CN116434091A - Building outer wall skin anti-drop monitoring method, medium and system - Google Patents
Building outer wall skin anti-drop monitoring method, medium and system Download PDFInfo
- Publication number
- CN116434091A CN116434091A CN202310545900.3A CN202310545900A CN116434091A CN 116434091 A CN116434091 A CN 116434091A CN 202310545900 A CN202310545900 A CN 202310545900A CN 116434091 A CN116434091 A CN 116434091A
- Authority
- CN
- China
- Prior art keywords
- falling
- wall skin
- skin
- image
- building
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 108
- 238000000034 method Methods 0.000 title claims abstract description 33
- 230000008859 change Effects 0.000 claims abstract description 40
- 239000011159 matrix material Substances 0.000 claims abstract description 39
- 230000002265 prevention Effects 0.000 claims abstract description 8
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 239000003550 marker Substances 0.000 claims description 42
- 238000012549 training Methods 0.000 claims description 42
- 239000013598 vector Substances 0.000 claims description 20
- 238000002474 experimental method Methods 0.000 claims description 7
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 238000012937 correction Methods 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 4
- 238000007664 blowing Methods 0.000 claims description 3
- 238000004062 sedimentation Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 230000009286 beneficial effect Effects 0.000 description 4
- 239000004570 mortar (masonry) Substances 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 230000006872 improvement Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 2
- 238000003491 array Methods 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 239000000853 adhesive Substances 0.000 description 1
- 230000001070 adhesive effect Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000008602 contraction Effects 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000006073 displacement reaction Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 229910052742 iron Inorganic materials 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000000873 masking effect Effects 0.000 description 1
- 239000003973 paint Substances 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/17—Terrestrial scenes taken from planes or by drones
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
- G06V10/225—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition based on a marking or identifier characterising the area
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/176—Urban or other man-made structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Remote Sensing (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Image Processing (AREA)
- Working Measures On Existing Buildindgs (AREA)
Abstract
The invention provides a method, a medium and a system for monitoring the falling-off prevention of an outer wall skin of a building, which belong to the technical field of building safety and comprise the following steps: acquiring an outer wall skin image containing an outer wall skin monitoring mark array of a building, wherein the outer wall skin monitoring mark array is formed by arranging thin markers attached or coated on the outer wall skin of the building in an array manner; preprocessing an outer wall skin image and extracting features to obtain a monitoring mark array image; comparing the monitoring mark array image with the standard monitoring mark array image to obtain a change matrix of the monitoring mark array; calculating a change matrix by using a pre-trained exterior wall skin falling model to obtain an exterior wall skin falling area and the remaining time of wall skin falling; and sending the obtained falling area of the outer wall skin and the residual falling time of the wall skin to related personnel. The method can realize early warning of the falling-off condition of the building outer wall skin, and effectively realize early prevention and treatment of the falling-off of the building outer wall skin.
Description
Technical Field
The invention belongs to the technical field of building safety, and particularly relates to a building outer wall skin anti-falling monitoring method, medium and system.
Background
The falling of the building outer wall skin, especially the falling of the high-rise building outer wall skin, can cause great threat to pedestrians, and in recent years, accidents of injury caused by the falling of the building outer wall occur too much. There are many reasons for the peeling of the building exterior wall skin, the following are some common reasons:
1. construction quality problems: in the construction process, the base layer of the wall body is not treated in place, so that the outer wall skin is not firmly bonded with the wall body and is easy to fall off. For example, the base layer is not cleaned up, the thickness of the spread mortar is not uniform, etc.
2. Material quality problems: the quality of the used outer wall skin material is unqualified, such as mortar, paint, putty and the like, and the outer wall skin can be possibly fallen off.
3. The construction process problems are as follows: in the construction process, the construction process of the outer wall skin is not standard, for example, the time for coating mortar is too long, so that the adhesive force of the dried mortar is reduced, and the outer wall Pi Yi falls off.
4. Environmental factors: the environmental factors of the building may also cause the exterior skin to fall off. Such as climate change, temperature, humidity, etc., may cause the exterior wall skin of the building to fall off.
5. Design problem: the design of the building is unreasonable, so that the outer wall bears excessive wind pressure or expansion with heat and contraction with cold and the like, and the outer wall skin is separated.
6. Repair lost over time: the service life of the building is longer, the outer wall skin is aged, the compression resistance and the tensile resistance are reduced, and the outer wall skin is easy to fall off.
7. Human factors: in the use process of the building, the building is impacted and extruded by external force, and the outer wall skin can be possibly separated.
In the prior art, a method for early warning the falling-off condition of the building outer wall skin is lacked, and the early prevention and treatment of the falling-off condition of the building outer wall skin cannot be effectively realized.
Disclosure of Invention
In view of the above, the invention provides a method, medium and system for monitoring the falling off of building outer wall skin, which can realize early warning of the falling off condition of the building outer wall skin and effectively realize the early prevention and treatment of the falling off of the building outer wall skin.
The invention is realized in the following way:
the first aspect of the invention provides a building exterior wall skin anti-falling monitoring method, which comprises the following steps:
s10, acquiring an outer wall skin image containing a building outer wall skin monitoring mark array, wherein the building outer wall skin monitoring mark array is formed by arranging thin markers attached or coated on building outer wall skins in an array manner;
s20, preprocessing the outer wall skin image and extracting features to obtain a monitoring mark array image;
s30, comparing the monitoring mark array image with a standard monitoring mark array image to obtain a change matrix of the monitoring mark array;
s40, calculating the change matrix by using a pre-trained exterior wall skin falling model to obtain an exterior wall skin falling area and a wall skin falling residual time;
and S50, sending the obtained falling area of the outer wall skin and the remaining falling time of the outer wall skin to related personnel for early warning or treatment.
The thin marker can be a light and thin infrared reflecting film, an IR reflecting film or a marker directly marked by adopting a color different from that of the outer wall skin.
On the basis of the technical scheme, the anti-falling monitoring method for the building outer wall skin can be further improved as follows:
in the step of acquiring the outer wall skin image comprising the building outer wall skin monitoring mark array, the acquiring device of the outer wall skin image is a camera, the camera is arranged at the bottom of the building, the camera can cover all outer wall skins on one side of the building, and the camera is a plurality of all outer wall skins for covering each side of the building; the camera is a camera with the resolution ratio of more than 4K.
The camera is arranged at the lower part of the building and is 2-3 meters away from the ground, so that the situation that people are injured due to high falling impact of the camera caused by ageing of equipment for installing the camera can be prevented; the iron stand can be arranged on an external wall of a building, and can also be arranged on an external vertical rod, such as a telegraph pole, a street lamp pole and the like.
Further, the step of preprocessing the exterior skin image and extracting features to obtain a monitoring mark array image specifically includes:
performing trapezoidal correction, denoising, histogram equalization, contrast enhancement and gray scale processing on the outer wall skin image to obtain a first image;
dividing the first image into different divided areas, wherein each divided area contains a thin marker;
marking the thin marker of each segmented region as a marked image;
and extracting the region corresponding to the mark image in the first image, removing other background regions and monitoring the mark array image.
The step of dividing the first image into different areas can be realized by threshold segmentation, edge detection, area growth and other methods;
the step of marking the thin marker of each divided area can be realized by using methods such as connected domain analysis, shape descriptors, key point detection and the like;
the step of removing other background areas can be realized by methods such as image masking, area extraction and the like.
Further, the standard monitoring mark array image obtaining step comprises the following steps:
selecting a first image shot by a camera containing all the thin markers as a basic exterior skin image;
taking the midpoint of the bottom edge and the plane of the top edge of the building single-sided outer wall as a basic plane according to the basic outer wall skin image, and taking the projection of the building single-sided outer wall on the basic plane as a standard wall surface;
projecting the monitoring mark array image on the basic plane to obtain a projection image;
setting a plurality of mark areas on the projection image by taking the center point of each thin mark on the projection image as the center;
and according to the relative positions of the camera and the standard wall surface, an observation point is established, and all marked areas on the standard wall surface observed by the observation point are obtained and used as standard monitoring mark array images.
Further, the step of obtaining the change matrix of the monitoring mark array includes:
establishing a basic coordinate system on the basic plane, wherein the origin of the basic coordinate system is the lower left point of a standard wall surface, the horizontal direction of the basic plane is used as a horizontal axis, the vertical axis is used as a vertical axis, the standard vector of each marking area is marked as X= [ Xa, xb, xc, xd ], xa represents the abscissa of the marking area, xb represents the ordinate of the marking area, xc represents the perimeter of the marking area, and Xd represents the area of the marking area;
establishing a monitoring vector for each thin marker on the projection image, wherein the monitoring vector is marked as Y= [ Ya, yb, yc and Yd ], wherein Ya represents the abscissa of the thin marker on the projection image, yb represents the ordinate of the thin marker on the projection image, yc represents the perimeter of the thin marker on the projection image, and Yd represents the area of the thin marker on the projection image;
calculating a standard vector of each marking area, and calculating a difference value of a monitoring vector of the thin marker on the corresponding projection image, wherein Dxy=X-Y;
and establishing a change matrix of Dxy corresponding to all the marked areas.
Further, the step of calculating the change matrix by using a pre-trained exterior wall skin falling model to obtain an exterior wall skin falling area and a wall skin falling residual time specifically comprises the following steps:
building a training set: the training set comprises more than 100 groups of experimental data, wherein the experiment is a small-scene experiment, and the specific scene is that an experimental single-sided wall with the height of 3 meters and the width of 2 meters is established, a thin marker is attached or coated on the experimental single-sided wall to form an experimental building outer wall skin monitoring mark array, an experimental standard monitoring mark array image is obtained, and the experimental single-sided wall is subjected to simulated sedimentation, simulated shaking and simulated blowing with multiple grades and multiple angles until the experimental single-sided wall is subjected to large-area outer wall skin falling, and the large-area outer wall skin falling is more than 3 square meters;
recording an experimental building outer wall skin monitoring mark array as a training array set every 1 second;
taking the moment of each falling of the outer wall skin as a splitting moment, recording the area proportion of the corresponding falling wall skin, and splitting the training array set into a plurality of training array subsets;
in each training array subset, calculating a change matrix of each experimental building exterior wall skin monitoring mark array and each experimental building exterior wall skin monitoring mark array, marking the change matrix as an experimental change matrix, and recording a time interval between a moment corresponding to the experimental change matrix and the last element in the training array subset as the falling residual time of the experimental wall skin;
the training input of the training set is an experimental change matrix corresponding to each experimental building exterior wall skin monitoring mark array in the training array set; the output of the training set is the area proportion of the falling wall skin corresponding to the experimental building outer wall skin monitoring mark array and the falling residual time of the experimental wall skin;
modeling and training: establishing an outer wall skin falling model prototype by using a convolutional neural network, and training by using the training set to obtain an outer wall skin falling model;
using a trained model: and calculating the change matrix by using an outer wall skin falling model to obtain corresponding falling wall skin area proportion and wall skin falling residual time, and calculating the falling area according to the falling wall skin area proportion and the total area of the outer wall skin.
The beneficial effects of adopting above-mentioned improvement scheme are: because actual outer wall skin data are difficult to collect, an experimental mode is adopted, a large amount of actual wall skin data are not required to be collected, an outer wall skin model is built by using the experimental data, the efficiency is higher, and the danger of actual data collection is avoided.
Further, the thin tag is rectangular.
The beneficial effects of adopting above-mentioned improvement scheme are: the rectangular mark is used, and the image shot by the camera has an edge angle, so that better identification degree can be achieved compared with a round shape or an oval shape.
Further, in the step of selecting the basic exterior wall skin image shot by the camera containing all the thin markers, if the thin markers are blocked in the exterior wall skin image shot by the camera, the unmanned aerial vehicle is adopted to orthogonally shoot each blocked thin marker of the exterior wall skin of the building, the image of the blocked thin markers is obtained, and the corresponding positions of the basic exterior wall skin image are marked according to the corresponding sizes of the thin markers.
The beneficial effects of adopting above-mentioned improvement scheme are: the camera arranged at the bottom of the building is only used for monitoring the building outer wall skin, the visual angle of the thin marker is smaller and smaller along with the increase of the height due to the fact that the building is higher, such as a high-rise residential building, and meanwhile, the size of the thin marker in the visual field of the camera is smaller and smaller due to the limitation of resolution, so that the detection is not facilitated. Moreover, because the building outer wall skin is uneven, shielding of a plurality of thin markers can be generated, and therefore, the inspection shooting unmanned aerial vehicle is utilized to shoot the building outer wall skin, and the images of all monitoring mark arrays can be effectively obtained.
A second aspect of the present invention provides a computer readable storage medium having stored therein program instructions for executing the above-described building exterior skin anti-drop monitoring method when the program instructions are run.
A third aspect of the present invention provides a building exterior skin anti-falling monitoring system, comprising the computer readable storage medium.
Compared with the prior art, the anti-falling monitoring method, medium and system for the building outer wall skin provided by the invention have the beneficial effects that: according to the invention, the camera is used for monitoring the single-sided outer wall skin of the building, the residual falling time and falling area of the outer wall skin of the building are calculated according to the monitored image and are sent to related personnel for processing, and early warning or processing on the falling situation of the outer wall skin of the building can be effectively realized; on the other hand, because the actual outer wall skin falling data are difficult to collect, a large amount of actual wall skin falling data are not required to be collected in an experimental mode, an outer wall skin falling model is built by using the experimental data, the efficiency is higher, and the danger of actual data collection is avoided; further, due to the high construction and resolution limitations, the thin markers are relatively deformed in the field of view of the camera, and are of a smaller size, which is not conducive to detection. Moreover, because the building outer wall skin is uneven, shielding of a plurality of thin markers can be generated, and therefore, the inspection shooting unmanned aerial vehicle is utilized to shoot the building outer wall skin, and the images of all monitoring mark arrays can be effectively obtained.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments of the present invention will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for monitoring the falling-off prevention of an outer wall skin of a building.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, based on the embodiments of the invention, which are apparent to those of ordinary skill in the art without inventive faculty, are intended to be within the scope of the invention.
Thus, the following detailed description of the embodiments of the invention, as presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, based on the embodiments of the invention, which are apparent to those of ordinary skill in the art without inventive faculty, are intended to be within the scope of the invention.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures.
In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc. indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings are merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the apparatus or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include one or more such feature. In the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
As shown in fig. 1, the first aspect of the present invention provides a flow chart of a method for monitoring anti-falling of building exterior wall skin, which comprises the following steps:
s10, acquiring an outer wall skin image containing a building outer wall skin monitoring mark array, wherein the building outer wall skin monitoring mark array is formed by arranging thin markers attached or coated on building outer wall skins in an array manner;
s20, preprocessing an outer wall skin image and extracting features to obtain a monitoring mark array image;
s30, comparing the monitoring mark array image with the standard monitoring mark array image to obtain a change matrix of the monitoring mark array;
s40, calculating a change matrix by using a pre-trained exterior wall skin falling model to obtain an exterior wall skin falling area and a wall skin falling residual time;
and S50, sending the obtained falling area of the outer wall skin and the remaining falling time of the outer wall skin to related personnel for early warning or treatment.
The step of sending the obtained outer wall skin falling area and the wall skin falling residual time to related personnel, or directly sending the obtained outer wall skin falling area and the wall skin falling residual time to related systems, wherein the related personnel are general property personnel or building operation and maintenance constructors; the relevant system is typically a building alarm system or directly an alarm.
In the above technical scheme, in the step of acquiring the exterior skin image including the building exterior skin monitoring mark array, the exterior skin image acquiring device is a camera, the camera is arranged at the bottom of the building, the camera can cover all exterior skins of one side of the building, and the camera is a plurality of all exterior skins for covering each side of the building; the camera is a camera with resolution ratio of more than 4K.
Further, in the above technical solution, the steps of preprocessing the exterior wall skin image and extracting the features to obtain the monitoring mark array image specifically include:
performing trapezoidal correction, denoising, histogram equalization, contrast enhancement and gray scale treatment on the outer wall skin image to obtain a first image;
dividing the first image into different divided areas, wherein each divided area contains a thin marker;
marking the thin marker of each divided area as a marked image;
and extracting the region corresponding to the mark image in the first image, removing other background regions and monitoring the mark array image.
Because the camera is arranged at the bottom of the building, the image obtained by shooting the single-sided wall of the whole building has a certain trapezoidal deformation, the outer wall image is restored to be rectangular by adopting a trapezoidal correction mode, the condition that the single-sided wall of the building is rectangular is generally considered, if the single-sided wall of the building is arc-shaped, a projection judgment mode is needed to be added, and the arc-shaped single-sided wall is projected to be a rectangular single-sided wall;
further, in the above technical solution, the standard monitoring flag array image acquiring step includes:
selecting a first image shot by a camera containing all the thin markers as a basic exterior skin image;
taking the midpoint of the bottom edge and the plane of the top edge of the building single-sided outer wall as a basic plane according to the basic outer wall skin image, and taking the projection of the building single-sided outer wall on the basic plane as a standard wall surface;
projecting the monitoring mark array image on a basic plane to obtain a projection image;
setting a plurality of mark areas on the projection image with the center point of each thin mark on the projection image as the center;
and according to the relative positions of the camera and the standard wall surface, establishing an observation point, and acquiring all marked areas on the standard wall surface observed by the observation point as standard monitoring mark array images.
Further, in the above technical solution, the step of obtaining the change matrix of the monitoring flag array includes:
a basic coordinate system is established on a basic plane, an origin of the basic coordinate system is a lower left point of a standard wall surface, a horizontal direction of the basic plane is used as a horizontal axis, a vertical axis of the basic plane is used as a vertical axis, standard vectors of each marking area are marked as X= [ Xa, xb, xc and Xd ], wherein Xa represents an abscissa of the marking area, xb represents an ordinate of the marking area, xc represents a perimeter of the marking area;
establishing a monitoring vector for each thin marker on the projection image, wherein the monitoring vector is marked as Y= [ Ya, yb, yc and Yd ], wherein Ya represents the abscissa of the thin marker on the projection image, yb represents the ordinate of the thin marker on the projection image, yc represents the circumference of the thin marker on the projection image, and Yd represents the area of the thin marker on the projection image;
calculating a standard vector of each marking area, and calculating a difference value of the monitoring vector of the thin marker on the corresponding projection image, wherein Dxy=X-Y;
and establishing a change matrix of Dxy corresponding to all the marked areas.
Further, in the above technical solution, the step of calculating the change matrix by using a pre-trained exterior wall skin falling model to obtain the exterior wall skin falling area and the remaining time of wall skin falling specifically includes:
building a training set: the training set comprises more than 100 groups of experimental data, and the experiment is a small-scene experiment, and the specific scene is that an experimental single-sided wall with the height of 3 meters and the width of 2 meters is established, a thin marker is attached or coated on the experimental single-sided wall to form an experimental building outer wall skin monitoring mark array, an experimental standard monitoring mark array image is obtained, and the experimental single-sided wall is subjected to simulated sedimentation, simulated shaking and simulated blowing at a plurality of grades and a plurality of angles until the experimental single-sided wall is subjected to large-area outer wall skin falling, and the large-area outer wall skin falling is the wall skin falling off more than 3 square meters;
recording an experimental building outer wall skin monitoring mark array as a training array set every 1 second;
taking the moment of each falling of the outer wall skin as a splitting moment, recording the area proportion of the corresponding falling wall skin, and splitting the training array set into a plurality of training array subsets;
in each training array subset, calculating a change matrix of each experimental building exterior wall skin monitoring mark array and each experimental building exterior wall skin monitoring mark array, marking the change matrix as an experimental change matrix, and recording a time interval between a moment corresponding to the experimental change matrix and the last element in the training array subset as the residual time of falling off of the experimental wall skin;
the training input of the training set is an experimental change matrix corresponding to each experimental building exterior skin monitoring mark array in the training array set; the output of the training set is the area proportion of the falling wall skin corresponding to the building outer wall skin monitoring mark array for experiments and the falling residual time of the experimental wall skin;
modeling and training: establishing an outer wall skin falling model prototype by using a convolutional neural network, and training by using a training set to obtain an outer wall skin falling model;
using a trained model: and calculating the change matrix by using the outer wall skin falling model to obtain the corresponding falling wall skin area proportion and the wall skin falling residual time, and calculating the falling area according to the falling wall skin area proportion and the total area of the outer wall skin.
The convolutional neural network comprises an input layer, a backbone layer, a full connection layer and an output layer, wherein the backbone layer is DenseNet121; the backbone network DenseNet121 is formed by sequentially stacking 1 Dense Block containing 6 convolution layers, 1 Dense Block containing 12 convolution layers, 1 Dense Block containing 24 convolution layers and 1 Dense Block containing 16 convolution layers; the input of each convolution layer in the Dense Block is the output of all previous convolution layers; the backbone layer is used for receiving input experimental change matrix extraction characteristics, and the full-connection layer is used for flattening output of the backbone layer into a vector;
the Output layer outputs the full connection layer through an activation function Softmax to obtain an Output result, wherein the Output result is a vector and is recorded as an Output vector output= [ S, T ], S represents the area proportion of the falling wall skin, and T represents the falling residual time corresponding to S.
Because the input is a change matrix, the output is a vector, and the general convolutional neural network can process the vector, the internal structure setting of the convolutional neural network is not needed to be paid much attention to.
Further, in the above technical solution, the thin tag is rectangular.
Further, in the above technical solution, in the step of selecting the basic exterior skin image captured by the camera including all the thin markers, if the thin markers are blocked in the exterior skin image captured by the camera, the unmanned aerial vehicle is adopted to orthogonally capture each blocked thin marker of the exterior skin of the building, obtain the image of the blocked thin marker, and mark the corresponding position of the basic exterior skin image according to the corresponding size of the thin marker.
A second aspect of the present invention provides a computer readable storage medium having stored therein program instructions for executing the above-described building exterior skin anti-drop monitoring method when the program instructions are run.
A third aspect of the present invention provides a building exterior skin anti-falling monitoring system, comprising the computer readable storage medium.
Specifically, the principle of the invention is as follows: the method is characterized in that the displacement is monitored, a thin marker is used for marking, when part of the outer wall skin is displaced, the corresponding thin marker is driven to displace, and after the camera is fixed at the bottom of the building and the thin marker is displaced, the center point, the area and the perimeter of the thin marker are changed in an image acquired by the camera due to the problem of visual angle; monitoring a building outer wall skin monitoring mark array in an experimental mode, simulating settlement, shaking and intervention of strong wind on an experimental single-sided wall, recording a monitoring mark array image immediately before each outer wall skin is dropped off, calculating a change matrix with a standard monitoring mark array image, establishing a correlation between the change matrix and the wall skin drop-off area ratio, establishing a model by using a neural network algorithm, and analyzing the correlation to obtain an outer wall skin drop-off model; and calculating the currently monitored exterior skin monitoring mark array image by using the exterior skin falling model to obtain the exterior skin falling area and the exterior skin falling residual time.
The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (10)
1. The method for monitoring the falling-off prevention of the building outer wall skin is characterized by comprising the following steps of:
s10, acquiring an outer wall skin image containing a building outer wall skin monitoring mark array, wherein the building outer wall skin monitoring mark array is formed by arranging thin markers attached or coated on building outer wall skins in an array manner;
s20, preprocessing the outer wall skin image and extracting features to obtain a monitoring mark array image;
s30, comparing the monitoring mark array image with a standard monitoring mark array image to obtain a change matrix of the monitoring mark array;
s40, calculating the change matrix by using a pre-trained exterior wall skin falling model to obtain an exterior wall skin falling area and a wall skin falling residual time;
and S50, sending the obtained falling area of the outer wall skin and the remaining falling time of the outer wall skin to related personnel for early warning or treatment.
2. The method for monitoring the falling off of building exterior skin according to claim 1, wherein in the step of acquiring exterior skin images comprising a building exterior skin monitoring mark array, the acquiring device of the exterior skin images is a camera, the camera is deployed at the bottom of a building, the camera can cover all exterior skins on one side of the building, and the camera is a plurality of all exterior skins for covering each side of the building; the camera is a camera with the resolution ratio of more than 4K.
3. The method for monitoring the falling off of the building exterior wall skin according to claim 2, wherein the steps of preprocessing the exterior wall skin image and extracting features to obtain a monitoring mark array image specifically comprise:
performing trapezoidal correction, denoising, histogram equalization, contrast enhancement and gray scale processing on the outer wall skin image to obtain a first image;
dividing the first image into different divided areas, wherein each divided area contains a thin marker;
marking the thin marker of each segmented region as a marked image;
and extracting the region corresponding to the mark image in the first image, removing other background regions and monitoring the mark array image.
4. The method for monitoring the falling off of the building exterior wall skin according to claim 3, wherein the step of obtaining the standard monitoring mark array image is as follows:
selecting a first image shot by a camera containing all the thin markers as a basic exterior skin image;
taking the midpoint of the bottom edge and the plane of the top edge of the building single-sided outer wall as a basic plane according to the basic outer wall skin image, and taking the projection of the building single-sided outer wall on the basic plane as a standard wall surface;
projecting the monitoring mark array image on the basic plane to obtain a projection image;
setting a plurality of mark areas on the projection image by taking the center point of each thin mark on the projection image as the center;
and according to the relative positions of the camera and the standard wall surface, an observation point is established, and all marked areas on the standard wall surface observed by the observation point are obtained and used as standard monitoring mark array images.
5. The method for monitoring the anti-falling off of the building exterior wall skin according to claim 4, wherein the step of obtaining the change matrix of the monitoring mark array is as follows:
establishing a basic coordinate system on the basic plane, wherein the origin of the basic coordinate system is the lower left point of a standard wall surface, the horizontal direction of the basic plane is used as a horizontal axis, the vertical axis is used as a vertical axis, the standard vector of each marking area is marked as X= [ Xa, xb, xc, xd ], xa represents the abscissa of the marking area, xb represents the ordinate of the marking area, xc represents the perimeter of the marking area, and Xd represents the area of the marking area;
establishing a monitoring vector for each thin marker on the projection image, wherein the monitoring vector is marked as Y= [ Ya, yb, yc and Yd ], wherein Ya represents the abscissa of the thin marker on the projection image, yb represents the ordinate of the thin marker on the projection image, yc represents the perimeter of the thin marker on the projection image, and Yd represents the area of the thin marker on the projection image;
calculating a standard vector of each marking area, and calculating a difference value of a monitoring vector of the thin marker on the corresponding projection image, wherein Dxy=X-Y;
and establishing a change matrix of Dxy corresponding to all the marked areas.
6. The method for monitoring the falling off of the exterior wall skin of the building according to claim 5, wherein the step of calculating the change matrix by using a pre-trained exterior wall skin falling off model to obtain the falling off area of the exterior wall skin and the falling off residual time of the exterior wall skin specifically comprises the following steps:
building a training set: the training set comprises more than 100 groups of experimental data, wherein the experiment is a small-scene experiment, and the specific scene is that an experimental single-sided wall with the height of 3 meters and the width of 2 meters is established, a thin marker is attached or coated on the experimental single-sided wall to form an experimental building outer wall skin monitoring mark array, an experimental standard monitoring mark array image is obtained, and the experimental single-sided wall is subjected to simulated sedimentation, simulated shaking and simulated blowing with multiple grades and multiple angles until the experimental single-sided wall is subjected to large-area outer wall skin falling, and the large-area outer wall skin falling is more than 3 square meters;
recording an experimental building outer wall skin monitoring mark array as a training array set every 1 second;
taking the moment of each falling of the outer wall skin as a splitting moment, recording the area proportion of the corresponding falling wall skin, and splitting the training array set into a plurality of training array subsets;
in each training array subset, calculating a change matrix of each experimental building exterior wall skin monitoring mark array and each experimental building exterior wall skin monitoring mark array, marking the change matrix as an experimental change matrix, and recording a time interval between a moment corresponding to the experimental change matrix and the last element in the training array subset as the falling residual time of the experimental wall skin;
the training input of the training set is an experimental change matrix corresponding to each experimental building exterior wall skin monitoring mark array in the training array set; the output of the training set is the area proportion of the falling wall skin corresponding to the experimental building outer wall skin monitoring mark array and the falling residual time of the experimental wall skin;
modeling and training: establishing an outer wall skin falling model prototype by using a convolutional neural network, and training by using the training set to obtain an outer wall skin falling model;
using a trained model: and calculating the change matrix by using an outer wall skin falling model to obtain corresponding falling wall skin area proportion and wall skin falling residual time, and calculating the falling area according to the falling wall skin area proportion and the total area of the outer wall skin.
7. The method for monitoring the falling off of an exterior wall skin of a building according to claim 6, wherein the thin-shaped marker is rectangular.
8. The method for monitoring the falling off prevention of the building exterior skin according to claim 4, wherein in the step of selecting the basic exterior skin image shot by the camera containing all the thin markers, if the thin markers are blocked in the exterior skin image shot by the camera, the unmanned aerial vehicle is adopted to orthogonally shoot each blocked thin marker of the building exterior skin, the image of the blocked thin markers is obtained, and the corresponding positions of the basic exterior skin image are marked according to the corresponding sizes of the thin markers.
9. A computer readable storage medium, wherein program instructions are stored in the computer readable storage medium, and when the program instructions are executed, the program instructions are used to perform the building exterior skin anti-drop monitoring method according to any one of claims 1-8.
10. A building exterior skin fall-off prevention monitoring system comprising the computer readable storage medium of claim 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310545900.3A CN116434091B (en) | 2023-05-15 | 2023-05-15 | Building outer wall skin anti-drop monitoring method, medium and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310545900.3A CN116434091B (en) | 2023-05-15 | 2023-05-15 | Building outer wall skin anti-drop monitoring method, medium and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116434091A true CN116434091A (en) | 2023-07-14 |
| CN116434091B CN116434091B (en) | 2024-08-02 |
Family
ID=87087411
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310545900.3A Active CN116434091B (en) | 2023-05-15 | 2023-05-15 | Building outer wall skin anti-drop monitoring method, medium and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116434091B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117169086A (en) * | 2023-09-08 | 2023-12-05 | 中建安装集团有限公司 | Method, medium and system for detecting construction quality of underground waterproof layer of building |
| CN117935097A (en) * | 2024-03-25 | 2024-04-26 | 中冶建筑研究总院（深圳）有限公司 | Building surface analysis method and system based on unmanned aerial vehicle wall crawling |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110044964A (en) * | 2019-04-25 | 2019-07-23 | 湖南科技大学 | Architectural coating layer debonding defect recognition methods based on unmanned aerial vehicle thermal imaging video |
| CN112258482A (en) * | 2020-10-23 | 2021-01-22 | 广东博智林机器人有限公司 | Building exterior wall mortar flow drop detection method and device |
| CN112435253A (en) * | 2020-12-08 | 2021-03-02 | 深圳创维数字技术有限公司 | Wall falling detection method and device and readable storage medium |
| CN113610782A (en) * | 2021-07-21 | 2021-11-05 | 武汉理工大学 | Building deformation monitoring method and equipment and storage medium |
| US20220076019A1 (en) * | 2020-09-04 | 2022-03-10 | Zillow, Inc. | Automated Analysis Of Image Contents To Determine The Acquisition Location Of The Image |
| CN114708195A (en) * | 2022-03-11 | 2022-07-05 | 湖南科技大学 | Glass curtain wall safety detection method and device and computer equipment |
| CN115574864A (en) * | 2022-10-10 | 2023-01-06 | 武汉誉钱乐房地产有限公司 | Building dangerous house intelligent monitoring management system based on data acquisition |
-
2023
- 2023-05-15 CN CN202310545900.3A patent/CN116434091B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110044964A (en) * | 2019-04-25 | 2019-07-23 | 湖南科技大学 | Architectural coating layer debonding defect recognition methods based on unmanned aerial vehicle thermal imaging video |
| US20220076019A1 (en) * | 2020-09-04 | 2022-03-10 | Zillow, Inc. | Automated Analysis Of Image Contents To Determine The Acquisition Location Of The Image |
| CN112258482A (en) * | 2020-10-23 | 2021-01-22 | 广东博智林机器人有限公司 | Building exterior wall mortar flow drop detection method and device |
| CN112435253A (en) * | 2020-12-08 | 2021-03-02 | 深圳创维数字技术有限公司 | Wall falling detection method and device and readable storage medium |
| CN113610782A (en) * | 2021-07-21 | 2021-11-05 | 武汉理工大学 | Building deformation monitoring method and equipment and storage medium |
| CN114708195A (en) * | 2022-03-11 | 2022-07-05 | 湖南科技大学 | Glass curtain wall safety detection method and device and computer equipment |
| CN115574864A (en) * | 2022-10-10 | 2023-01-06 | 武汉誉钱乐房地产有限公司 | Building dangerous house intelligent monitoring management system based on data acquisition |
Non-Patent Citations (1)
| Title |
|---|
| 曹艳玲;袁义宏;: "基于特征分布的建筑墙体裂缝图像识别方法", 沈阳工业大学学报, no. 02, 26 February 2018 (2018-02-26), pages 117 - 122 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117169086A (en) * | 2023-09-08 | 2023-12-05 | 中建安装集团有限公司 | Method, medium and system for detecting construction quality of underground waterproof layer of building |
| CN117935097A (en) * | 2024-03-25 | 2024-04-26 | 中冶建筑研究总院（深圳）有限公司 | Building surface analysis method and system based on unmanned aerial vehicle wall crawling |
| CN117935097B (en) * | 2024-03-25 | 2024-05-31 | 中冶建筑研究总院（深圳）有限公司 | Building surface analysis method and system based on unmanned aerial vehicle wall crawling |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116434091B (en) | 2024-08-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116434091B (en) | Building outer wall skin anti-drop monitoring method, medium and system | |
| CN111241959B (en) | Method for detecting personnel not wearing safety helmet through construction site video stream | |
| CN109117858B (en) | Method and device for monitoring icing of wind driven generator blade | |
| CN104123544A (en) | Video analysis based abnormal behavior detection method and system | |
| CN113903081A (en) | Visual identification artificial intelligence alarm method and device for images of hydraulic power plant | |
| CN113743256B (en) | Intelligent early warning method and device for site safety | |
| CN105139429B (en) | A kind of fire detection method based on flame notable figure and spatial pyramid histogram | |
| US20220414901A1 (en) | Real-time detection method of block motion based on feature point recognition | |
| CN116631158A (en) | Landslide early warning device based on image recognition and computer vision | |
| CN111062373A (en) | Hoisting process danger identification method and system based on deep learning | |
| CN106548113A (en) | Image-recognizing method and system | |
| CN112258490A (en) | Low-emissivity coating intelligent damage detection method based on optical and infrared image fusion | |
| CN103530988B (en) | A kind of lawn is trampleed on and the method for early warning of the extent of injury | |
| CN111723656A (en) | Smoke detection method and device based on YOLO v3 and self-optimization | |
| CN115880231A (en) | Power transmission line hidden danger detection method and system based on deep learning | |
| CN111046822A (en) | Large vehicle anti-theft method based on artificial intelligence video identification | |
| van den Hengel et al. | Automatic camera placement for large scale surveillance networks | |
| CN114792405A (en) | Overhead line foreign matter intrusion prevention monitoring method | |
| CN110276379A (en) | A kind of the condition of a disaster information rapid extracting method based on video image analysis | |
| CN111950518B (en) | Video image enhancement method for violent behavior recognition | |
| CN105335985B (en) | A kind of real-time capturing method and system of docking aircraft based on machine vision | |
| WO2022104798A1 (en) | 5g-based unmanned electronic traffic police duty system | |
| CN110490117B (en) | Parking event determination method and system based on image depth information | |
| CN108322710B (en) | Crowd evacuation device and system | |
| CN105574490A (en) | Vehicle brand identification method and system based on headlight image characteristics |
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 |