CN115482217B - Transformer substation electric shock prevention video detection method based on Gaussian mixture model separation algorithm - Google Patents
Transformer substation electric shock prevention video detection method based on Gaussian mixture model separation algorithm Download PDFInfo
- Publication number
- CN115482217B CN115482217B CN202211146602.9A CN202211146602A CN115482217B CN 115482217 B CN115482217 B CN 115482217B CN 202211146602 A CN202211146602 A CN 202211146602A CN 115482217 B CN115482217 B CN 115482217B
- Authority
- CN
- China
- Prior art keywords
- personnel
- profile
- preset
- mixture model
- person
- 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
- 239000000203 mixture Substances 0.000 title claims abstract description 54
- 238000001514 detection method Methods 0.000 title claims abstract description 53
- 238000000926 separation method Methods 0.000 title claims abstract description 35
- 230000002265 prevention Effects 0.000 title claims description 6
- 238000012545 processing Methods 0.000 claims abstract description 14
- 238000001914 filtration Methods 0.000 claims abstract description 5
- 238000009826 distribution Methods 0.000 claims description 93
- 238000010586 diagram Methods 0.000 claims description 30
- 230000008859 change Effects 0.000 claims description 27
- 230000002776 aggregation Effects 0.000 claims description 15
- 238000004220 aggregation Methods 0.000 claims description 15
- 238000000034 method Methods 0.000 claims description 14
- 230000003044 adaptive effect Effects 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000003068 static effect Effects 0.000 claims description 4
- 238000012360 testing method Methods 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 206010014357 Electric shock Diseases 0.000 description 19
- 230000009286 beneficial effect Effects 0.000 description 12
- 238000010276 construction Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000011897 real-time detection Methods 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20036—Morphological image processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a transformer substation anti-electric shock video detection method based on a Gaussian mixture model separation algorithm, which comprises the following steps: collecting a video to be detected, and processing the video to be detected based on a Gaussian mixture model separation algorithm to obtain a target image; performing image binarization processing on the target image to obtain a binary image; performing open operation on the binary image by using a kernel of 10x10 to remove noise and performing closed operation to connect adjacent areas to obtain a result image; determining the minimum circumscribed positive rectangle of the result image by adopting an object contour detection method; filtering the minimum external positive rectangle according to a preset critical judgment condition to obtain a detection image; judging whether the outline of the detection image is intersected with a preset warning area or not, and generating a detection result.
Description
Technical Field
The invention relates to the technical field of substation anti-electric shock video detection, in particular to a substation anti-electric shock video detection method based on a Gaussian mixture model separation algorithm.
Background
Electric power is a high-risk field of safe production, and in recent years, each unit of an electric power system is put into a large amount of resources to strengthen on-site safety control. In the past, manual supervision is time-consuming and labor-consuming, risk early warning is not timely, evidence is difficult to obtain after that, and data utilization is low. With the continuous development of the technologies of the internet of things, edge computing, cloud computing and artificial intelligence, the video real-time analysis technology based on the edge computing is applied.
In the construction process of transformer substation interval, because the construction of workers can be illicitly carried higher and longer articles (such as ladders and the like), once the articles are too close to or touch the high-voltage wires, high-voltage electric air breakdown is extremely easy to generate to cause electric shock accidents. According to 35KV-110KV substation design specification, the safety clearance between charged parts of different phases of 35KV is 400mm; the safe clear distance between the charged portions of the 110KV different phases is 1000mm.
The application provides a transformer substation anti-electric shock video detection method based on a Gaussian mixture model separation algorithm, which utilizes the Gaussian mixture model separation algorithm to rapidly separate a foreground and a background, and detects the construction process of a transformer substation interval in real time, so that the accident occurrence probability is reduced, and the electric shock risk is prevented.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a transformer substation anti-electric shock video detection method based on a Gaussian mixture model separation algorithm, which utilizes the Gaussian mixture model separation algorithm to rapidly separate the foreground and the background, and carries out real-time detection on the construction process of a transformer substation interval, thereby reducing the occurrence probability of accidents and preventing the electric shock risk.
A transformer substation anti-electric shock video detection method based on a Gaussian mixture model separation algorithm comprises the following steps:
Collecting a video to be detected, and processing the video to be detected based on a Gaussian mixture model separation algorithm to obtain a target image; performing image binarization processing on the target image to obtain a binary image; performing open operation on the binary image by using a kernel of 10x10 to remove noise and performing closed operation to connect adjacent areas to obtain a result image; determining the minimum circumscribed positive rectangle of the result image by adopting an object contour detection method; filtering the minimum external positive rectangle according to a preset critical judgment condition to obtain a detection image; judging whether the outline of the detection image is intersected with a preset warning area or not, and generating a detection result.
As one embodiment of the invention, the adaptive Gaussian mixture model is estimated according to an online expectation maximization algorithm to obtain a Gaussian mixture model separation algorithm.
As an embodiment of the present invention, establishing the adaptive gaussian mixture model includes: respectively acquiring probability P (X N) of X N of any pixel value in the training image at time N and front B distributions which are used as models of scene and background; determining an update equation of the gaussian component according to the probability P (X N) and the model of the first B distributions used as a background of the scene; and updating the Gaussian components of the matching test values of each participation in establishing the self-adaptive Gaussian mixture model according to the updating equation until the preset condition is met, so as to obtain the self-adaptive Gaussian mixture model.
As an embodiment of the present invention, the calculation formula of the probability P (X N) includes:
；
Wherein, Is the weight parameter of the kth Gaussian component,/>Is the normal distribution of the kth gaussian component;
normal distribution of gaussian components/> Expressed as:
；
Wherein, Is the mean vector of the component distribution parameters in the D dimension,/>Is/>Covariance of Gaussian components, D is the dimension of the input vector x, x is the independent variable of the normal distribution of the components,/>For the probability of x at time N, I is the identity matrix,/>The variance value of unit distribution;
At the same time, K distributions are based on fitness values And (5) sequencing.
As an embodiment of the present invention, the value of B is:
；
where the threshold T is the smallest part of the model that is distributed to act as a scene background, B e B, The proportion of time reserved for the j-th color.
As one embodiment of the present invention, the update equation includes:
；
；
；
；
；
Wherein, Is the kth Gaussian component,/>Defining a time constant for determining the change;
meanwhile, if none of the K distributions matches the pixel value to be measured, the least probable component is replaced with a distribution having the current value as its mean, initial high variance, and low weight parameter.
As one embodiment of the present invention, estimating the adaptive gaussian mixture model according to an online expectation maximization algorithm includes:
The adaptive gaussian mixture model starts to be estimated according to the expected full statistical update equation and then switches to the L-component window version when the first L samples are processed.
As an embodiment of the present invention, performing image binarization processing on a target image includes:
；
Wherein x, y is the abscissa of the pixel point, For this point value, thresh is a threshold, maxval denotes a new pixel value that is assigned when the pixel value exceeds the threshold.
As an embodiment of the present invention, a transformer substation anti-electric shock video detection method based on a gaussian mixture model separation algorithm further includes:
acquiring the outline of the detection images, and generating a personnel outline distribution map of each detection image;
acquiring a yellow personnel profile distribution diagram of which the distance between the personnel profile distribution diagram and a preset warning area is smaller than a preset safety distance;
an orange personnel profile distribution diagram with the aggregation degree of the personnel profiles larger than a preset aggregation degree threshold value in the yellow personnel profile distribution diagram is obtained; the aggregation degree of the personnel profiles is determined according to the number of the personnel profiles in the yellow personnel profile distribution diagram and the distance between every two personnel profiles;
Judging whether a continuous preset orange personnel profile distribution map exists or not, if so, determining the continuous preset orange personnel profile distribution map as a red personnel profile distribution map, and sending out a site risk alarm;
And determining a dangerous value corresponding to the red personnel profile according to the change amplitude of the personnel profile in the continuously preset orange personnel profile, and sending reminding alarm information to the manager of the corresponding level according to the dangerous value.
As one embodiment of the present invention, determining a hazard value corresponding to a red personnel profile according to a continuously preset amplitude of variation of personnel profiles in an orange personnel profile comprises:
determining the current state of each personnel outline according to the variation amplitude of the personnel outline in the continuous preset orange personnel outline distribution diagram; wherein the current state includes a moving state and a stationary state;
when the change amplitude of the profile of the current person in the preset time is smaller than the preset change amplitude, the profile of the current person is in a static state, otherwise, the profile of the current person is in a motion state;
Determining the movement trend of the outline of the person in a movement state based on the change amplitude;
acquiring the nearest distance between each personnel outline in a motion state and the nearest other personnel outline in the motion state;
And calculating the go through dangers rate of the person corresponding to the person profile closest to the preset guard area in the current red person profile distribution diagram according to the movement trend of all the person profiles in the movement state and the closest distance between each person profile in the movement state and the nearest other person profiles in the movement state, and taking the go through dangers rate as a dangerous value of the corresponding red person profile distribution diagram.
The beneficial effects of the invention are as follows:
1. the Gaussian mixture model separation algorithm is used, and the self-adaptive background modeling method is provided, so that illumination change and shadow recognition can be well processed;
2. the detection efficiency is high, and real-time video detection can be realized on low-power-consumption edge computing equipment (xavier nx).
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention may be realized and attained by the structure particularly pointed out in the written description and drawings.
The technical scheme of the invention is further described in detail through the drawings and the embodiments.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
Fig. 1 is a method flowchart of a transformer substation electric shock prevention video detection method based on a gaussian mixture model separation algorithm in an embodiment of the invention;
Fig. 2 is a specific flowchart related to an alarm function in a transformer substation anti-electric shock video detection method based on a gaussian mixture model separation algorithm in an embodiment of the invention;
Fig. 3 is a flowchart of a judgment process for judging whether a worker has potential safety hazards due to aggregation in a critical area in a substation anti-electric shock video detection method based on a gaussian mixture model separation algorithm in an embodiment of the invention.
Detailed Description
The preferred embodiments of the present invention will be described below with reference to the accompanying drawings, it being understood that the preferred embodiments described herein are for illustration and explanation of the present invention only, and are not intended to limit the present invention.
Referring to fig. 1, an embodiment of the present invention provides a method for detecting a transformer substation anti-electric shock video based on a gaussian mixture model separation algorithm, including:
S101, acquiring a video to be detected;
S102, processing a video to be detected based on a Gaussian mixture model separation algorithm to obtain a target image;
s103, performing image binarization processing on the target image to obtain a binary image;
S104, performing open operation on the binary image by using a kernel of 10x10 to remove noise points and performing closed operation to connect adjacent areas to obtain a result image;
S105, determining the minimum circumscribed positive rectangle of the result image by adopting an object contour detection method;
S106, filtering the minimum external positive rectangle according to a preset critical judgment condition to obtain a detection image;
s107, judging whether the outline of the detection image is intersected with a preset warning area or not, and generating a detection result;
The working principle of the technical scheme is as follows: collecting a video to be tested of a transformer substation, processing the video to be tested based on a Gaussian mixture model separation algorithm to obtain a target image, and performing image binarization processing on the target image to obtain a binary image; then, performing open operation on the binary image by using a kernel of 10x10 to remove noise generated by video jitter and compression, performing close operation on the binary image by using the kernel of 10x10 to connect similar areas to obtain a result image, performing object contour detection on the result image to obtain a minimum circumscribed positive rectangle of the contour, filtering out the rectangle with the rectangular area smaller than an area threshold value by using the rectangular area, wherein the threshold value is generally set to be within a range of 100-300, and 150 is taken as the threshold value in the example; finally, setting all filtered object contours as rects, setting a certain object contour as rect, setting a warning area as ALARMRECT, judging whether the contour of a detection image is intersected with a preset warning area, and generating a detection result; as can be seen from the detection result shown in FIG. 3, the method can well separate the background from the foreground, reduce noise problems caused by video jitter, illumination change and the like, and can rapidly detect the changed part of the video so as to detect whether an object intrudes into a warning area;
The beneficial effects of the technical scheme are as follows: and the foreground and the background are rapidly separated by using a Gaussian mixture model separation algorithm, real-time detection is performed aiming at the construction process of the transformer substation interval, the accident occurrence probability is reduced, and the electric shock risk is prevented.
Referring to fig. 2, in one embodiment, a method for detecting a transformer substation anti-electric shock video based on a gaussian mixture model separation algorithm further includes determining whether to generate alarm information according to a detection result, and continuing to perform subsequent detection on the collected video after the current detection flow is finished;
The working principle and beneficial effects of the technical scheme are as follows: based on fig. 1, if an intersection exists between a certain object contour rect and a warning area ALARMRECT, namely rect n ALARMRECT > 0, an alarm is generated, and the pseudo code is as follows:
bool alarm = false;
for ( var rect in rects)
if ( rect ∩ alarmRect > 0 )
alarm = true;
the alarm can timely give out an alarm to related personnel, thereby being beneficial to reducing the accident occurrence probability.
In one embodiment, estimating the adaptive Gaussian mixture model according to an online expectation maximization algorithm to obtain a Gaussian mixture model separation algorithm;
The beneficial effects of the technical scheme are as follows: the Gaussian mixture model separation algorithm is beneficial to rapidly separating the foreground and the background.
In one embodiment, building an adaptive gaussian mixture model includes: respectively acquiring probability P (X N) of X N of any pixel value in the training image at time N and front B distributions which are used as models of scene and background; determining an update equation of the gaussian component according to the probability P (X N) and the model of the first B distributions used as a background of the scene; updating the Gaussian components of the matching test values of each of the self-adaptive Gaussian mixture models according to the updating equation until the preset conditions are met, so as to obtain the self-adaptive Gaussian mixture model;
The calculation formula of the probability P (X N) comprises:
；
Wherein, Is the weight parameter of the kth Gaussian component,/>Is the normal distribution of the kth gaussian component;
Normal distribution of kth gaussian component Expressed as:
；
Wherein, Is the mean vector of the component distribution parameters in the D dimension,/>Is/>Covariance of Gaussian components, D is the dimension of the input vector x, x is the independent variable of the normal distribution of the components,/>For the probability of x at time N, I is the identity matrix,/>The variance value of unit distribution; further, x is the argument range,/>Representation refers to the probability of a certain value in a range;
k distributions are based on fitness values Sequencing;
updating the equation includes:
；
；
；
；
；
Wherein, Is/>Gaussian component,/>Defining a time constant for determining the change;
Meanwhile, if none of the K distributions is matched with the pixel value to be detected, the least possible component is replaced by a distribution taking the current value as the mean value, the initial high variance and the low weight parameter;
The working principle and beneficial effects of the technical scheme are as follows: each pixel in the image is formed by K Gaussian distribution mixed modeling; the value of a pixel at time N is The probability of (2) can be written as: ; wherein/> Is/>Weight parameters of the gaussian component; /(I)Is the normal distribution of the kth component, expressed as; Wherein/>Is an average value/>Is/>Covariance of the components; k distributions are based on fitness value/>Ordering, the first B distributions are used as a model of the scene background, where B is estimated as/>；
The threshold T is the smallest part of the background model, B e B,The ratio of time left for the j-th color, w representing the ratio of time left for this color, b representing the distribution of the first b smaller colors; in other words, it is the minimum prior probability of the background in the scene; background subtraction is performed by marking one foreground pixel, the standard deviation of any pixel from any B distribution exceeding 2.5 standard deviations; the gaussian component of the first matching test value will be updated by the following update equation；
；
；
；
；
Wherein the method comprises the steps ofIs the kth gaussian component; /(I)Defining a time constant for determining the change; if none of the K distributions match the pixel value, the least likely component is replaced with a distribution having the current value as its mean, initial high variance, and low weight parameters.
In one embodiment, estimating the adaptive Gaussian mixture model according to an online expectation-maximization algorithm includes:
starting to estimate the adaptive Gaussian mixture model according to an expected full statistics update equation, and then switching to an L-event window version when the first L samples are processed;
The working principle and beneficial effects of the technical scheme are as follows: we start estimating the gaussian mixture model by the expected full statistical update equation and then switch to the L-party window version when processing the first L samples; before all L samples can be collected, the expected adequate statistical update equation provides a good estimate at the beginning; this initial estimation improves the accuracy of the estimation and also improves the performance of the tracker, allowing for fast convergence on a stable background model; the L-party window update equation takes precedence over the latest data, so that the tracker can adapt to the change of the environment;
the on-line EM algorithm for which sufficient statistics are expected is shown in the left column, while the L-event window version is shown in the right column:
；
the tracker can adapt to the change of the environment, and the performance of the tracker and the accuracy of estimation are improved.
The online EM algorithm for which sufficient statistics are expected is shown in the left column, while the L-party window version is shown in the right column.
In one embodiment, performing image binarization processing on a target image includes:
；
Wherein x, y is the abscissa of the pixel point, For the value of this point, thresh is a threshold, maxval denotes a new pixel value that is assigned when the pixel value exceeds the threshold;
In the present embodiment of the present invention, in the present embodiment, ， />；
The beneficial effects of the technical scheme are as follows: it is beneficial to highlight the contours of the target image.
Referring to fig. 3, in an embodiment, a method for detecting a transformer substation anti-electric shock video based on a gaussian mixture model separation algorithm further includes:
s201, acquiring the outline of the detection image, and generating a personnel outline distribution map of each detection image;
S202, acquiring a yellow personnel profile distribution diagram with a preset guard area smaller than a preset safety distance from the personnel profile distribution diagram;
S203, acquiring an orange personnel profile map with the aggregation degree of personnel profiles in the yellow personnel profile map being greater than a preset aggregation degree threshold; the aggregation degree of the personnel profiles is determined according to the number of the personnel profiles in the yellow personnel profile distribution diagram and the distance between every two personnel profiles;
S204, judging whether a continuous preset orange personnel profile distribution map exists or not, if so, determining the continuous preset orange personnel profile distribution map as a red personnel profile distribution map, and sending out an on-site risk alarm;
S205, determining a dangerous value corresponding to the red personnel profile according to the change amplitude of the personnel profile in the continuously preset orange personnel profile, and sending reminding alarm information to the manager of the corresponding level according to the dangerous value;
the working principle of the technical scheme is as follows: acquiring the outline of the detection images, and generating a personnel outline distribution map of each detection image; then, a yellow personnel profile distribution diagram with the preset guard area smaller than a preset safety distance in the personnel profile distribution diagram is obtained; wherein the yellow personnel profile is a partial or complete personnel profile; an orange personnel profile distribution diagram with the aggregation degree of the personnel profiles larger than a preset aggregation degree threshold value in the yellow personnel profile distribution diagram is obtained; the aggregation degree of the personnel profiles is determined according to the number of the personnel profiles in the yellow personnel profile distribution diagram and the distance between every two personnel profiles; the more people contours in the yellow people contour distribution map, the higher the aggregation level; the closer the distance between every two person contours in the yellow person contour distribution map is, the higher the aggregation degree is; continuously acquiring the outline of the detection image, and reserving a preset personnel outline graph each time; for example, 4 personnel profile graphs need to be reserved, and 4 personnel profile graphs are reserved currently, then after the next personnel profile graph is resolved, a reserved sequence is added, and the personnel profile graph with the earliest time in the original reserved sequence is deleted; judging whether a continuous preset orange personnel profile distribution map exists or not, if so, determining the continuous preset orange personnel profile distribution map as a red personnel profile distribution map, and sending out a site risk alarm; for example, 4 personnel profile graphs need to be reserved, 3 preset personnel profile graphs are reserved, and if 3 continuous orange personnel profile graphs exist in the currently reserved 4 personnel profile graphs, the continuous 3 personnel profile graphs are determined to be red personnel profile graphs; finally, determining a dangerous value corresponding to the red personnel profile according to the change amplitude of the personnel profile in the continuously preset orange personnel profile, and sending reminding alarm information to the manager of the corresponding level according to the dangerous value;
the beneficial effects of the technical scheme are as follows: when the transformer substation works, working staff can have the possibility of gathering office work because of working needs, when people gather, the situations such as collision falling are likely to occur, if the gathering staff is at the edge of the warning area at the moment, the falling staff is likely to fall in the warning area at the moment, so that potential safety hazards are caused, through the technical scheme, the safe falling distance is additionally set up (namely, the preset safe distance), and the personnel gathering degree is analyzed, so that the accident occurrence probability is further reduced, and because the personnel quality of constructors is different, the constructors pay attention to when the alarm sent by the system is not available, the danger value is set up, the corresponding danger alarm is sent to the manager of the corresponding grade according to the danger value, and the electric shock prevention reminding effect is further improved.
In one embodiment, determining the hazard value for the red personnel profile from the magnitude of the change in personnel profile in the continuously preset orange personnel profile comprises:
determining the current state of each personnel outline according to the variation amplitude of the personnel outline in the continuous preset orange personnel outline distribution diagram; wherein the current state includes a moving state and a stationary state;
when the change amplitude of the profile of the current person in the preset time is smaller than the preset change amplitude, the profile of the current person is in a static state, otherwise, the profile of the current person is in a motion state;
Determining the movement trend of the outline of the person in a movement state based on the change amplitude;
acquiring the nearest distance between each personnel outline in a motion state and the nearest other personnel outline in the motion state;
According to the motion trend of all the personnel contours in the motion state and the nearest distance between each personnel contour in the motion state and the nearest other personnel contours in the motion state, calculating to obtain the go through dangers rate of the personnel corresponding to the personnel contour closest to the preset guard area in the current red personnel contour distribution map, and taking the go through dangers rate as the dangerous value of the corresponding red personnel contour distribution map;
The working principle of the technical scheme is as follows: determining the current state of each personnel outline according to the variation amplitude of the personnel outline in the continuous preset orange personnel outline distribution diagram; wherein the current state includes a moving state and a stationary state; the change amplitude refers to the change amplitude of the personnel outline of the same personnel, which is determined by any personnel outline in the current orange personnel outline distribution diagram according to the existing position and any personnel outline in the next orange personnel outline distribution diagram according to the existing position, and the action amplitude of the two personnel outlines; when the change amplitude of the profile of the current person in the preset time is smaller than the preset change amplitude, the profile of the current person is in a static state, otherwise, the profile of the current person is in a motion state; the preset time refers to acquisition time of continuously preset orange personnel profile distribution diagrams, and the variation amplitude of the current personnel profile in the preset time refers to variation amplitude of the current personnel profile in the continuously preset orange personnel profile distribution diagrams; determining the movement trend of the outline of the person in a movement state based on the change amplitude; movement tendencies, i.e., movement directions, include movement in a direction facing the contour of the nearest other person and movement in an opposite direction to the contour of the nearest other person; acquiring the nearest distance between each personnel outline in a motion state and the nearest other personnel outline in the motion state; wherein the nearest distance between the two persons in motion state is the same; finally, according to the movement trend of all the personnel contours in the movement state and the nearest distance between each personnel contour in the movement state and the nearest other personnel contours in the movement state, calculating to obtain the go through dangers rate of the personnel corresponding to the personnel contour closest to the preset guard area in the current red personnel contour distribution map, and taking the go through dangers rate as a dangerous value of the corresponding red personnel contour distribution map; typically, the calculation is started by the collision of the outermost personnel outline of the gathered personnel to the inner personnel outline; the inner personnel are the personnel closest to the preset warning area; the calculation formula of go through dangers rate is preferably: wherein/> For go through dangers rates of persons corresponding to the person profile closest to the preset guard area in the current red person profile distribution diagram,/>For the number of person profiles in the current red person profile,/>Is the current state of personnel outline,/>When=1, the person profile is the outermost person profile, and when i=m-1, the person profile is the nearest person profile to the innermost person profile,/>The weight value corresponding to the current state of the current personnel outline is the same as/>In the motion state,/>=1, When/>In a stationary state,/>=0，/>For the trend of movement of personnel profile,/>A preset collision rate corresponding to the movement trend of the current personnel profile, wherein when the movement trend of the ith personnel profile faces the (i+1) th personnel profile,/>When the motion of the ith personnel outline tends not to face the (i+1) th personnel outline,/>=V, V > > V, l is the closest distance between the ith and the (i+1) th person profiles,/>For the risk weight values of the ith personnel outline and the (i+1) th personnel outline distance, the closer the distance l is,/>The larger;
the beneficial effects of the technical scheme are as follows: the probability of collision of the personnel is determined through the change amplitude of the personnel outline, so that go through dangers rates of the personnel closest to the preset warning area are determined, dangerous values corresponding to the red personnel outline distribution diagram are obtained, and the warning effect of preventing electric shock is improved.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (8)
1. A transformer substation electric shock prevention video detection method based on a Gaussian mixture model separation algorithm is characterized by comprising the following steps: collecting a video to be detected, and processing the video to be detected based on a Gaussian mixture model separation algorithm to obtain a target image; performing image binarization processing on the target image to obtain a binary image; performing open operation on the binary image by using a kernel of 10x10 to remove noise and performing closed operation to connect adjacent areas to obtain a result image; determining the minimum circumscribed positive rectangle of the result image by adopting an object contour detection method; filtering the minimum external positive rectangle according to a preset critical judgment condition to obtain a detection image; judging whether the outline of the detection image is intersected with a preset warning area or not, and generating a detection result;
acquiring the outline of the detection images, and generating a personnel outline distribution map of each detection image;
acquiring a yellow personnel profile distribution diagram of which the distance between the personnel profile distribution diagram and a preset warning area is smaller than a preset safety distance;
an orange personnel profile distribution diagram with the aggregation degree of the personnel profiles larger than a preset aggregation degree threshold value in the yellow personnel profile distribution diagram is obtained; the aggregation degree of the personnel profiles is determined according to the number of the personnel profiles in the yellow personnel profile distribution diagram and the distance between every two personnel profiles;
Judging whether a continuous preset orange personnel profile distribution map exists or not, if so, determining the continuous preset orange personnel profile distribution map as a red personnel profile distribution map, and sending out a site risk alarm;
Determining a dangerous value corresponding to the red personnel profile according to the change amplitude of the personnel profile in the continuously preset orange personnel profile, and sending reminding alarm information to the manager of the corresponding level according to the dangerous value;
Determining a dangerous value corresponding to the red personnel profile according to the variation amplitude of the personnel profile in the orange personnel profile continuously preset, wherein the dangerous value comprises the following steps:
determining the current state of each personnel outline according to the variation amplitude of the personnel outline in the continuous preset orange personnel outline distribution diagram; wherein the current state includes a moving state and a stationary state;
when the change amplitude of the profile of the current person in the preset time is smaller than the preset change amplitude, the profile of the current person is in a static state, otherwise, the profile of the current person is in a motion state;
Determining the movement trend of the outline of the person in a movement state based on the change amplitude;
acquiring the nearest distance between each personnel outline in a motion state and the nearest other personnel outline in the motion state;
According to the motion trend of all the personnel contours in the motion state and the nearest distance between each personnel contour in the motion state and the nearest other personnel contours in the motion state, calculating to obtain the go through dangers rate of the personnel corresponding to the personnel contour closest to the preset guard area in the current red personnel contour distribution map, and taking the go through dangers rate as the dangerous value of the corresponding red personnel contour distribution map;
wherein, the calculation formula of go through dangers rate is: Wherein dangerous is go through dangers rate of a person corresponding to a person profile closest to a preset guard area in the current red person profile, m is the number of person profiles in the current red person profile, g is the current state of the person profile, i=1 indicates that the person profile is the outermost person profile, i=m-1 indicates that the person profile is the person profile closest to the innermost person profile, Q g,i is a weight value corresponding to the current state of the current person profile, Q g,i =1 when g is in a moving state, Q g,i =0 when g is in a stationary state, d is a moving trend of the person profile, H d,i is a preset collision rate corresponding to a moving trend of the current person profile, wherein H d,i =v when the moving trend of the i person profile faces the i+1th person profile, H d,i =v, V > V when the moving trend of the i person profile does not face the i+1th person profile, L is the nearest distance L to the i+1th person profile, and L is a distance L35, the distance L is larger than the nearest person profile.
2. The substation anti-electric shock video detection method based on the Gaussian mixture model separation algorithm according to claim 1 is characterized in that the adaptive Gaussian mixture model is estimated according to an online expectation maximization algorithm to obtain the Gaussian mixture model separation algorithm.
3. The substation anti-electric shock video detection method based on the Gaussian mixture model separation algorithm according to claim 2, wherein the establishing of the adaptive Gaussian mixture model comprises the following steps: respectively acquiring probability P (X N) of X N of any pixel value in the training image at time N and front B distributions which are used as models of scene and background; determining an update equation of the gaussian component according to the probability P (X N) and the model of the first B distributions used as a background of the scene; and updating the Gaussian components of the matching test values of each participation in establishing the self-adaptive Gaussian mixture model according to the updating equation until the preset condition is met, so as to obtain the self-adaptive Gaussian mixture model.
4. The substation anti-electric shock video detection method based on the Gaussian mixture model separation algorithm according to claim 3, wherein the calculation formula of the probability P (X N) comprises:
Where w k is the weight parameter of the kth gaussian component, η (x N;θk) is the normal distribution of the kth gaussian component;
the normal distribution η (x N;θk) of the kth gaussian component is expressed as:
Mu k is a mean vector of component distribution parameters in D dimension, sigma k＝σk 2 I is covariance of kth Gaussian component, D is dimension of input vector x, x is independent variable of component normal distribution, x N is probability of x at N time, I is an identity matrix, sigma k is variance value of unit distribution, and K is number of Gaussian distribution in the mixed model;
At the same time, the K distributions are ordered according to fitness value w k/σk.
5. The substation electric shock prevention video detection method based on the Gaussian mixture model separation algorithm, according to claim 3, wherein the value of B is as follows:
Where the threshold T is the smallest part of the model that is distributed to act as a scene background, b.epsilon.B, w j is the proportion of time left for the j-th color.
6. The method for detecting the electric shock video of the transformer substation based on the Gaussian mixture model separation algorithm according to claim 4, wherein the updating equation comprises the following steps:
Where ω k is the kth gaussian component and 1/α defines the time constant that determines the variation;
Meanwhile, if none of the K distributions matches the pixel value to be measured, the least probable component is replaced with a distribution having the current value as its mean, initial high variance, and low weight parameter.
7. The substation anti-electric shock video detection method based on the gaussian mixture model separation algorithm according to claim 2, wherein the estimating the adaptive gaussian mixture model according to the online expectation maximization algorithm comprises:
The adaptive gaussian mixture model starts to be estimated according to the expected full statistical update equation and then switches to the L-component window version when the first L samples are processed.
8. The substation anti-electric shock video detection method based on the Gaussian mixture model separation algorithm according to claim 1, wherein the image binarization processing is performed on the target image, and the method comprises the following steps:
where x, y is the abscissa of the pixel point, g (x, y) is the value of the point, thresh is the threshold, maxval is the new pixel value that is assigned when the pixel value exceeds the threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211146602.9A CN115482217B (en) | 2022-09-21 | 2022-09-21 | Transformer substation electric shock prevention video detection method based on Gaussian mixture model separation algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211146602.9A CN115482217B (en) | 2022-09-21 | 2022-09-21 | Transformer substation electric shock prevention video detection method based on Gaussian mixture model separation algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115482217A CN115482217A (en) | 2022-12-16 |
| CN115482217B true CN115482217B (en) | 2024-05-10 |
Family
ID=84423693
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211146602.9A Active CN115482217B (en) | 2022-09-21 | 2022-09-21 | Transformer substation electric shock prevention video detection method based on Gaussian mixture model separation algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115482217B (en) |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4916450A (en) * | 1988-05-12 | 1990-04-10 | Radar Control Systems Corporation | Radar system for headway control of a vehicle |
| CN102180139A (en) * | 2011-04-22 | 2011-09-14 | 钟延沁 | Proximity trigger safety air bag and safety control method thereof |
| CN102826423A (en) * | 2011-06-14 | 2012-12-19 | 株式会社日立制作所 | Elevator safe operation system and safe operation method |
| DE102013214305A1 (en) * | 2013-07-22 | 2015-01-22 | Bayerische Motoren Werke Aktiengesellschaft | Method for avoiding a collision of a vehicle with at least one object, driver assistance system and vehicle |
| CN105953762A (en) * | 2016-07-07 | 2016-09-21 | 国网山东省电力公司经济技术研究院 | Power transmission line modeling method used for net distance measurement and distance measurement device thereof |
| CN108274469A (en) * | 2018-04-23 | 2018-07-13 | 东北大学 | Vacuum mechanical-arm anticollision detecting system and detection method based on multidimensional vision sensor |
| WO2018130016A1 (en) * | 2017-01-10 | 2018-07-19 | 哈尔滨工业大学深圳研究生院 | Parking detection method and device based on monitoring video |
| JP2018206107A (en) * | 2017-06-06 | 2018-12-27 | 東芝デベロップメントエンジニアリング株式会社 | Approaching object detecting device |
| CN111985388A (en) * | 2020-08-18 | 2020-11-24 | 深圳市自行科技有限公司 | Pedestrian attention detection driving assistance system, device and method |
| CN113012315A (en) * | 2021-02-23 | 2021-06-22 | 浙江三一装备有限公司 | Safety monitoring system and method for working machine and working machine |
| CN113052125A (en) * | 2021-04-09 | 2021-06-29 | 内蒙古科电数据服务有限公司 | Construction site violation image recognition and alarm method |
| CN113344874A (en) * | 2021-06-04 | 2021-09-03 | 温州大学 | Pedestrian boundary crossing detection method based on Gaussian mixture modeling |
| CN113890193A (en) * | 2021-09-29 | 2022-01-04 | 国网内蒙古东部电力有限公司呼伦贝尔供电公司 | Working state monitoring method and system for electric tracing system of transformer substation circuit breaker |
| CN114261389A (en) * | 2021-11-29 | 2022-04-01 | 东风商用车有限公司 | Running track collision risk analysis method of automatic driving commercial vehicle |
| CN114708508A (en) * | 2022-04-21 | 2022-07-05 | 广东电网有限责任公司东莞供电局 | Method for identifying moving small animal |
| CN114913420A (en) * | 2022-05-17 | 2022-08-16 | 中国矿业大学 | Dynamic obstacle monitoring system and monitoring method for anti-impact drilling robot |
-
2022
- 2022-09-21 CN CN202211146602.9A patent/CN115482217B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4916450A (en) * | 1988-05-12 | 1990-04-10 | Radar Control Systems Corporation | Radar system for headway control of a vehicle |
| CN102180139A (en) * | 2011-04-22 | 2011-09-14 | 钟延沁 | Proximity trigger safety air bag and safety control method thereof |
| CN102826423A (en) * | 2011-06-14 | 2012-12-19 | 株式会社日立制作所 | Elevator safe operation system and safe operation method |
| DE102013214305A1 (en) * | 2013-07-22 | 2015-01-22 | Bayerische Motoren Werke Aktiengesellschaft | Method for avoiding a collision of a vehicle with at least one object, driver assistance system and vehicle |
| CN105953762A (en) * | 2016-07-07 | 2016-09-21 | 国网山东省电力公司经济技术研究院 | Power transmission line modeling method used for net distance measurement and distance measurement device thereof |
| WO2018130016A1 (en) * | 2017-01-10 | 2018-07-19 | 哈尔滨工业大学深圳研究生院 | Parking detection method and device based on monitoring video |
| JP2018206107A (en) * | 2017-06-06 | 2018-12-27 | 東芝デベロップメントエンジニアリング株式会社 | Approaching object detecting device |
| CN108274469A (en) * | 2018-04-23 | 2018-07-13 | 东北大学 | Vacuum mechanical-arm anticollision detecting system and detection method based on multidimensional vision sensor |
| CN111985388A (en) * | 2020-08-18 | 2020-11-24 | 深圳市自行科技有限公司 | Pedestrian attention detection driving assistance system, device and method |
| CN113012315A (en) * | 2021-02-23 | 2021-06-22 | 浙江三一装备有限公司 | Safety monitoring system and method for working machine and working machine |
| CN113052125A (en) * | 2021-04-09 | 2021-06-29 | 内蒙古科电数据服务有限公司 | Construction site violation image recognition and alarm method |
| CN113344874A (en) * | 2021-06-04 | 2021-09-03 | 温州大学 | Pedestrian boundary crossing detection method based on Gaussian mixture modeling |
| CN113890193A (en) * | 2021-09-29 | 2022-01-04 | 国网内蒙古东部电力有限公司呼伦贝尔供电公司 | Working state monitoring method and system for electric tracing system of transformer substation circuit breaker |
| CN114261389A (en) * | 2021-11-29 | 2022-04-01 | 东风商用车有限公司 | Running track collision risk analysis method of automatic driving commercial vehicle |
| CN114708508A (en) * | 2022-04-21 | 2022-07-05 | 广东电网有限责任公司东莞供电局 | Method for identifying moving small animal |
| CN114913420A (en) * | 2022-05-17 | 2022-08-16 | 中国矿业大学 | Dynamic obstacle monitoring system and monitoring method for anti-impact drilling robot |
Non-Patent Citations (2)
| Title |
|---|
| Identification of independent structural shocks in the presence of multiple Gaussian components;Simone Maxand;《Econometrics and Statistics》;20181103;第16卷;第55-68页 * |
| 基于图像处理的变电站视频智能分析研究;朱红岷 等;《计算机工程与应用》;20180401;第54卷(第07期);第264-270页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115482217A (en) | 2022-12-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106407928B (en) | Transformer composite insulator casing monitoring method and system based on raindrop identification | |
| CN103632158B (en) | Forest fire prevention monitor method and forest fire prevention monitor system | |
| CN111428617A (en) | Video image-based distribution network violation maintenance behavior identification method and system | |
| CN111144232A (en) | Transformer substation electronic fence monitoring method based on intelligent video monitoring, storage medium and equipment | |
| CN112347916B (en) | Video image analysis-based power field operation safety monitoring method and device | |
| CN106412501A (en) | Construction safety behavior intelligent monitoring system based on video and monitoring method thereof | |
| CN104463253B (en) | Passageway for fire apparatus safety detection method based on adaptive background study | |
| CN111488799B (en) | Falling object identification method and system based on image identification | |
| CN113903081A (en) | Visual identification artificial intelligence alarm method and device for images of hydraulic power plant | |
| CN111222478A (en) | Construction site safety protection detection method and system | |
| CN110443196A (en) | Fire-fighting road occupying detection method based on SSIM algorithm | |
| CN113313005A (en) | Power transmission conductor on-line monitoring method and system based on target identification and reconstruction | |
| CN113469654B (en) | Multi-level safety control system of transformer substation based on intelligent algorithm fuses | |
| CN110120155A (en) | A kind of chemical industry plant area vehicle overload overload intelligent monitoring and alarming system | |
| CN112132043B (en) | Fire fighting channel occupation self-adaptive detection method based on monitoring video | |
| CN111738336A (en) | Image detection method based on multi-scale feature fusion | |
| CN114782897A (en) | Dangerous behavior detection method and system based on machine vision and deep learning | |
| CN112308032B (en) | Intelligent cooperative linkage control method for underground air door through video detection | |
| CN115100813B (en) | Intelligent community system based on digital twins | |
| CN111476979A (en) | Intelligent security and stability maintenance method and system based on multi-model analysis | |
| CN114782988A (en) | Construction environment-oriented multi-stage safety early warning method | |
| CN107222726A (en) | Electric power facility external force damage prevention early warning scheme | |
| CN115482217B (en) | Transformer substation electric shock prevention video detection method based on Gaussian mixture model separation algorithm | |
| CN104077591A (en) | Intelligent and automatic computer monitoring system | |
| CN114387662A (en) | Method for identifying unsafe behaviors of underground personnel in coal mine |
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 |