CN116309592B - Fire control channel illegal stop detection method - Google Patents
Fire control channel illegal stop detection method Download PDFInfo
- Publication number
- CN116309592B CN116309592B CN202310579431.7A CN202310579431A CN116309592B CN 116309592 B CN116309592 B CN 116309592B CN 202310579431 A CN202310579431 A CN 202310579431A CN 116309592 B CN116309592 B CN 116309592B
- Authority
- CN
- China
- Prior art keywords
- target
- connected domain
- image
- target connected
- difference 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.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 17
- 230000000877 morphologic effect Effects 0.000 claims abstract description 13
- 238000000034 method Methods 0.000 claims description 28
- 238000010606 normalization Methods 0.000 claims description 8
- 238000004891 communication Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 abstract description 2
- 238000013459 approach Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 241000949477 Toona ciliata Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention relates to the technical field of image processing, and discloses a fire control channel illegal stop detection method, which comprises the following steps: acquiring a difference image of a video frame image and a standard image, and judging whether the difference image contains a connected domain or not by using gray values of pixel points in the difference image and neighbor pixel points of the pixel points; when the difference images contain connected domains, acquiring each connected domain in each difference image in a set time for matching, and taking the matched target connected domain in each difference image and other difference images as a target connected domain set; judging whether the target connected domain is an automobile or not by using the symmetry degree and the morphological value of all the target connected domains in the target connected domain set; when the target connected domain is judged to be an automobile, illegal stopping occurs in the fire-fighting access area; effectively avoiding potential safety hazards caused by the fact that illegal vehicles cannot be accurately identified due to the fact that other moving objects exist in a background area, and the vehicles in a fire-fighting channel cannot be processed in time.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a fire fighting access illegal stop detection method.
Background
At present, although automobiles are popularized, the parking space cannot meet the requirements, a plurality of owners can easily park the automobiles in a fire-fighting channel for the convenience of the owners, so that great potential safety hazards are caused, whether the fire-fighting channel is provided with an illegal parking of the automobiles is judged, and whether the fire-fighting channel is provided with the illegal parking of the automobiles is judged by detecting images of positions of the fire-fighting channel through cameras arranged beside the fire-fighting channel.
At present, whether a fire fighting access is provided with a vehicle is judged, whether the image in the fire fighting access is provided with the vehicle is determined by dividing the image in the shot fire fighting access, and at present, a frame difference method is adopted to distinguish a foreground area from a background area when the image is divided, and because the background area has certain motion, a target area obtained by the method possibly contains interference information affecting judgment, whether the fire fighting access is provided with the vehicle is judged inaccurately, and great potential safety hazards are caused when the vehicle which is in the fire fighting access is not timely processed.
Disclosure of Invention
The invention is used for solving the problem of inaccurate judgment of whether a illegal parking vehicle exists in a fire-fighting channel at present, and provides a fire-fighting channel illegal parking detection method capable of accurately separating a background area from a foreground area in the fire-fighting channel, accurately judging whether the fire-fighting channel has the illegal parking vehicle or not through the separated area, comprising the following steps:
acquiring a standard image of a fire fighting access area and a video frame image in a video of the fire fighting access area;
sequentially obtaining difference images of the video frame image and the standard image, and judging whether the difference images contain connected domains or not by using gray values of pixel points in the difference images and neighbor pixel points of the pixel points;
when the difference image contains a connected domain, the difference image is used as an initial difference image; acquiring a video frame image corresponding to the initial difference image as an initial video frame image, traversing the difference image of the video frame image and the standard image in a set time after the initial video frame image is acquired; stopping traversing when the connected domain does not exist in the difference image of the video frame image and the standard image within the traversing setting time, and newly acquiring the next difference image containing the connected domain; taking the video frame image corresponding to the difference image as a new initial video frame image, traversing to obtain the difference image of the video frame image and the standard image in a set time after the new initial video frame image is obtained, and sequentially iterating;
stopping iteration when all difference images of all video frame images and standard images in the set time are traversed to have connected domains;
matching each connected domain in each difference image obtained in the set time, and taking the matched target connected domain in each difference image and other difference images as a target connected domain set;
acquiring the central line of each target communicating domain in the target communicating domain set, and acquiring the symmetry degree of the target communicating domain by utilizing the regions on two sides of the central line of each target communicating domain;
acquiring a morphological value of each target connected domain in the target connected domain set by utilizing the area of each target connected domain and the convex hull area of the target connected domain;
judging whether the target connected domain is an automobile or not by using the symmetry degree and the morphological value of all the target connected domains in the target connected domain set;
when the target connected domain is judged to be an automobile, illegal stop occurs in the fire-fighting access area.
Further, the method for obtaining the symmetry degree of the target connected domain by utilizing the regions at two sides of the center line of the target connected domain comprises the following steps:
dividing the target communication domain through the center line to obtain the overlapping areas of the areas on the two sides of the center line after division;
and obtaining the symmetry degree of the target communicating region according to the superposition area of the areas at the two sides of the central line and the total area of the target communicating region.
Further, the method for judging whether the target connected domain is an automobile or not by using the symmetry degree and the morphological value of all the target connected domains in the target connected domain set comprises the following steps:
utilizing the symmetry degree and the morphology value of all the target connected domains in the target connected domain set to obtain the probability that the target connected domain is an automobile;
comparing the probability of the obtained target connected domain being the automobile with a set threshold value to judge whether the target connected domain is the automobile or not;
the expression of the probability of obtaining the target connected domain as the automobile is as follows:
wherein:a probability value indicating that the target connected domain is an automobile;the average value of symmetry degrees of all the target connected domains in the target connected domain set;the maximum value of the symmetry degree of the target connected domain in the target connected domain set;the minimum value of the symmetry degree of the target connected domain in the target connected domain set;is the first in the target connected domain setMorphology values of the individual target connected domains;the method is an average value of morphological values of all the target connected domains in the target connected domain set;representing the number of the target connected domains in the target connected domain set;representing a normalization function;representing an exponential function.
Further, the method for judging whether the difference image contains the connected domain or not by using the gray value of the pixel point in the difference image and the gray value of the neighborhood pixel point of the pixel point comprises the following steps:
acquiring the uniformity degree of each pixel point according to the gray value of each pixel point in the difference image and the gray difference value of each pixel point and the pixel points in the neighborhood of the pixel point;
obtaining the target degree of each pixel point according to the gray value, the mixing degree and the number of the gray values in the pixel point field, wherein the number is nonzero;
acquiring a target pixel point according to the target degree of each pixel point of the difference image;
judging whether the difference image contains the connected domain or not by using whether the obtained target pixel point in the difference image can form the connected domain or not, if so, obtaining all the connected domains in the difference image by using the obtained target pixel point.
Further, the method for obtaining the target degree of each pixel point comprises the following steps:
the expression of the target degree is:
wherein:represent the firstTarget degree of each pixel point;represent the firstThe number of gray value non-zero pixel points in the periphery of each pixel point;represent the firstGray values of the individual pixels;represent the firstTarget degree of each pixel point;representing the normalization function.
Further, the method for obtaining the center line of each target connected domain in the target connected domain set includes:
obtaining the motion direction of the target connected domain by using the successfully matched target connected domain in each difference image;
and a straight line parallel to the movement direction is made through the center point of the target communicating region, and the straight line is taken as the center line of the target communicating region.
The beneficial effects of the invention are as follows: according to the invention, the target communicating domains which are matched with each other in the difference image are determined through the difference image of the video frame image and the standard fire-fighting channel, and whether the target communicating domains are automobiles or not is judged by utilizing the symmetry degree and the form value of the target communicating domains, so that whether the fire-fighting channel has an illegal automobile or not can be more accurately judged; effectively avoiding potential safety hazards caused by the fact that illegal vehicles cannot be accurately identified due to the fact that other moving objects exist in a background area, and the vehicles in a fire-fighting channel cannot be processed in time.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
FIG. 1 is a schematic flow chart of the method of the invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Examples
The invention as shown in fig. 1 provides a fire-fighting access breach detection method, comprising:
because the fire-fighting channel is fixed, the camera used for detecting the fire-fighting channel is also fixed, when the monitoring room is carried out on the pictures of the fire-fighting channel through the camera, all the pictures of the monitored fire-fighting channel (i.e. the pictures where the fire-fighting channel is located) are fixed, when the standard image of the fire-fighting channel is acquired, the pictures without any environmental influence on the fire-fighting channel are acquired as the standard image through the camera in the embodiment (i.e. only the fire-fighting channel and other fixed and immovable facilities exist in the pictures); the camera for shooting the fire-fighting channel can be arranged right in front of and behind the fire-fighting channel and can also be arranged on the side face of the fire-fighting channel, and in order to facilitate the technical scheme provided by the embodiment to be more intuitively known by the person in the technical field, the camera is arranged right in front of or right behind the fire-fighting channel, so that the image of the fire-fighting channel can be more intuitively acquired; meanwhile, the camera adopted by the embodiment is a high-definition camera, so that images at the fire-fighting channel can be accurately recorded; acquiring a standard image of a fire-fighting channel and a video of a fire-fighting channel area through the camera;
sequentially utilizing each frame of image in the video of the firefighting channel area to make a difference with a standard image to obtain a difference image, wherein the obtained difference image is obtained by adopting an algorithm commonly used in the prior art to obtain the difference image, and the specific method of the difference image is not described in detail;
the first aspect is whether the fire-fighting channel has automobiles or not, and the second aspect is that whether the automobiles in the fire-fighting channel stop for more than the time of the illegal stop, if the automobiles only stop on the fire-fighting channel temporarily, then the fire-fighting channel is opened, and the behavior is not considered to be illegal stop behavior occupying the fire-fighting channel; the embodiment detects the illegal situation of the fire fighting access based on the detection result;
in order to obtain the parking time of the vehicle in the fire-fighting channel, when the scheme adopted by the embodiment is adopted, whether a connected domain exists in a difference image obtained by taking a difference between a video shot by a camera and a standard image, when the connected domain exists, taking a frame image corresponding to the difference image as the start of a target frame image, starting timing, and acquiring all target videos of the target frame image in a set time; when all difference images of the video frame image and the standard image in the target video image have the condition that the connected domain does not exist (the condition that the objects entering the fire-fighting channel leave in the set time is explained), directly judging whether the fire-fighting channel is illicitly stopped or not is stopped; judging whether the connected domain exists in the newly acquired difference image of the video frame image and the standard image, taking the video frame image as a target frame image, re-acquiring the target image in a set time, judging whether the connected domain exists in the difference image or not as an automobile when the connected domain exists in all the difference images of the video frame image and the standard image in the target video image, and judging whether illegal stop occurs or not according to whether the connected domain exists in the automobile or not.
The time set in this embodiment is 5 minutes, that is, if the time for which the vehicle is stopped in the fire-fighting access exceeds 5 minutes, it is judged that it is illegal to stop, and if it does not exceed 5 minutes, it is not judged; the set time is set by the user; in this example, the distance was described with a set time of 5 minutes.
Specifically, a camera acquires a video frame image of each frame fire-fighting channel area, and makes a difference between the video frame image and a standard image to obtain a difference image, and judging whether the difference image contains a connected domain or not by using gray values of pixel points in the difference image and neighborhood pixel points of the pixel points;
since the gray value of a single pixel in the difference image cannot represent what, when the gray value of a pixel in the difference image (the gray value obtained by making the difference between the gray value of the pixel in the video frame image and the gray value of the pixel in the standard image) is relatively large, it cannot be said that the pixel is the target pixel, but it cannot be said that the pixel is not the target pixel; therefore, whether the pixel is a target pixel can be comprehensively judged according to some gray scale change conditions of the neighborhood around the pixel. The more the number of non-zero pixels in the surrounding neighborhood, the less the probability that the pixel is noise can be based on the number of non-zero pixels in the surrounding neighborhood. The more uniform the gray distribution, the greater the probability that this point is the target, in relation to the gray distribution of surrounding neighborhood pixel points.
In view of this, in this embodiment, the uniformity degree of each pixel point and its neighboring pixel points in the difference image is calculated first, and then whether the pixel point is a target pixel point is determined according to the uniformity degree of each pixel point and the gray value of each pixel point, so that all the target pixel points in the difference image can be extracted more accurately, specifically:
acquiring the uniformity degree of each pixel point according to the gray value of each pixel point in the difference image and the gray difference value of each pixel point and the pixel points in the neighborhood of the pixel point;
the uniformity of each pixel point is obtained as follows:
in the method, in the process of the invention,first, theGray level difference values between each pixel point and each pixel point in the neighborhood,represent the firstThe gray value of each pixel point,represent the firstThe number of the pixel points in the neighborhood of the pixel points;represent the firstUniformity of individual pixel points;
obtaining the target degree of each pixel point according to the gray value, the mixing degree and the number of the gray values in the pixel point field, wherein the number is nonzero;
the expression of the target degree is:
wherein:represent the firstTarget degree of each pixel point;represent the firstThe number of gray value non-zero pixel points in the periphery of each pixel point;represent the firstGray values of the individual pixels;represent the firstTarget degree of each pixel point;representing a normalization function;
after normalization treatment, makeThe range of the value of (2) is withinBetween them, the effect is to map the values toOver the interval.
Through the obtained target degree of each pixel pointComparing with the set threshold, the threshold set in the embodiment is 0.7, and when the target degree of the pixel is greater than the set threshold 0.7, the pixel is taken as the target pixel; and obtaining all the target pixel points in the difference image, and obtaining all the connected domains in the difference image according to the region formed by the obtained target pixel points.
When the difference image contains a connected domain, the difference image is used as an initial difference image; acquiring a video frame image corresponding to the initial difference image as an initial video frame image, traversing the difference image of the video frame image and the standard image in a set time after the initial video frame image is acquired; stopping traversing when the connected domain does not exist in the difference image of the video frame image and the standard image within the traversing setting time, and newly acquiring the next difference image containing the connected domain; taking the video frame image corresponding to the difference image as a new initial video frame image, traversing to obtain the difference image of the video frame image and the standard image in a set time after the new initial video frame image is obtained, and sequentially iterating;
stopping iteration when all difference images of all video frame images and standard images in the set time are traversed to have connected domains;
matching each connected domain in each difference image obtained in the set time, and taking the matched target connected domain in each difference image and other difference images as a target connected domain set;
in the method for obtaining all connected domains in the difference image, because the difference image is obtained by making a difference between a video frame image with a fire-fighting channel area and a standard image, a plurality of zero pixel points exist in the difference image, in theory, if no object exists in the fire-fighting channel, all the pixel points in the difference image are zero pixel points after making the difference, namely, the pixel points with zero gray scale, in order to accurately extract the connected domains in the difference image;
when a connected domain appears in a difference image obtained after the difference between the video frame image and the standard image (which indicates that an object enters a fire-fighting channel area), timing is started, the video frame image corresponding to the difference image is taken as an initial video frame image, and after the initial video frame image is obtained, the video frame image within 5 minutes after the initial video frame image is obtained between each frame image and the standard image, the difference image of each video frame image and the standard image is obtained at the same time when the video frame image within 5 minutes is obtained, and the connected domain of the difference image is obtained, if no connected domain appears in the difference image, the connected domain does not appear in the difference image of the video frame image and the standard image obtained at the 3 rd minute after the target frame image, then the object which enters the fire-fighting channel is considered to leave the fire-fighting channel, the object stays in the fire-fighting channel and does not belong to the fire-fighting channel, so that whether the object which does not belong to the fire-fighting channel area is illegal stop is judged, whether the object which enters the fire-fighting channel area is an automobile is judged, whether the connected domain appears in the new video frame image is obtained from the new video frame image or the new connected domain appears in the new set time when the new connected domain appears in the fire-fighting channel area; entering a new iteration process; when the connected domain always appears and does not disappear in the obtained video frame image within 5 minutes, the iteration is ended, and whether the connected domain is an automobile or not is judged.
In the embodiment, whether the connected domain in the difference image is an automobile connected domain is judged according to the following mode;
because the objects entering the fire-fighting passing area not only contain vehicles but also pedestrians or animals, when the objects exist in the difference image, no method exists for judging whether the fire-fighting channel is in illegal stop or not; therefore, it is necessary to judge the objects present in the fire-fighting access area, and these moving objects may have an influence on the judgment of the communication area of the automobile. The motion of the automobile and the motion of the pedestrian are greatly different, although the automobile and the human have symmetry, the symmetry degree of the pedestrian is reduced due to the swing of the hands and the legs when the pedestrian moves, the differential image obtained by a frame difference method when the vehicle moves is always rectangular, but the shape of the differential image of the pedestrian is relatively large.
Because there may be multiple connected domains in the difference images (two objects enter the fire-fighting channel region at the same time), when judging the connected domains, firstly matching the connected domains in all the difference images (the difference images obtained through 5 minutes of video), obtaining connected domains matched with other difference images in each difference image (the mutually matched connected domains are the same object), and then taking the successfully matched target connected domains as a target connected domain set (the obtained difference images are arranged according to the time of the video frame, and the successfully matched target connected domains in the obtained difference images are ordered according to the sequence of the difference images to obtain a target connected domain sequence); specifically, the corner detection is performed on the connected domains in each difference image to obtain the corner in each connected domain, then the obtained corner in each connected domain is matched with the corner in other connected domains, the successfully matched connected domains form a target connected domain set, and the corner detection and the corner matching method adopted in the embodiment are common means in the technical field and are not described in detail herein.
According to the change of the same corner point of each target communicating domain, which is detected by each target communicating domain in the target communicating domain set and the target communicating domain adjacent to the target communicating domain, the movement direction of the object corresponding to each target communicating domain; then, a straight line parallel to the movement direction is made through the center point of each target communicating domain, and the straight line is taken as the center line of the target communicating domain; dividing each target communication domain corresponding to the central line into a left half region and a right half region, symmetrically covering the left half region into a right half region image, or symmetrically covering the right half region into the left half region to obtain the overlapping area of the two regions, and obtaining the symmetry degree of each target region by utilizing the overlapping area of the left half region and the right half region divided by the target region and the total area of the target region;
the expression of the symmetry degree of each target area is:
wherein:representing the first in the target connected domain setThe degree of symmetry of the individual target areas;representing the first in the target connected domain setThe overlapping area of the left and right half areas of the target areas divided by the center line,representing the first in the target connected domain setThe total area of the target areas, when an image is completely symmetrical, the area of the overlapped part and the total area contrast should approach to 0.5, when the total area of the image is unchanged, the bigger the area of the overlapped part, the higher the symmetry degree of the object, the smaller the area of the overlapped part, the lower the symmetry degree of the object, and the two are in proportion.
Meanwhile, convex hull detection is carried out on each target area in the target connected domain set, the area of a convex hull area of each target area after convex hull detection is obtained, a convex hull algorithm-Graham algorithm commonly used at present is adopted when the convex hull detection is carried out on each target area, and the method for detecting the convex hull of the target area is not described in detail.
Obtaining a morphological value of the target connected domain by combining the total area of the target region after obtaining the convex hull area of the target region in the target connected domain set;
the morphological value of each target connected domain is expressed as:
wherein:representing the first in the target connected domain setMorphology values for the individual target areas;representing the first in the target connected domain setTotal area of the individual target areas;representing the first in the target connected domain setConvex hull areas of the target areas; the area ratio of the object area to the convex hull area can reflect the shape of the object area, and the larger the area ratio is, the more the object approaches to a rectangle, and the more regular the shape of the object is. The shape of the object can be judged by the area ratio.
Because the time intervals among the frames are equal in the acquired video frame images, when an object is detected to enter the fire-fighting access area, the target area in each frame is changed, and the probability that the target area is a vehicle can be obtained by the change condition of the symmetry degree of the target area and the deformation degree of the target area within a period of time; the symmetry degree of the automobile is not changed greatly in the running process, and the deformation degree of the automobile is also very small. The symmetry degree of the automobile is very high, but the symmetry degree of the pedestrian is not low, but the symmetry degree and the shape change of the pedestrian are relatively severe due to the swing of the hands and the legs during the movement of the pedestrian. Whether the area is a vehicle or not can be judged according to the change condition of the symmetry degree and the deformation degree. The more intense the change in degree of symmetry, the less likely the vehicle is to be targeted, and the greater the degree of deformation, the less likely the target area is to be a vehicle, inversely proportional. The higher the symmetry degree of the target area is, the larger the probability that the area is a vehicle is, so that whether the target area is an automobile is obtained, the standard deviation of all the forms of the target in the period of time is calculated, the change distribution of the target form can be seen, if the form change of the target area is not obvious, the standard deviation is small, the target form distribution is concentrated, and the two forms are in direct proportion. Calculating the range of the symmetry degree in the period of time to represent the change condition of the symmetry degree, wherein the larger the range is, the more severe the change is, and the smaller the range is, the more gentle the change degree is; for this reason, the present embodiment uses the symmetry degree and the morphology value of all the target connected domains to determine whether the target connected domain is an automobile.
Specifically, the probability that the target connected domain is an automobile is utilized by the symmetry degree and the morphological value of all the target connected domains in the target connected domain set;
the expression of the probability that the target connected domain in the target connected domain set is an automobile is:
wherein:a probability value indicating that the target connected domain in the target connected domain set is an automobile;the average value of symmetry degrees of the target connected domains in the target connected domain set;the maximum value of symmetry degree of the target connected domain in the target connected domain set;the minimum value of the degree of the calling of the target connected domain in the target connected domain set;the first of the target connected domain setsMorphology values of the individual target connected domains;the mean value of the morphological values of the target connected domains in the target connected domain set;representing the number of the target connected domains in the target connected domain set;representing a normalization function;representing an exponential function;representing the deformation degree of a target region in the target connected region set in the motion process;
by passing throughThe normalization algorithm is carried out, and the value range is withinBetween them, the effect is to map the values toOn the interval; when the probability that the target connected domain is an automobile is greater than the threshold, the threshold is set to be 0.75 in the embodiment, at the moment, the object entering the fire-fighting access area is considered to be an automobile, and meanwhile, the time that the automobile stays in the fire-fighting access area exceeds the set time for 5 minutes, so that the condition that the fire-fighting access is out of stop is judged.
In order to facilitate the person skilled in the art to more clearly understand how to perform distance description on whether the target connected domain is an automobile in the embodiment; making a difference between a video image in a set time in the video and a standard fire-fighting access area image to obtain a difference image of each frame; detecting connected domains of the difference images of each frame to obtain the connected domains in each difference image, wherein 10 difference images are adopted to carry out distance description in the embodiment; detecting a plurality of connected domains in each difference image, and then respectively matching the connected domains in each difference image with the connected domains in other difference images to obtain a matching result, wherein two target connected domain sets are obtained, namely target connected domain sets A (A1, A2, A3, A4, A5, A6, A7, A8, A9 and A10); a set of target connected domains B (B1, B2, B3, B4, B5, B6, B7, B8, B9, B10); calculating the symmetry degree and the morphology value of each connected domain in the target connected domain set A and the target connected domain set B respectively, and utilizing the symmetry degree and the morphology value of each target connected domain in the target connected domain set A to obtain the probability value that the target connected domain in the target connected domain set B is an automobile (because the matching successfully indicates that all connected domains are one object); the symmetry degree and the shape value of each target connected domain in the target connected domain set B are equal to the probability value of the automobile in the target connected domain set B, and then whether the object corresponding to the target connected domain set B and the object corresponding to the target connected domain set B are illegal vehicles or not is judged according to the probability value.
Meanwhile, when judging that the automobile is out of stop in the fire-fighting access area, the alarm can be sent out to the property or the residence in the fire-fighting access area to inform the fire-fighting access area of the out-of-stop information, and the worker is dispatched at the first time by the property or the residence to process the out-of-stop information, so that the fire-fighting access is prevented from being occupied for a long time to cause great potential safety hazards.
Meanwhile, the embodiment also provides a fire-fighting channel illegal stop alarm system, which comprises a camera arranged in the fire-fighting channel area and used for detecting the fire-fighting channel area; a GPS, a client and a processor for locating the message channel area; the method comprises the steps that a camera sends a video frame image of a shot fire fighting area to a processor, the processor receives a difference value image of each frame image and a fire fighting area standard image, detects connected domains in the difference value image, when the connected domains exist in the difference value image (an object is stated to enter a fire fighting channel area), the processor takes the video frame image corresponding to the difference value image of the obtained connected domains as a video frame image for starting timing, counts whether the connected domains in the difference value image exist all the time in a next counted time period, if not, then the connected domains exist all the time in a specified time period after the next obtained difference value image exists in the connected domains, sequentially iterates, and when the connected domains exist all the time in the specified time period, a connected domain set formed by the target connected domains is obtained in the specified time period, and is matched, the degree of symmetry and the form value of each target connected domain in the target connected domain set are utilized to calculate and judge whether the connected domains enter the fire fighting channel area as a first time period, if the vehicle is stopped in the fire fighting channel area is stopped, the first time period is stopped, and the fire fighting information is stopped by a client is stopped, and the fire fighting information is stopped in the fire fighting channel is stopped.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (5)
1. A fire channel breach detection method, comprising:
acquiring a standard image of a fire fighting access area and a video frame image in a video of the fire fighting access area;
sequentially obtaining difference images of the video frame image and the standard image, and judging whether the difference images contain connected domains or not by using gray values of pixel points in the difference images and neighbor pixel points of the pixel points;
when the difference image contains a connected domain, the difference image is used as an initial difference image; acquiring a video frame image corresponding to the initial difference image as an initial video frame image, traversing the difference image of the video frame image and the standard image in a set time after the initial video frame image is acquired; stopping traversing when the connected domain does not exist in the difference image of the video frame image and the standard image within the traversing setting time, and re-acquiring the next difference image containing the connected domain; taking the video frame image corresponding to the difference image as a new initial video frame image, traversing to obtain the difference image of the video frame image and the standard image in a set time after the new initial video frame image is obtained, and sequentially iterating;
stopping iteration when all difference images of all video frame images and standard images in the traversing set time have connected domains;
matching each connected domain in each difference image obtained in the set time, and taking the matched target connected domain in each difference image and other difference images as a target connected domain set;
acquiring the central line of each target communicating domain in the target communicating domain set, and acquiring the symmetry degree of the target communicating domain by utilizing the regions on two sides of the central line of each target communicating domain;
acquiring a morphological value of each target connected domain in the target connected domain set by utilizing the area of each target connected domain and the convex hull area of the target connected domain;
judging whether the target connected domain is an automobile or not by using the symmetry degree and the morphological value of all the target connected domains in the target connected domain set;
when the target connected domain is judged to be an automobile, illegal stopping occurs in the fire-fighting access area;
the method for judging whether the target connected domain is an automobile or not by utilizing the symmetry degree and the morphological value of all the target connected domains in the target connected domain set comprises the following steps: utilizing the symmetry degree and the morphology value of all the target connected domains in the target connected domain set to obtain the probability that the target connected domain is an automobile;
comparing the probability of the obtained target connected domain being the automobile with a set threshold value to judge whether the target connected domain is the automobile or not;
the expression of the probability of obtaining the target connected domain as the automobile is as follows:
wherein:a probability value indicating that the target connected domain is an automobile; />The average value of symmetry degrees of all the target connected domains in the target connected domain set; />The maximum value of the symmetry degree of the target connected domain in the target connected domain set; />The minimum value of the symmetry degree of the target connected domain in the target connected domain set; />Is the +.>Morphology values of the individual target connected domains; />The method is an average value of morphological values of all the target connected domains in the target connected domain set; />Representing the number of the target connected domains in the target connected domain set; />Representing a normalization function; />Representing an exponential function.
2. The fire passage stopping detection method according to claim 1, wherein the method for obtaining the symmetry degree of the target communicating region by using the regions on both sides of the center line of the target communicating region comprises:
dividing the target communication domain through the center line to obtain the overlapping areas of the areas on the two sides of the center line after division;
and obtaining the symmetry degree of the target communicating region according to the superposition area of the areas at the two sides of the central line and the total area of the target communicating region.
3. The fire control channel stop violation detection method according to claim 1, wherein the method for judging whether the difference image contains the connected domain or not by using the gray values of the pixel point in the difference image and the neighborhood pixel point of the pixel point comprises the following steps:
acquiring the uniformity degree of each pixel point according to the gray value of each pixel point in the difference image and the gray difference value of each pixel point and the pixel points in the neighborhood of the pixel point;
obtaining the target degree of each pixel point according to the gray value and the uniformity degree of each pixel point and the number of the gray values in the neighborhood of the pixel point, wherein the gray values are non-zero;
acquiring a target pixel point according to the target degree of each pixel point of the difference image;
judging whether the difference image contains the connected domain or not by using whether the obtained target pixel point in the difference image can form the connected domain or not, if so, obtaining all the connected domains in the difference image by using the obtained target pixel point.
4. The fire fighting access stop detection method according to claim 3, wherein the method for obtaining the target degree of each pixel point is as follows:
the expression of the target degree is:
wherein:indicate->Target degree of each pixel point; />Indicate->The number of gray value non-zero pixel points in the periphery of each pixel point; />Indicate->Gray values of the individual pixels; />Indicate->Uniformity of individual pixel points; />Representing the normalization function.
5. The fire fighting access breach detection method of claim 1, wherein the method of obtaining a center line of each of the set of target connected domains includes:
obtaining the motion direction of the target connected domain by using the successfully matched target connected domain in each difference image;
and a straight line parallel to the movement direction is made through the center point of the target communicating region, and the straight line is taken as the center line of the target communicating region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310579431.7A CN116309592B (en) | 2023-05-23 | 2023-05-23 | Fire control channel illegal stop detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310579431.7A CN116309592B (en) | 2023-05-23 | 2023-05-23 | Fire control channel illegal stop detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116309592A CN116309592A (en) | 2023-06-23 |
| CN116309592B true CN116309592B (en) | 2023-08-11 |
Family
ID=86799996
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310579431.7A Active CN116309592B (en) | 2023-05-23 | 2023-05-23 | Fire control channel illegal stop detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116309592B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036490A (en) * | 2014-05-13 | 2014-09-10 | 重庆大学 | Foreground segmentation method applied to mobile communication network transmission |
| CN110197479A (en) * | 2019-05-23 | 2019-09-03 | 中国科学院力学研究所 | A kind of modified quality determining method in surface |
| CN115995091A (en) * | 2023-02-09 | 2023-04-21 | 清华大学 | Method and device for reading flow chart, electronic equipment and storage medium |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109901567B (en) * | 2017-12-08 | 2020-06-23 | 百度在线网络技术（北京）有限公司 | Method and apparatus for outputting obstacle information |
-
2023
- 2023-05-23 CN CN202310579431.7A patent/CN116309592B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036490A (en) * | 2014-05-13 | 2014-09-10 | 重庆大学 | Foreground segmentation method applied to mobile communication network transmission |
| CN110197479A (en) * | 2019-05-23 | 2019-09-03 | 中国科学院力学研究所 | A kind of modified quality determining method in surface |
| CN115995091A (en) * | 2023-02-09 | 2023-04-21 | 清华大学 | Method and device for reading flow chart, electronic equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 基于毫米波雷达与机器视觉信息融合的道路车辆检测与跟踪算法研究;张翔;中国优秀硕士学位论文全文数据库工程科技II辑(第04期);第C035-314页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116309592A (en) | 2023-06-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10212397B2 (en) | Abandoned object detection apparatus and method and system | |
| US9008359B2 (en) | Detection of static object on thoroughfare crossings | |
| CN110738150B (en) | Camera linkage snapshot method and device and computer storage medium | |
| CN110867083B (en) | Vehicle monitoring method, device, server and machine-readable storage medium | |
| WO2019089441A1 (en) | Exclusion zone in video analytics | |
| CN112367475B (en) | Traffic incident detection method and system and electronic equipment | |
| CN110718041B (en) | Method, device and system for preventing children from getting lost and storage medium | |
| CN104239847B (en) | Driving warning method and electronic device for vehicle | |
| WO2021008039A1 (en) | Systems and methods for object monitoring | |
| CN106560861A (en) | Intelligent supervision method based on computer vision | |
| KR20160037480A (en) | Method for establishing region of interest in intelligent video analytics and video analysis apparatus using the same | |
| CN101877132B (en) | Interactive event processing method and device used for motion tracking | |
| CN116309592B (en) | Fire control channel illegal stop detection method | |
| KR101522757B1 (en) | Method for removing noise of image | |
| CN116434161B (en) | Method and system for judging whether parking behavior based on high-order video is credible | |
| CN110913209B (en) | Camera shielding detection method and device, electronic equipment and monitoring system | |
| CN112800846A (en) | High-altitude parabolic monitoring method and device, electronic equipment and storage medium | |
| CN110008957B (en) | District illegal stop warning method based on license plate recognition and image variance algorithm | |
| CN111931745A (en) | Vehicle detection method and device, electronic equipment and storage medium | |
| CN114699702B (en) | Fire fighting equipment detection method and related device | |
| KR101407394B1 (en) | System for abandoned and stolen object detection | |
| CN107609498B (en) | Data processing method of computer monitoring system | |
| CN113378728A (en) | Monitoring method, system, electronic equipment and computer readable storage medium | |
| CN114399537B (en) | Vehicle tracking method and system for target personnel | |
| CN113112814B (en) | Snapshot method and device without stopping right turn and computer storage medium |
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 |