019-05-14 | 2019-08-09 | 深圳供电局有限公司 | Precarious position monitoring method, device and the computer equipment of operating personnel |
| WO2020024845A1 (en) * | 2018-08-01 | 2020-02-06 | 京东方科技集团股份有限公司 | Positioning method and apparatus |
| CN111753712A (en) * | 2020-06-22 | 2020-10-09 | 中国电力科学研究院有限公司 | Method, system and equipment for monitoring safety of power production personnel |
| CN112000094A (en) * | 2020-07-20 | 2020-11-27 | 山东科技大学 | Single-and-double-eye combined high-voltage transmission line hardware fitting online identification and positioning system and method |
-
2021
- 2021-05-21 CN CN202110560448.9A patent/CN113283496B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108846848A (en) * | 2018-06-25 | 2018-11-20 | 广东电网有限责任公司电力科学研究院 | A kind of the operation field method for early warning and device of fusion UWB positioning and video identification |
| WO2020024845A1 (en) * | 2018-08-01 | 2020-02-06 | 京东方科技集团股份有限公司 | Positioning method and apparatus |
| CN110060299A (en) * | 2019-04-18 | 2019-07-26 | 中国测绘科学研究院 | Danger source identifies and positions method in passway for transmitting electricity based on binocular vision technology |
| CN110111016A (en) * | 2019-05-14 | 2019-08-09 | 深圳供电局有限公司 | Precarious position monitoring method, device and the computer equipment of operating personnel |
| CN111753712A (en) * | 2020-06-22 | 2020-10-09 | 中国电力科学研究院有限公司 | Method, system and equipment for monitoring safety of power production personnel |
| CN112000094A (en) * | 2020-07-20 | 2020-11-27 | 山东科技大学 | Single-and-double-eye combined high-voltage transmission line hardware fitting online identification and positioning system and method |
Non-Patent Citations (1)
| Title |
|---|
| 祝晶 等: "一种配网带电作业机器人的目标线路识别与定位算法研究", 《工业控制计算机》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115471977A (en) * | 2022-07-31 | 2022-12-13 | 贵州电网有限责任公司 | Safe distance early warning system based on depth camera |
| CN115471977B (en) * | 2022-07-31 | 2023-12-01 | 贵州电网有限责任公司 | Safety distance early warning system based on depth camera |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113283496B (en) | 2022-07-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102412627B (en) | Image identification-based intelligent transformer substation state monitoring system | |
| CN111259892B (en) | Inspection method, inspection device, inspection equipment and inspection medium for state of indicator lamp | |
| CN109240311B (en) | Outdoor electric power field construction operation supervision method based on intelligent robot | |
| CN109214280B (en) | Shop identification method and device based on street view, electronic equipment and storage medium | |
| CN112229524A (en) | Body temperature screening method and device, terminal equipment and storage medium | |
| CN111753712A (en) | Method, system and equipment for monitoring safety of power production personnel | |
| CN109871764A (en) | A kind of abnormal behaviour recognition methods, device and storage medium | |
| CN112434669B (en) | Human body behavior detection method and system based on multi-information fusion | |
| CN113283496B (en) | Method applied to intelligent identification of dangerous actions of distribution network uninterrupted operation | |
| CN110874866B (en) | Video-based three-dimensional monitoring method and system for transformer substation | |
| CN115495540B (en) | Intelligent route identification method, system and medium for robot inspection | |
| CN108761290A (en) | Robot and its piping lane electric compartment method for inspecting, system, equipment, storage medium | |
| CN109992011B (en) | Secondary alignment method, device and equipment for inspection robot holder | |
| CN109948613A (en) | A kind of Infrared image recognition and device of arrester | |
| CN114665608B (en) | Intelligent sensing inspection system and method for transformer substation | |
| CN104426090A (en) | Intelligent cable inspection robot | |
| CN113343998A (en) | Reading monitoring system and method for electric power mechanical meter, computer equipment and application | |
| CN113925391A (en) | Tumble detection method and device based on cleaning robot and cleaning robot | |
| CN114898523A (en) | Power transmission line image monitoring method and device with distance measuring function | |
| CN103438819A (en) | Transformer substation tubular busbar deflection monitoring method | |
| CN114187511A (en) | Intelligent monitoring method for power transmission panoramic big data | |
| CN114041741B (en) | Data processing unit, processing device, surgical system, surgical instrument, and medium | |
| CN116206255A (en) | Dangerous area personnel monitoring method and device based on machine vision | |
| CN202334001U (en) | Image recognition-based intelligent transformer substation status monitoring system | |
| CN113449688B (en) | Power transmission tree obstacle recognition system based on image and laser point cloud data fusion |
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 |