CN117710809A - Intelligent detection method and system for building outer wall falling - Google Patents
Intelligent detection method and system for building outer wall falling Download PDFInfo
- Publication number
- CN117710809A CN117710809A CN202311607489.4A CN202311607489A CN117710809A CN 117710809 A CN117710809 A CN 117710809A CN 202311607489 A CN202311607489 A CN 202311607489A CN 117710809 A CN117710809 A CN 117710809A
- Authority
- CN
- China
- Prior art keywords
- image data
- visual
- area
- data
- infrared
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 37
- 230000000007 visual effect Effects 0.000 claims abstract description 166
- 238000004458 analytical method Methods 0.000 claims abstract description 41
- 238000000034 method Methods 0.000 claims abstract description 30
- 238000012216 screening Methods 0.000 claims abstract description 20
- 238000012545 processing Methods 0.000 claims abstract description 19
- 238000001931 thermography Methods 0.000 claims abstract description 19
- 238000001914 filtration Methods 0.000 claims description 15
- 238000000605 extraction Methods 0.000 claims description 13
- 238000003708 edge detection Methods 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 9
- 238000012706 support-vector machine Methods 0.000 claims description 9
- 230000003044 adaptive effect Effects 0.000 claims description 6
- 230000002708 enhancing effect Effects 0.000 claims description 6
- 238000009499 grossing Methods 0.000 claims description 6
- 238000013136 deep learning model Methods 0.000 claims description 4
- 230000003449 preventive effect Effects 0.000 abstract description 4
- 230000007547 defect Effects 0.000 description 7
- 230000009286 beneficial effect Effects 0.000 description 4
- 238000005336 cracking Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000005855 radiation Effects 0.000 description 3
- 239000000284 extract Substances 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000006740 morphological transformation Effects 0.000 description 2
- 230000032683 aging Effects 0.000 description 1
- 239000011449 brick Substances 0.000 description 1
- 238000009435 building construction Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000003331 infrared imaging Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012634 optical imaging Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- 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 provides an intelligent detection method and system for building outer wall falling, wherein the method comprises the following steps: collecting and extracting the external vertical surface characteristics of a building, and making an aircraft track of the unmanned aerial vehicle; controlling the unmanned aerial vehicle to fly according to the flight track, and acquiring data to obtain outer-facade infrared image data and outer-facade visual image data; analyzing the outer facade infrared image data based on an infrared thermal imaging analysis model, and screening out an infrared threat area; analyzing and processing the outer visual image data and the infrared threat area position data based on the visual analysis model, and performing secondary judgment on the visual danger areas with interference areas removed to obtain danger data; and outputting dangerous data and sending out early warning signals. The invention can perform multi-angle, full-coverage and close-range falling detection on the building outer wall, so that staff can comprehensively analyze the building outer wall and take preventive measures to avoid safety accidents.
Description
Technical Field
The invention relates to the technical field of intelligent detection, in particular to an intelligent detection method and system for building outer wall falling.
Background
With the continuous expansion of university cities and the increase of the number of people, the problems of building construction of universities are increasingly prominent. Generally, the service life of the teaching building of a school is about 30-50 years, while dormitory buildings and teaching buildings of most universities are built in the last century, the service life reaches or even exceeds the expected life, and the damage such as aging, cracking, hollowing, tile falling and the like of a plurality of buildings is caused. Accidents such as falling off and injuring people of each school outer wall brick are frequent, and the accidents cause serious threat to personal safety of teachers and students. Therefore, it is necessary to periodically detect the facing layer of the outer wall of the campus building, and how to quickly and simply eliminate the potential safety hazard becomes a key for preventing such an event from happening continuously.
At present, common detection modes comprise a visual detection method, a hammering method and the like, but the methods have the defects of high cost, long period, low efficiency and the like due to dependence on a scaffold or a crane; the infrared thermal image detection can be used for analyzing and judging the change of the external temperature of the wall surface, but is easily limited by the observation distance and elevation angle and influenced by the material quality and color of the outer elevation material, so that the error and uncertainty are high; the computer vision method can be used for detecting and rapidly analyzing a large amount of data, so that time and resources are saved, but the computer vision technology still causes larger errors under the influence of illumination and visual angles. In the face of the complex building facade damage detection field, the method is obviously difficult to adapt to the new requirements of the development trend.
Therefore, how to accurately and efficiently detect the falling-off condition of the building outer wall becomes a problem which needs to be solved by workers in the current field.
Disclosure of Invention
In view of the above, it is necessary to provide an intelligent detection method and system for the falling of the building outer wall, so as to achieve the purpose of accurately and efficiently detecting the falling condition of the building outer wall.
In order to achieve the above object, the present invention provides an intelligent detection method for building exterior wall falling, comprising:
collecting surface image data of a building body, extracting external elevation features of the building body, and making an aircraft track of the unmanned aerial vehicle based on the external elevation features of the building body;
controlling the unmanned aerial vehicle to fly according to the flight track, and acquiring data to obtain outer-facade infrared image data and outer-facade visual image data;
analyzing the outer facade infrared image data based on an infrared thermal imaging analysis model, screening out an infrared threat area if threat exists, and determining the position data of the infrared threat area;
analyzing and processing the visual image data of the outer facade based on the visual analysis model, screening out visual dangerous areas if dangers exist, and performing secondary judgment on the visual dangerous areas excluding the interference areas to obtain dangerous data, wherein the dangerous data comprise falling dangerous area data and falling dangerous area position data;
and outputting dangerous data and sending out early warning signals.
In one possible implementation, collecting building surface image data, extracting building external facade features, and formulating an aircraft trajectory of the unmanned aerial vehicle based on the building external facade features, including:
collecting surface image data of a building body, and extracting external elevation features of the building body;
dividing the external vertical face of the building into a plurality of rectangular areas based on the external vertical face characteristics of the building;
performing obstacle removing treatment on the rectangular area to obtain a target rectangular area;
and establishing the unmanned aerial vehicle flight track traversing the target rectangular area based on a grid mode.
In one possible implementation, the facade image data is acquired based on an infrared camera assembly and a visual camera assembly onboard the drone.
In one possible implementation, the infrared thermal imaging analysis model is a deep learning model based on a support vector machine algorithm, and the kernel function of the support vector machine is a radial basis kernel function.
In one possible implementation manner, the analyzing and processing are performed on the visual image data of the outer surface based on the visual analysis model, and the secondary judging is performed on the threat area with the interference area removed, so as to obtain the dangerous data, including:
screening the facade visual image data, and removing the facade visual image data with definition lower than a preset threshold value to obtain first visual image data;
filtering and enhancing the first visual image data to obtain second visual image data;
extracting features of the second visual image data to obtain a visual hazard area, and removing an interference area in the visual hazard area to obtain third visual image data;
and performing secondary judgment on the visual danger area in the third visual image data to obtain danger data.
In one possible implementation, filtering and enhancing the first visual image data to obtain the second visual image data includes:
performing filtering processing on the first visual image data based on the adaptive smoothing filtering;
and carrying out contrast enhancement processing on the first visual image data based on a histogram gray stretching method to obtain second visual image data.
In one possible implementation manner, feature extraction is performed on the second visual image data, and an interference area in the threat area is removed, so as to obtain third visual image data, including:
and carrying out feature extraction on the second visual image data based on an image edge detection method, and removing an interference area in the visual danger area based on a curve feature extraction method to obtain third visual image data.
In one possible implementation, rejecting interference regions in a visual hazard region based on a curve feature extraction method includes:
acquiring a crack skeleton in the second visual image based on an image refinement algorithm, and removing a burr area and a fracture area in the crack skeleton by utilizing image closing operation to obtain a target crack skeleton;
extracting pixel coordinates of a target crack skeleton, and calculating a linear characteristic fitting equation based on the pixel coordinates of the target crack skeleton;
and judging the linear characteristic fitting equation by using the curvature to obtain a visual hazard area, and removing a linear interference area in the target crack skeleton to obtain third visual image data.
In one possible implementation manner, performing secondary judgment on the visual hazard area in the third visual image data to obtain hazard data includes:
and performing secondary judgment on the visual hazard area in the third visual image data based on the infrared threat area so as to judge the difference area between the infrared threat area and the visual hazard area and obtain hazard data.
In order to achieve the above object, the present invention further provides an intelligent detection system for building exterior wall falling, comprising:
the flight path planning module is used for collecting surface image data of the building body, extracting external elevation features of the building body and formulating an aircraft path of the unmanned aerial vehicle based on the external elevation features of the building body;
the data acquisition module is used for controlling the unmanned aerial vehicle to fly according to the flight track and acquiring data to obtain outer-facade infrared image data and outer-facade visual image data;
the infrared thermal imaging analysis module is used for analyzing the outer vertical face infrared image data based on the infrared thermal imaging analysis model, screening out an infrared threat area if threat exists, and determining the position data of the infrared threat area;
the visual analysis module is used for analyzing and processing the visual image data of the outer facade based on the visual analysis model, screening out dangerous areas if dangers exist, and carrying out secondary judgment on the visual dangerous areas excluding the interference areas to obtain dangerous data, wherein the dangerous data comprise falling dangerous area data and falling dangerous area position data;
and the output module is used for outputting dangerous data and sending out early warning signals.
The beneficial effects of adopting the embodiment are as follows: collecting surface image data of a building body, extracting external elevation features of the building body, and making an aircraft track of the unmanned aerial vehicle based on the external elevation features of the building body; controlling the unmanned aerial vehicle to fly according to the flight track, and acquiring data to obtain outer-facade infrared image data and outer-facade visual image data; analyzing the outer facade infrared image data based on an infrared thermal imaging analysis model, screening out an infrared threat area, and determining infrared threat area position data; analyzing and processing the outer visual image data and the infrared threat area position data based on the visual analysis model, and performing secondary judgment on the visual danger areas with interference areas removed to obtain danger data; and outputting dangerous data and sending out early warning signals. The invention can perform multi-angle, full-coverage and close-range falling detection on the building outer wall, so that staff can comprehensively analyze the building outer wall and take preventive measures to avoid safety accidents.
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 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 schematic flow chart of an embodiment of an intelligent detection method for building exterior wall falling off provided by the invention;
fig. 2 is a schematic flow chart of step S11 in an embodiment of a method for intelligently detecting falling of an outer wall of a building according to the present invention;
fig. 3 is a schematic flow chart of step S14 in an embodiment of a method for intelligently detecting falling of an outer wall of a building according to the present invention;
fig. 4 is a schematic structural diagram of an embodiment of an intelligent detection system for building exterior wall falling.
Detailed Description
The technical solutions in 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. It will be apparent that the described embodiments are only some, but not all, embodiments of the invention. 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.
Some of the block diagrams shown in the figures are functional entities and do not necessarily correspond to physically or logically separate entities. These functional entities may be implemented in software or in one or more hardware modules or integrated circuits or in different networks and/or processor systems and/or microcontroller systems.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the invention. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
Fig. 1 is a schematic flow chart of an embodiment of an intelligent detection method for building exterior wall falling.
Referring to fig. 1, the invention provides an intelligent detection method for building exterior wall falling, which comprises the following steps:
s11, acquiring building surface image data, extracting building external elevation features, and making an airplane track of the unmanned aerial vehicle based on the building external elevation features;
s12, controlling the unmanned aerial vehicle to fly according to the flight track, and acquiring data to obtain outer-facade infrared image data and outer-facade visual image data;
s13, analyzing the outer vertical face infrared image data based on an infrared thermal imaging analysis model, screening out an infrared threat area if threat exists, and determining infrared threat area position data;
s14, analyzing and processing the visual image data of the outer facade based on the visual analysis model, screening out dangerous areas if danger exists, and performing secondary judgment on the visual dangerous areas excluding the interference areas to obtain dangerous data, wherein the dangerous data comprise falling dangerous area data and falling dangerous area position data;
s15, outputting dangerous data and sending out early warning signals.
The beneficial effects of adopting the embodiment are as follows: collecting surface image data of a building body, extracting external elevation features of the building body, and making an aircraft track of the unmanned aerial vehicle based on the external elevation features of the building body; controlling the unmanned aerial vehicle to fly according to the flight track, and acquiring data to obtain outer-facade infrared image data and outer-facade visual image data; analyzing the outer facade infrared image data based on an infrared thermal imaging analysis model, screening out an infrared threat area, and determining infrared threat area position data; analyzing and processing the outer visual image data and the infrared threat area position data based on the visual analysis model, and performing secondary judgment on the visual danger areas with interference areas removed to obtain danger data; and outputting dangerous data and sending out early warning signals. The invention can perform multi-angle, full-coverage and close-range falling detection on the building outer wall, so that staff can comprehensively analyze the building outer wall and take preventive measures to avoid safety accidents.
It will be appreciated that the present embodiment includes mainly two aspects, namely data acquisition and data processing. The data acquisition can be carried by the unmanned aerial vehicle and is collected by the infrared camera shooting assembly and the visual camera shooting assembly, the data processing part is used for analyzing and judging the acquired infrared data through an infrared imaging analysis model, outputting an infrared threat area when a defect exists, extracting the position of the infrared threat area, analyzing and judging the acquired visual data through a visual analysis model, outputting a visual danger area when the visual analysis model judges that a curve exists, comprehensively judging the infrared threat area and the visual danger area, determining the danger area, and outputting dangerous data and early warning signals.
Fig. 2 is a schematic flow chart of step S11 in an embodiment of a method for intelligently detecting falling of an external wall of a building according to the present invention.
Referring to fig. 2, step S11 includes:
s111, collecting surface image data of a building body, and extracting external elevation features of the building body;
s112, dividing the external vertical face of the building into a plurality of rectangular areas based on the external vertical face characteristics of the building;
s113, performing obstacle removing processing on the rectangular area to obtain a target rectangular area;
s114, establishing an unmanned aerial vehicle flight track traversing the target rectangular area based on a grid mode.
It can be appreciated that the unmanned aerial vehicle flight trajectory can be set by a component carrying a GPS system, and the setting rules include: for a vertical flat facade, the flight path starts at a predetermined corner, moves up in the vertical direction to the next break point, then proceeds down, repeating this pattern until the entire facade is recorded and the drone moves to the next facade in a similar manner; for a horizontal flat facade, the path should start at a predetermined corner and continue to move to the right, then move upwards by a break and move to the left in a linear fashion, repeating this pattern until the entire facade is recorded; after capturing the building surface image data, the drone may capture a thermal image of the rooftop in a similar grid fashion, starting at one corner and moving in a horizontal or vertical fashion along the superimposed grid until the rooftop area is scanned intact.
Preferably, before data acquisition, the flight path needs to be formulated to avoid obstacles or interference, and meanwhile, the flight path needs to be continuous as much as possible, so that the traceability of the position of the acquired information of the operation data is ensured. Meanwhile, the relationship between the unmanned aerial vehicle and the surrounding environment is checked in real time. When the wind speed is proper, the unmanned aerial vehicle can be unlocked for taking off after confirming that the unmanned aerial vehicle has a safe taking-off condition.
Further, the facade image data is acquired based on an infrared camera assembly and a visual camera assembly carried on the unmanned aerial vehicle.
In one embodiment, the infrared thermal imaging analysis model is a deep learning model based on a support vector machine algorithm, and the kernel function of the support vector machine is a radial basis kernel function.
It is understood that the infrared camera assembly receives the infrared radiation energy distribution pattern of the detected area by using the infrared detector and the optical imaging objective lens, and reflects the infrared radiation energy distribution pattern to the photosensitive element of the infrared detector, thereby obtaining an infrared thermal image corresponding to the thermal distribution field of the surface of the detected area. The invisible infrared radiation on the surface of the outer wall is converted into a thermal image (different colors represent different temperature areas of the surface to be measured) through the external temperature change, and then the whole temperature distribution condition analysis is carried out on the thermal image through the infrared thermal imaging analysis model, so that whether threat exists is judged.
Specifically, the infrared thermal imaging analysis model comprises a deep learning model based on a support vector machine algorithm, and the embodiment trains an automatic defect identification model by establishing a small sample surface defect identification data set. A Radial Basis Function (RBF) kernel function with strong learning ability and wide application range is selected as a kernel function of a support vector machine. The optimal segmentation plane can be constructed through a support vector machine algorithm, and the shape feature vector of the defect area based on screening is mapped to a high-dimensional space, so that the automatic screening of the defect area and the non-defect area is realized.
Fig. 3 is a schematic flow chart of step S14 in an embodiment of the intelligent detection method for building exterior wall falling provided by the present invention.
Referring to fig. 3, step S14 includes:
s141, screening the outer facade visual image data, and removing the outer facade visual image data with definition lower than a preset threshold value to obtain first visual image data;
s142, filtering and enhancing the first visual image data to obtain second visual image data;
s143, extracting features of the second visual image data to obtain a visual danger area, and eliminating interference areas in the visual danger area to obtain third visual image data;
s144, performing secondary judgment on the visual danger area in the third visual image data to obtain danger data.
The analysis and processing of the facade visual image data and the infrared threat area position data by the visual analysis model can be divided into three steps: image screening, preprocessing, and object extraction and recognition are described separately below.
In step S141, screening the facade visual image data to screen out building exterior wall images meeting the quality requirement, and deleting the pictures with serious blurring and low definition to obtain first visual image data; step S142 uses adaptive smoothing filtering to replace a four-neighborhood average smoothing algorithm in SFC combination method to extract the characteristics of the first visual image data, and obtains second visual image data. Therefore, the filtering parameters are automatically adjusted according to different conditions, so that the algorithm applicability is stronger; step S143, extracting features of the second visual image data, namely extracting a region possibly with cracks to obtain a visual dangerous region, and eliminating an interference region in the visual dangerous region, wherein the interference region can comprise structures such as wall parting lines, partial windows and railings, and the recognition accuracy of the cracks can be effectively improved by eliminating the structures such as the wall parting lines, the partial windows and the railings; and step S144, performing secondary judgment on the visual hazard area in the third visual image data to obtain hazard data, thereby further improving the crack identification precision.
In one embodiment, step S142 includes:
performing filtering processing on the first visual image data based on the adaptive smoothing filtering;
and carrying out contrast enhancement processing on the first visual image data based on a histogram gray stretching method to obtain second visual image data.
In one use scene, enhancing the contrast of the image by using a histogram gray stretching method; performing adaptive filtering smoothing on the image for 1 time and Laplacian sharpening for 1 time, and repeating the steps for 4 times; performing contrast enhancement treatment on the crack image for 2 times by using an improved histogram stretching method; the image of the crack was smoothed 2 times using adaptive filtering and contrast enhanced 5 times using a modified histogram stretching method. The method can effectively remove noise on the basis of keeping the image quality, improves the contrast of the image, and highlights the crack characteristics, thereby being beneficial to further crack identification work.
In one embodiment, feature extraction is performed on the second visual image data, and an interference area in the threat area is removed to obtain third visual image data, including:
and carrying out feature extraction on the second visual image data based on an image edge detection method, and removing an interference area in the visual danger area based on a curve feature extraction method to obtain third visual image data.
Further, the object extraction section adopts an image edge detection method that extracts edge information from a high frequency component of an image by differential operation, and then performs edge detection of second visual image data based on a gradient operator template edge detection algorithm. The edge detection operator has higher signal-to-noise ratio and detection precision, the embodiment uses an algorithm combining a maximum inter-class variance method (Otsu) method and a gradient histogram to calculate a high threshold value, and sets a low threshold value to be half of the high threshold value, and the embodiment adopts an improved algorithm to enhance the self-adaptive capacity of the edge detection operator.
The Otsu method is a method of automatically determining a threshold value to maximize an inter-class variance. Assuming that the image pixel is N, the gray scale range is [0, L-1]The number of pixels corresponding to the gray level i is n i The probability of occurrence is:
dividing pixels in the second visual image into two types of C0 and C1 according to a gray value by using a threshold T, wherein C0 consists of pixels with gray values of [0, T ], C1 consists of pixels with gray values of [ T+1, L-1], and for gray distribution probability, the gray distribution probability average value of the second visual image is as follows:
u T and (3) representing a gray distribution probability average value of the second visual image, wherein the gray distribution probability average value of C0 and C1 is expressed as follows:
u 0 represents the gray distribution probability mean value of C0, u 1 Representing the gray level distribution probability mean value of C1;
further, the expression of the inter-class variance is:
wherein,representing the inter-class variance.
In the embodiment, T is sequentially valued in the range of [0, L-1], so that the T value with the largest inter-class variance is the optimal threshold value of the Otsu method. After the threshold value is obtained by the Otsu method, because the gradient histogram shows information such as the intensity of the edge, in order to reduce the loss of some necessary crack edge points, whether the currently obtained threshold value can cause the loss of the edge information can be more accurately observed by adopting the gradient histogram.
In one embodiment, in the image of the architectural surface crack acquired by the drone, there is typically a target area suspected of cracking and an interference area that is not cracking. Therefore, the extracted edge information not only comprises the edge of the crack in the image, but also comprises the edge information of a wall parting line, a part of window and the like, and inconvenience is brought to the identification and measurement of the crack, and the observation shows that the interference information is generally window frames, railing structural joints and the like, and has regular straight line characteristics, and the crack is mostly an irregular and non-directional curve, so that the invention uses the curvature of the crack line as the standard for eliminating the interference area.
In one embodiment, rejecting interference regions in a visual risk region based on a curve feature extraction method includes:
acquiring a crack skeleton in the second visual image based on an image refinement algorithm, and removing a burr area and a fracture area in the crack skeleton by utilizing image closing operation to obtain a target crack skeleton;
extracting pixel coordinates of a target crack skeleton, and calculating a linear characteristic fitting equation based on the pixel coordinates of the target crack skeleton;
and judging the linear characteristic fitting equation by using the curvature to obtain a visual hazard area, and removing a linear interference area in the target crack skeleton to obtain third visual image data.
It can be understood that, to obtain the linear characteristic of the crack, further obtain the curvature information thereof, on the basis of the two-vision image, firstly, an image thinning algorithm is used to obtain the crack skeleton. Then, eliminating partial tiny burrs or fracture conditions through image closing operation, wherein line segment-shaped structural elements are selected as templates for morphological transformation according to image features of cracks; the morphological transformation and refinement of the building outer wall cracks realize the transformation of the binary image after edge detection into a single-pixel line. On the basis, the coordinates of all pixel points forming the line in the binary image are required to be extracted, and the linear characteristics of the lines are described according to a mathematical equation fitted by the coordinates; after fitting the polynomial equation of the line, the curvature of any point on the line is calculated. According to the result, if a straight line is fitted, the straight line can be deleted, and if a curve is fitted, under the condition that no other special interference objects exist, the existence of cracks in the crack image of the building outer wall can be judged, and the crack image needs to be processed.
Further, performing secondary judgment on the visual hazard area in the third visual image data to obtain hazard data, including:
and performing secondary judgment on the visual hazard area in the third visual image data based on the infrared threat area so as to judge the difference area between the infrared threat area and the visual hazard area and obtain hazard data.
Fig. 4 is a schematic structural diagram of an embodiment of an intelligent detection system for building exterior wall falling.
Referring to fig. 4, the present invention further provides an intelligent detection system for building exterior wall falling, including:
the flight path planning module 41 is used for acquiring the surface image data of the building body, extracting the external vertical surface characteristics of the building body and making the plane path of the unmanned aerial vehicle based on the external vertical surface characteristics of the building body;
the data acquisition module 42 is used for controlling the unmanned aerial vehicle to fly according to the flight track and acquiring data to obtain outer-facade infrared image data and outer-facade visual image data;
the infrared thermal imaging analysis module 43 is configured to analyze the outer facade infrared image data based on the infrared thermal imaging analysis model, and if a threat exists, screen out an infrared threat area, and determine position data of the infrared threat area;
the visual analysis module 44 is configured to analyze and process the visual image data of the outer facade based on the visual analysis model, if there is a danger, screen out a dangerous area, and perform secondary judgment on the visual dangerous area with the interference area removed to obtain dangerous data, where the dangerous data includes data of a falling dangerous area and position data of the falling dangerous area;
and the output module 45 is used for outputting the dangerous data and sending out an early warning signal.
The beneficial effects of adopting the embodiment are as follows: the flight path planning module 41 collects the surface image data of the building body, extracts the external vertical surface characteristics of the building body, and formulates the plane path of the unmanned aerial vehicle based on the external vertical surface characteristics of the building body; the data acquisition module 42 controls the unmanned aerial vehicle to fly according to the flight track, and acquires data to obtain the outer-facade infrared image data and the outer-facade visual image data; the infrared thermal imaging analysis module 43 analyzes the outer surface infrared image data based on the infrared thermal imaging analysis model, screens out an infrared threat area, and determines infrared threat area position data; the visual analysis module 44 analyzes and processes the outer visual image data and the infrared threat area position data based on the visual analysis model, and performs secondary judgment on the visual danger area with the interference area removed to obtain danger data; the output module 45 outputs the dangerous data and sends out the early warning signal. The invention can perform multi-angle, full-coverage and close-range falling detection on the building outer wall, so that staff can comprehensively analyze the building outer wall and take preventive measures to avoid safety accidents.
The intelligent detection system for building exterior wall falling provided in the above embodiment can implement the technical scheme described in the above embodiment of the intelligent detection method for building exterior wall falling, and the specific implementation principle of each module or unit can be based on the corresponding content in the embodiment of the intelligent detection method for building exterior wall falling, which is not described herein again.
The above describes in detail the method and system for detecting the falling off of the building outer wall, and specific examples are applied to describe the principle and implementation of the invention, and the description of the above examples is only used for helping to understand the method and core idea of the invention; meanwhile, as those skilled in the art will vary in the specific embodiments and application scope according to the ideas of the present invention, the present description should not be construed as limiting the present invention in summary.
Claims (10)
1. An intelligent detection method for building exterior wall falling off is characterized by comprising the following steps:
collecting surface image data of a building body, extracting external elevation features of the building body, and formulating an airplane track of the unmanned aerial vehicle based on the external elevation features of the building body;
controlling the unmanned aerial vehicle to fly according to the flying track, and acquiring data to obtain outer-facade infrared image data and outer-facade visual image data;
analyzing the outer elevation infrared image data based on an infrared thermal imaging analysis model, screening out an infrared threat area if threat exists, and determining infrared threat area position data;
analyzing and processing the outer facade visual image data based on a visual analysis model, screening out visual dangerous areas if danger exists, and performing secondary judgment on the visual dangerous areas excluding the interference areas based on the infrared threat areas to obtain dangerous data, wherein the dangerous data comprise falling dangerous area data and falling dangerous area position data;
and outputting the dangerous data and sending out an early warning signal.
2. The intelligent detection method for building exterior wall falling off according to claim 1, wherein the steps of collecting building surface image data, extracting building exterior elevation features, and making an aircraft track of the unmanned aerial vehicle based on the building exterior elevation features comprise:
collecting surface image data of a building body, and extracting external elevation features of the building body;
dividing the external vertical face of the building into a plurality of rectangular areas based on the external vertical face characteristics of the building;
performing obstacle removing treatment on the rectangular region to obtain a target rectangular region;
and establishing the unmanned aerial vehicle flight track traversing the target rectangular area based on a grid mode.
3. The intelligent detection method for building exterior wall falling off according to claim 1, wherein the exterior wall image data is acquired based on an infrared camera assembly and a visual camera assembly mounted on the unmanned aerial vehicle.
4. The intelligent detection method for building exterior wall falling off according to claim 1, wherein the infrared thermal imaging analysis model is a deep learning model based on a support vector machine algorithm, and a kernel function of the support vector machine is a radial basis kernel function.
5. The intelligent detection method for building exterior wall falling according to claim 1, wherein the analyzing and processing the exterior wall visual image data based on the visual analysis model, screening a visual dangerous area if a danger exists, and performing secondary judgment on a dangerous area excluding an interference area based on the infrared threat area to obtain dangerous data, comprises:
screening the facade visual image data, and removing the facade visual image data with definition lower than a preset threshold value to obtain first visual image data;
filtering and enhancing the first visual image data to obtain second visual image data;
extracting features of the second visual image data to obtain a visual danger area, and removing an interference area in the visual danger area to obtain third visual image data;
and performing secondary judgment on the visual danger area in the third visual image data to obtain danger data.
6. The intelligent detection method for building exterior wall drop according to claim 5, wherein the filtering and enhancing the first visual image data to obtain second visual image data comprises:
filtering the first visual image data based on adaptive smoothing filtering;
and carrying out contrast enhancement processing on the first visual image data based on a histogram gray stretching method to obtain second visual image data.
7. The intelligent detection method for building exterior wall falling off according to claim 5, wherein the performing feature extraction on the second visual image data, removing the interference area in the threat area, and obtaining third visual image data includes:
and extracting features of the second visual image data based on an image edge detection method, and removing the interference area in the visual danger area based on a curve feature extraction method to obtain third visual image data.
8. The intelligent detection method for building exterior wall falling off according to claim 7, wherein the eliminating the interference area in the vision-dangerous area based on the curve feature extraction method comprises:
acquiring a crack skeleton in the second visual image based on an image refinement algorithm, and removing a burr area and a fracture area in the crack skeleton by utilizing an image closing operation to obtain a target crack skeleton;
extracting pixel coordinates of the target crack skeleton, and calculating a linear characteristic fitting equation based on the pixel coordinates of the target crack skeleton;
and judging the linear characteristic fitting equation by using curvature to obtain a visual hazard area, and removing a linear interference area in the target fracture skeleton to obtain third visual image data.
9. The intelligent detection method for building exterior wall falling according to claim 5, wherein the performing the secondary judgment on the visual hazard area in the third visual image data to obtain hazard data comprises:
and performing secondary judgment on the visual hazard area in the third visual image data based on the infrared threat area so as to judge the difference area between the infrared threat area and the visual hazard area and obtain hazard data.
10. An intelligent detection system for building exterior wall falling off, which is characterized by comprising:
the flight path planning module is used for collecting surface image data of the building body, extracting external elevation features of the building body and formulating an aircraft path of the unmanned aerial vehicle based on the external elevation features of the building body;
the data acquisition module is used for controlling the unmanned aerial vehicle to fly according to the flight track and acquiring data to obtain outer-facade infrared image data and outer-facade visual image data;
the infrared thermal imaging analysis module is used for analyzing the outer elevation infrared image data based on an infrared thermal imaging analysis model, screening out an infrared threat area if threat exists, and determining infrared threat area position data;
the visual analysis module is used for analyzing and processing the outer-facade visual image data based on a visual analysis model, if danger exists, a visual danger area is screened out, the visual danger area excluding the interference area is subjected to secondary judgment based on the infrared threat area, the visual danger area obtains danger data, and the danger data comprises falling danger area data and falling danger area position data;
and the output module is used for outputting the dangerous data and sending out an early warning signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311607489.4A CN117710809A (en) | 2023-11-27 | 2023-11-27 | Intelligent detection method and system for building outer wall falling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311607489.4A CN117710809A (en) | 2023-11-27 | 2023-11-27 | Intelligent detection method and system for building outer wall falling |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117710809A true CN117710809A (en) | 2024-03-15 |
Family
ID=90154361
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311607489.4A Pending CN117710809A (en) | 2023-11-27 | 2023-11-27 | Intelligent detection method and system for building outer wall falling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117710809A (en) |
-
2023
- 2023-11-27 CN CN202311607489.4A patent/CN117710809A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105373135B (en) | A kind of method and system of aircraft docking guidance and plane type recognition based on machine vision | |
| CN103442209B (en) | Video monitoring method of electric transmission line | |
| CN106290388B (en) | A kind of insulator breakdown automatic testing method | |
| CN108230344B (en) | Automatic identification method for tunnel water leakage diseases | |
| CN110246130B (en) | Airport pavement crack detection method based on infrared and visible light image data fusion | |
| CN105158257B (en) | Slide plate measurement method and device | |
| CN100520362C (en) | Method for detecting forest fire fog based on colorful CCD image analysis | |
| KR101709751B1 (en) | An automatic monitoring system for dangerous situation of persons in the sea | |
| CN111126184B (en) | Post-earthquake building damage detection method based on unmanned aerial vehicle video | |
| CN107784652B (en) | Tower rapid detection method based on unmanned aerial vehicle image | |
| CN106056619A (en) | Unmanned aerial vehicle vision wire patrol method based on gradient constraint Radon transform | |
| CN112819094A (en) | Target detection and identification method based on structural similarity measurement | |
| EP2124194B1 (en) | Method of detecting objects | |
| CN109711256B (en) | Low-altitude complex background unmanned aerial vehicle target detection method | |
| CN111126183A (en) | Method for detecting damage of building after earthquake based on near-ground image data | |
| CN103679704A (en) | Video motion shadow detecting method based on lighting compensation | |
| CN105096305A (en) | Method and device for analyzing state of insulators | |
| CN110807765A (en) | Suspension insulator string inclination detection method and system based on image processing | |
| KR20200143149A (en) | Safety Inspection Check Method of Structure through Artificial Intelligence Analysis of Scan Image by Drone | |
| CN115035141A (en) | Tunnel crack remote monitoring and early warning method based on image processing | |
| CN113313107A (en) | Intelligent detection and identification method for multiple types of diseases on cable surface of cable-stayed bridge | |
| CN109521021A (en) | A kind of nuclear power plant containment shell appearance inspecting system and method | |
| CN113673614B (en) | Metro tunnel foreign matter intrusion detection device and method based on machine vision | |
| CN108460357B (en) | Windowing alarm detection system and method based on image recognition | |
| CN116309447B (en) | Dam slope crack detection method based on deep learning |
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 |