Old house roof plate crack detection method, system and device

Abstract

The invention relates to the technical field of image analysis, in particular to an old house roof plate crack detection method, system and device. The method comprises the following steps: acquiring an infrared thermal image of an old house roof plate; obtaining a suspected crack region according to the color channel values of the pixel points in the infrared thermal image; obtaining a reference crack area according to the position distribution of the suspected crack area; according to the size difference between the reference crack area and the suspected crack area corresponding to the reference crack area, the instability degree is obtained; dividing the reference crack region to obtain a reference category; obtaining a reasonable degree according to a pixel distribution similarity condition between the reference crack regions in the reference category; and detecting a real crack area of the roof plate of the old house according to the instability degree and the reasonability degree. According to the method, the instability degree and the reasonable degree are obtained, the crack area is accurately detected, the situation that a normal roof plate area is mistakenly recognized as the crack area is avoided, and the efficiency and accuracy of old house roof plate maintenance are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image analysis, and in particular to a method, system and device for detecting cracks in roof slabs of old buildings. Background Art

[0002] When detecting cracks in the roof slabs of old houses, since the thermal conductivity of the crack area is different from that of the normal roof slab, heat will accumulate in the crack area of ​​the roof slab to form a surface temperature gradient, which appears as an obvious bright area in the infrared thermal image. Therefore, in the existing method, the crack area of ​​the roof slab of the old house can be determined by obtaining the infrared thermal image of the roof slab of the old house and analyzing the infrared thermal image.

[0003] However, in actual situations, the roof slabs of old houses will be affected by factors such as the external environment and solar radiation. Due to the different solar radiation intensities in different time periods, the temperature distribution on the surface of the roof slabs of old houses will change. Under different solar radiation intensities, normal roof slab areas may also appear as obvious bright areas in infrared thermal images, resulting in inaccurate crack detection on the roof slabs of old houses, affecting the efficiency and accuracy of repairing the roof slabs of old houses, and wasting manpower and material resources. Summary of the invention

[0004] In order to solve the technical problem that the normal roof area of ​​old houses is misidentified as a crack area due to different solar radiation intensities, resulting in inaccurate crack detection of the roof of old houses, the purpose of the present invention is to provide a method, system and device for detecting cracks in the roof of old houses. The technical solutions adopted are as follows: In a first aspect, an embodiment of the present invention provides a method for detecting cracks in the roof of an old house, the method comprising the following steps: Obtain infrared thermal images of the roof of old buildings under different lighting conditions; According to the color channel value of each pixel point in each infrared thermal image, the suspected crack area in each infrared thermal image is obtained; All suspected crack areas in infrared thermal images are mapped to the same designated image, and suspected crack areas that overlap in the designated image but are not in the same infrared thermal image are constructed into a set. The areas where the suspected crack areas in each set overlap are all taken as target areas. According to the number of suspected crack areas corresponding to each target area, reference crack areas are screened out; according to the size difference between each reference crack area and each corresponding suspected crack area, the instability degree of each reference crack area is obtained; The reference crack regions are divided into reference categories based on the width of each reference crack region; the rationality of each reference crack region is obtained according to the similarity of pixel distribution of each reference crack region in each reference category with other reference crack regions; According to the instability and rationality of each reference crack area, the real crack area of ​​the roof slab of the old house is detected.

[0005] Furthermore, the method for obtaining the suspected crack area is: For any infrared thermal image, convert the infrared thermal image into a pseudo-color image; For any pixel in the pseudo-color image, obtaining a difference between a normalized red channel value and a normalized green channel value of the pixel as a first value; Obtain the difference between the normalized red channel value and the normalized blue channel value of the pixel as the second value; The sum of the first value and the second value is used as the red apparent value of the pixel; The result of normalizing the product of the normalized red channel value of the pixel and the red obvious value is taken as the red component degree of the pixel; Obtaining the red component degree of each pixel in the pseudo-color image, and when the red component degree is greater than a preset red component degree threshold, taking the corresponding pixel as a suspected crack pixel; The area formed by the suspected crack pixel points is used as the suspected crack area in the infrared thermal image.

[0006] Furthermore, the method for obtaining the reference crack area is: For any target area, the ratio of the number of suspected crack areas corresponding to the target area to the total number of infrared thermal images is used as the overlap degree of the target area; When the overlap degree is greater than a preset overlap degree threshold, the corresponding target area is used as a reference crack area.

[0007] Furthermore, the method for obtaining the instability degree is: For any reference crack region, obtain the area difference between the reference crack region and each corresponding suspected crack region, all of which are taken as the first difference; The mean of the first differences is taken as the instability degree of the reference crack region.

[0008] Furthermore, the method for obtaining the width of the reference crack area is: For any reference crack region, half of the number of edge pixels of the reference crack region is taken as the length of the reference crack region; The ratio of the area to the length of the reference crack region is taken as the width of the reference crack region.

[0009] Furthermore, the method for obtaining the reference category is: According to the width of each reference crack area, the reference crack area is divided by an adaptive clustering algorithm, and the divided categories are used as reference categories.

[0010] Furthermore, the method for obtaining the reasonable degree is: For any pixel point in any reference crack region within any reference category, the red component degree of each suspected crack region corresponding to the pixel point in the reference crack region is used as the reference red component degree; The mean value of the reference red component level is used as the target red component level of the pixel; Obtain the target red component level of each pixel in the reference crack area, and determine the target red component level range of the reference crack area; Obtain the overlapping length of the target red component degree range between the reference crack area and each other reference crack area in the reference category, and use them as the target length; The result of normalizing the mean value of the target length is taken as the reasonableness of the reference crack area.

[0011] Furthermore, the method for obtaining the real crack area is: According to the instability degree and rationality degree of each reference crack area, the true degree of each reference crack area is obtained; wherein the instability degree is negatively correlated with the true degree, and the rationality degree is positively correlated with the true degree; When the authenticity is greater than a preset authenticity threshold, the corresponding reference crack region is taken as the true crack region.

[0012] In a second aspect, another embodiment of the present invention provides a system for detecting cracks in roof slabs of old buildings, the system comprising: Infrared thermal image acquisition module, used to acquire infrared thermal images of roof panels of old houses under different lighting conditions; A suspected crack region acquisition module is used to acquire the suspected crack region in each infrared thermal image according to the color channel value of each pixel point in each infrared thermal image; The instability degree acquisition module is used to map all suspected crack areas in the infrared thermal images to the same designated image, construct the suspected crack areas that overlap in the designated image and are not in the same infrared thermal image into a set, and take the areas where the suspected crack areas in each set overlap as target areas, and screen out reference crack areas according to the number of suspected crack areas corresponding to each target area; and obtain the instability degree of each reference crack area according to the size difference between each reference crack area and each corresponding suspected crack area; A reasonable degree acquisition module is used to divide the reference crack area into reference categories based on the width of each reference crack area; and to acquire the reasonable degree of each reference crack area according to the similarity of pixel distribution between each reference crack area in each reference category and other reference crack areas; The real crack area acquisition module is used to detect the real crack area of ​​the roof slab of the old house according to the instability and rationality of each reference crack area.

[0013] In a third aspect, another embodiment of the present invention provides a device for detecting cracks in the roof slabs of old buildings, the device comprising: a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above methods are implemented.

[0014] The present invention has the following beneficial effects: The present invention first obtains the suspected crack area in each infrared thermal image according to the color channel value of each pixel point in each infrared thermal image, which is conducive to the subsequent accurate and efficient detection of crack areas in the roof slabs of old houses; considering that some normal roof slab areas are mistakenly identified as suspected crack areas due to the influence of light intensity, in order to avoid the subsequent misidentification of some normal roof slab areas as crack areas, the suspected crack areas in all infrared thermal images are mapped to the same designated image, and the suspected crack areas that overlap in the designated image and are not in the same infrared thermal image are constructed as a set, and the areas where the suspected crack areas in each set overlap are used as target areas, and the possible corresponding positions of the crack areas are preliminarily determined; in order to analyze the possibility that each target area is a crack area and avoid the normal roof slab area being mistakenly identified as a crack area, the reference crack area is screened out according to the number of suspected crack areas corresponding to each target area, and the corresponding position of the crack area is further determined, which is conducive to the subsequent The crack areas are detected more accurately; further, based on the size difference between each reference crack area and each corresponding suspected crack area, the instability degree of each reference crack area is obtained, accurately reflecting the possibility that each reference crack area is not a real crack area; in order to further analyze the possibility that each reference crack area is a real crack area, the reference crack areas are divided to obtain reference categories, and then based on the similarity of pixel distribution between each reference crack area and other reference crack areas in each reference category, the rationality of each reference crack area is obtained, accurately reflecting the possibility that each reference crack area is a real crack area; finally, based on the instability and rationality of each reference crack area, the degree to which each reference crack area is a real crack area is analyzed, and then the real crack areas of the roof slabs of old houses are accurately detected, effectively avoiding the normal roof slab areas from being misidentified as crack areas, improving the efficiency and accuracy of roof slab repair of old houses, and avoiding the waste of manpower and material resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A schematic flow chart of a method for detecting cracks in the roof of an old building provided by an embodiment of the present invention; Figure 2 A flow chart of a method for obtaining a suspected crack area provided by an embodiment of the present invention; Figure 3A flow chart of a method for obtaining a reasonable degree provided by an embodiment of the present invention; Figure 4 A structural diagram of a crack detection system for roof slabs of old buildings provided by one embodiment of the present invention; Figure 5 A schematic diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of a method, system and device for detecting cracks in the roof of an old house proposed by the present invention, its specific implementation method, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0018] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0019] The following is a detailed description of a method, system and device for detecting cracks in the roof of an old building provided by the present invention in conjunction with the accompanying drawings.

[0020] Embodiment 1: The present invention proposes a method for detecting cracks in the roof of an old house. Figure 1 , which shows a schematic flow chart of a method for detecting cracks in the roof of an old house provided by an embodiment of the present invention, the method comprising the following steps: Step S1: Obtain infrared thermal images of the roof of an old building under different lighting conditions.

[0021] It is known that the thermal conductivity of the crack area of ​​the roof of an old house is different from that of the normal area. Heat will accumulate in the crack area, and the temperature of the crack area is relatively high. Therefore, the crack area of ​​the roof of an old house is more obvious in the infrared thermal image obtained by the infrared thermal imager, that is, the bright area in the infrared thermal image. In actual situations, different solar radiation intensities will affect the temperature distribution in the analyzed roof area of ​​the old house. Under different solar radiation intensities, normal areas in different parts of the roof of the old house will also appear as bright areas in the infrared thermal image, resulting in inaccurate detection of the crack area of ​​the roof of the old house.

[0022] In order to accurately detect the crack area of ​​the roof slab of the old house, this embodiment first determines the roof slab area of ​​the old house to be analyzed, and then uses the infrared thermal imager carried by the drone to observe the roof slab area of ​​the old house to be analyzed at 1:00, 3:00, 5:00, 7:00, 9:00, 11:00, 13:00, 15:00, 17:00, 19:00, 21:00 and 23:00, and obtain 12 infrared thermal images. The implementer can set the time and number of infrared thermal images obtained by the infrared thermal imager according to actual conditions, which is not limited here. Among them, the spectral range of the infrared thermal imager is 7.5um-14um, the temperature sensitivity is 0.05℃, and the image resolution is 640×480 pixels. At this point, infrared thermal images of the roof slab of the old house under different lighting are obtained.

[0023] It is known that the crack area of ​​the roof of an old house must be a bright area in the infrared thermal image. In order to accurately detect the crack area in the roof of the old house to be analyzed later, this embodiment converts the infrared thermal image into a pseudo-color image. Since the color mapping table usually maps higher pixel values, i.e., pixels representing higher temperatures, to red, and red is visually associated with heat and high temperature, the bright area in the infrared thermal image will be displayed in red in the pseudo-color image, and then the red channel component corresponding to the high temperature area in the pseudo-color image is the largest, and the other color channel components are relatively small. It should be noted that the infrared thermal images that appear later are all defaulted to the converted pseudo-color images.

[0024] Step S2: Obtain the suspected crack area in each infrared thermal image according to the color channel value of each pixel in each infrared thermal image.

[0025] It is known that the temperature corresponding to the crack area is relatively high, and it must correspond to the red area in the pseudo-color image. Therefore, the component of the red channel value of the pixel corresponding to the crack area in the pseudo-color image is the largest. In addition, this embodiment decomposes the pseudo-color image into RGB three colors, determines the color channel values ​​of each pixel in each infrared thermal image, and obtains the proportion of the red channel value of each pixel by analyzing the color channel value of each pixel, thereby predicting the suspected crack pixel points and determining the suspected crack area in each infrared thermal image. Among them, the method of RGB three-color decomposition is a well-known technology and will not be repeated here. Therefore, this embodiment obtains the suspected crack area in each infrared thermal image according to the color channel value of each pixel in each infrared thermal image.

[0026] Preferably, in one possible implementation of this embodiment, the method for obtaining the suspected crack area can be found in Figure 2 , which shows a flow chart of a method for obtaining a suspected crack area provided in this embodiment, the method comprising the following steps: Step S201: Obtain the red component level.

[0027] For any pixel in any infrared thermal image, in order to accurately and efficiently analyze the red display of the pixel, this embodiment normalizes the values ​​of each color channel of the pixel, wherein this embodiment normalizes the values ​​of each color channel of the pixel by the maximum and minimum value normalization method. Then, the difference between the normalized red channel value of the pixel and the normalized green channel value is obtained as the first value; the difference between the normalized red channel value of the pixel and the normalized blue channel value is obtained as the second value; when the first value and the second value are both larger, it means that the red channel value of the pixel is more obvious, which indirectly reflects that the pixel is more likely to be a crack pixel, and then the addition result of the first value and the second value is used as the red obvious value of the pixel; in order to more accurately show the red display of the pixel, the product of the normalized red channel value of the pixel and the red obvious value is normalized as the red component degree of the pixel. It should be noted that this embodiment normalizes the product of the normalized red channel value of the pixel and the red obvious value by the norm normalization function. The greater the red component, the more likely the pixel is a crack pixel.

[0028] At this point, the red component level of each pixel in each infrared thermal image is obtained.

[0029] Step S202: Obtain suspected crack pixel points.

[0030] It is known that the greater the red component degree, the more likely the corresponding pixel point is a crack pixel point. Therefore, the preset red component degree threshold is set to 0.7 in this embodiment. The implementer can set the value of the preset red component degree threshold according to the actual situation, which is not limited here. When the red component degree is greater than the preset red component degree threshold, the corresponding pixel point is regarded as a suspected crack pixel point. So far, the suspected crack pixel points in each infrared thermal image are obtained.

[0031] Step S203: Acquire a suspected crack area.

[0032] For any infrared thermal image, the area formed by the suspected crack pixel points in the infrared thermal image is used as the suspected crack area in the infrared thermal image. So far, the suspected crack area in each infrared thermal image is determined.

[0033] Step S3: Map all suspected crack areas in the infrared thermal images to the same designated image, construct the suspected crack areas that overlap in the designated image and are not in the same infrared thermal image into a set, and use the areas where the suspected crack areas in each set overlap with each other as target areas. According to the number of suspected crack areas corresponding to each target area, select the reference crack areas; and obtain the instability degree of each reference crack area according to the size difference between each reference crack area and its corresponding suspected crack area.

[0034] In actual situations, due to the different light radiation intensities at different times, the heating conditions of the roof of old houses will be different at different times. At certain times, some normal areas of the roof of old houses have higher temperatures, and then some suspected crack areas in different infrared thermal images correspond to the normal areas of the roof. It is known that the position of the crack area of ​​the roof of old houses is fixed, and the temperature corresponding to the crack area must be relatively high. Therefore, the crack area must appear at the same position in all infrared thermal images, and the suspected crack area generated by the normal area of ​​the roof is less likely to remain at the same position in other infrared thermal images. Therefore, this embodiment maps the suspected crack areas in all infrared thermal images to the same designated image, and constructs the suspected crack areas that overlap in the designated image and are not in the same infrared thermal image into a set, wherein there must be overlapping areas between the suspected crack areas in the same set. The areas where the suspected crack areas in each set overlap are all taken as target areas.

[0035] When a target area corresponds to more suspected crack areas, it means that the target area appears as a crack area more often under different solar radiation intensities, which indirectly reflects that the target area is more likely to be a crack area on the roof of an old house. In addition, this embodiment selects reference crack areas that may be crack areas from the target area according to the number of suspected crack areas corresponding to each target area.

[0036] It is known that cracks will store heat under different solar radiation intensities, so the position of the bright area corresponding to the same crack area in the infrared thermal images obtained at different times is the same. Even if there is an impact of changes in solar radiation intensity, the bright area corresponding to the same crack, that is, the suspected crack area, has almost no position change. However, the temperature of the normal roof area changes under different solar radiation intensities, and the size of the suspected crack area formed in the same normal roof area will be significantly different in different infrared thermal images. In order to more accurately detect the crack area of ​​the roof of old houses, this embodiment obtains the degree of instability of each reference crack area based on the size difference between each reference crack area and each corresponding suspected crack area. Among them, the greater the degree of instability, the less likely the corresponding reference crack area is to be a crack area, and the more likely it is to be formed in the normal roof area due to the impact of solar radiation intensity.

[0037] Preferably, in a method that can be implemented in this embodiment, the method for obtaining the reference crack area is: for any target area, the more times the target area appears in different infrared thermal images, the more likely the target area is to be a crack area. It is known that the suspected crack area corresponding to the target area is in different infrared thermal images, and then this embodiment uses the ratio of the number of suspected crack areas corresponding to the target area to the total number of infrared thermal images as the overlap degree of the target area; it should be noted that the suspected crack area corresponding to the target area is the suspected crack area of ​​the target area constructed by overlapping, that is, the suspected crack area corresponding to the target area is a set. So far, the overlap degree of each target area is obtained. Among them, the greater the overlap degree, the more times the corresponding target area appears in different infrared thermal images, and the more likely the corresponding target area is to be a crack area; and then this embodiment sets the preset overlap degree threshold to 0.7, and the implementer can set the size of the preset overlap degree threshold according to the actual situation, which is not set here. When the overlap degree is greater than the preset overlap degree threshold, the corresponding target area is used as the reference crack area. So far, all reference crack areas are screened out.

[0038] Preferably, in a method that can be implemented in this embodiment, the method for obtaining the degree of instability is: for any reference crack area, the absolute value of the difference between the area of ​​the reference crack area and each suspected crack area corresponding to it is obtained, and all are taken as the first difference; the larger the first difference, the more inconsistent the size of the suspected crack areas in the same set corresponding to the reference crack area, reflecting that the reference crack area is less likely to be a crack area on the roof of an old house. In order to determine the possibility that the reference crack area is not a crack area, the mean of the first difference is taken as the degree of instability of the reference crack area. The greater the degree of instability, the less likely the reference crack area is to be a real crack area. Among them, the method for obtaining the area of ​​the reference crack area and the suspected crack area is a well-known technology and will not be described in detail.

[0039] At this point, the instability degree of each reference crack area is obtained.

[0040] Step S4: Divide the reference crack areas into reference categories based on the width of each reference crack area; obtain the rationality of each reference crack area according to the similarity of pixel distribution between each reference crack area in each reference category and other reference crack areas.

[0041] It is known that cracks of different widths gather different amounts of heat, so cracks of different widths display different degrees of red in pseudo-color images. Considering the possibility that the reference crack area is a normal roof area of ​​an old house, in order to accurately detect the crack area, this embodiment first divides the reference crack area based on the width of each reference crack area to obtain a reference category; it is known that the red display degree of the reference crack area corresponding to cracks of similar widths in the pseudo-color image is similar, and there is a deviation in the red display degree of the reference crack area corresponding to the normal roof area in the pseudo-color image. Therefore, this embodiment obtains the reasonableness of each reference crack area based on the similarity of the pixel distribution of each reference crack area in each reference category with other reference crack areas. The greater the reasonableness, the more likely the corresponding reference crack area is to be a real crack area.

[0042] Preferably, in one possible implementation of this embodiment, the method for obtaining the reasonable degree can be found in Figure 3 , which shows a flow chart of a method for obtaining a reasonable degree provided in this embodiment, the method comprising the following steps: Step S301: Obtain the width of a reference crack area.

[0043] In actual situations, the reference crack regions are irregular and the width of the reference crack regions may be uneven. Therefore, this embodiment calculates the width of each reference crack region based on the area and length of each reference crack region.

[0044] In one possible implementation of this embodiment, the width of the reference crack region is obtained by: for any reference crack region, half of the number of edge pixels of the reference crack region is used as the length of the reference crack region; it should be noted that the length of the reference crack region is an approximate length. The ratio of the area to the length of the reference crack region is used as the width of the reference crack region. The method for obtaining edge pixels is a well-known technique and will not be described in detail.

[0045] Step S302: Obtain reference categories.

[0046] The width of each reference crack region is obtained, and then the reference crack region is divided according to the width of each reference crack region by an adaptive clustering algorithm, and the divided categories are all used as reference categories. The adaptive clustering algorithm is a well-known technology and will not be described in detail. It should be noted that the width of the reference crack region in each reference category is similar.

[0047] Step S303: Obtaining the rationality.

[0048] For any reference crack region within any reference category, the more similar the red display degree of the reference crack region in the pseudo-color image is to that of other reference crack regions within the reference category, the more likely the reference crack region is to be a real crack region.

[0049] In order to obtain the red display situation corresponding to each reference crack area, for any pixel point in any reference crack area within any reference category, the red component degree of the pixel point in each suspected crack area corresponding to the reference crack area is used as the reference red component degree; the mean of the reference red component degrees is used as the target red component degree of the pixel point; the target red component degree of each pixel point in the reference crack area is obtained, and the target red component degree range of the reference crack area is determined, that is, the range formed by the minimum target red component degree and the maximum target red component degree in the reference crack area.

[0050] Further obtain the overlapping length of the target red component degree range between the reference crack area and each other reference crack area in the reference category, and use them as the target length; the larger the target length, the more similar the red display degree of the reference crack area is to the corresponding other reference crack area, which indirectly reflects that the reference crack area is more likely to be a real crack area. In order to accurately indicate the possibility that the reference crack area is a real crack area, the mean of the target length is normalized as the reasonable degree of the reference crack area. The greater the reasonable degree, the more likely the reference crack area is to be a real crack area. It should be noted that in this embodiment, the mean of the target length is normalized by the norm normalization function.

[0051] At this point, the reasonableness of each reference crack area is obtained.

[0052] Step S5: Detect the real crack area of ​​the roof slab of the old house according to the instability and rationality of each reference crack area.

[0053] It is known that the greater the instability, the less likely the corresponding reference crack region is to be a real crack region; the greater the rationality, the more likely the corresponding reference crack region is to be a real crack region. Furthermore, this embodiment detects the real crack region of the roof slab of an old house according to the instability and rationality of each reference crack region.

[0054] Preferably, in an achievable manner of this embodiment, the method for obtaining the real crack region is: according to the instability degree and rationality degree of each reference crack region, the real degree of each reference crack region is obtained; wherein the instability degree is negatively correlated with the real degree, and the rationality degree is positively correlated with the real degree; wherein the calculation formula of the real degree is: ; In the formula, is the true extent of the vth reference crack area; is the instability degree of the vth reference crack region; is the reasonableness of the vth reference crack area; is the first preset constant, which is greater than 0; norm is the normalization function. Set to 1 to avoid the denominator being 0. Implementers can set it according to actual conditions. The size of is not limited here; The greater the degree of authenticity, the more likely the corresponding reference crack region is a real crack region; therefore, this embodiment sets the preset degree of authenticity threshold to 0.8, and the implementer can set the value of the preset degree of authenticity threshold according to the actual situation, which is not limited here. When the degree of authenticity is greater than the preset degree of authenticity threshold, the corresponding reference crack region is regarded as a real crack region.

[0055] At this point, the crack detection of the roof slabs of old houses has been accurately detected, which effectively avoids the normal roof slab area from being misidentified as a crack area, which is conducive to maintenance personnel to carry out maintenance accurately and efficiently, avoiding the waste of manpower and material resources.

[0056] In summary, the present embodiment obtains an infrared thermal image of the roof of an old house; obtains a suspected crack area according to the color channel value of a pixel point in the infrared thermal image; obtains a reference crack area according to the position distribution of the suspected crack area; obtains the degree of instability according to the size difference between the reference crack area and the suspected crack area corresponding to it; divides the reference crack area to obtain a reference category; obtains a reasonable degree according to the similarity of pixel distribution between the reference crack areas in the reference category; detects the real crack area of ​​the roof of the old house according to the degree of instability and the reasonable degree. The present invention accurately detects the crack area by obtaining the degree of instability and the reasonable degree, avoids the normal roof area from being misidentified as a crack area, and improves the efficiency and accuracy of the maintenance of the roof of the old house.

[0057] Embodiment 2: The present invention also proposes a crack detection system for the roof of an old house, please refer to Figure 4 , which shows a structural diagram of a crack detection system for roof slabs of old houses provided by an embodiment of the present invention, the system includes: an infrared thermal image acquisition module 10, a suspected crack area acquisition module 20, an instability degree acquisition module 30, a reasonable degree acquisition module 40 and a real crack area acquisition module 50.

[0058] The infrared thermal image acquisition module 10 is used to acquire infrared thermal images of the roof of the old building under different lighting conditions.

[0059] The suspected crack region acquisition module 20 is used to acquire the suspected crack region in each infrared thermal image according to the color channel value of each pixel point in each infrared thermal image.

[0060] The instability degree acquisition module 30 is used to map the suspected crack areas in all infrared thermal images to the same designated image, construct the suspected crack areas that overlap in the designated image and are not in the same infrared thermal image into a set, and use the areas where the suspected crack areas in each set overlap with each other as target areas, and screen out reference crack areas according to the number of suspected crack areas corresponding to each target area; and obtain the instability degree of each reference crack area according to the size difference between each reference crack area and each corresponding suspected crack area.

[0061] The reasonableness acquisition module 40 is used to divide the reference crack areas into reference categories based on the width of each reference crack area; and obtain the reasonableness of each reference crack area according to the similarity of pixel distribution between each reference crack area in each reference category and other reference crack areas.

[0062] The real crack region acquisition module 50 is used to detect the real crack region of the roof slab of the old house according to the instability and rationality of each reference crack region.

[0063] It should be noted that the system provided in the above embodiment is only illustrated by the division of the above functional modules. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device is divided into different functional modules to complete all or part of the functions described above. In addition, the old house roof crack detection system and the old house roof crack detection method embodiment provided in the above embodiment belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0064] Embodiment 3: The present invention also proposes a device for detecting cracks in the roof of an old house, the device comprising a memory and a processor, wherein the memory stores an executable program code, and the processor is used to call and execute the executable program code to execute a method for detecting cracks in the roof of an old house provided in an embodiment of the present application. The device can be a chip, a component or a module, and the chip can include a connected processor and a memory; wherein the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute a method for detecting cracks in the roof of an old house provided in the above embodiment.

[0065] In addition, the present application embodiment also protects a computer device, see Figure 5 The computer device includes a memory 401, a processor 402, and a computer program 403 stored in the memory 401 and running on the processor 402, wherein when the processor 402 executes the computer program 403, the computer device can execute any of the above-mentioned methods for detecting cracks in the roof of an old house.

[0066] Embodiment 4: This embodiment also provides a computer-readable storage medium, in which a computer program code is stored. When the computer program code is run on a computer, the computer executes the above-mentioned related method steps to implement a method for detecting cracks in the roof of an old house provided in the above embodiment.

[0067] Embodiment 5: This embodiment also provides a computer program product. When the computer program product is run on a computer, the computer is caused to execute the above-mentioned related steps to implement a method for detecting cracks in the roof slab of an old house provided in the above-mentioned embodiment.

[0068] Among them, the device, computer-readable storage medium, computer program product or chip provided in this embodiment is used to execute the corresponding method provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method provided above, and will not be repeated here.

[0069] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0070] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for detecting cracks in the roof of an old house, characterized in that: The method comprises the following steps: Obtain infrared thermal images of the roof of old buildings under different lighting conditions; According to the color channel value of each pixel point in each infrared thermal image, the suspected crack area in each infrared thermal image is obtained; All suspected crack areas in infrared thermal images are mapped to the same designated image, and suspected crack areas that overlap in the designated image but are not in the same infrared thermal image are constructed into a set. The areas where the suspected crack areas in each set overlap are all taken as target areas. According to the number of suspected crack areas corresponding to each target area, reference crack areas are screened out; according to the size difference between each reference crack area and each corresponding suspected crack area, the instability degree of each reference crack area is obtained; The reference crack regions are divided into reference categories based on the width of each reference crack region; the rationality of each reference crack region is obtained according to the similarity of pixel distribution of each reference crack region in each reference category with other reference crack regions; According to the instability and rationality of each reference crack area, the real crack area of ​​the roof slab of the old house is detected.

2. A method for detecting cracks in the roof of an old house as claimed in claim 1, characterized in that: The method for obtaining the suspected crack area is: For any infrared thermal image, convert the infrared thermal image into a pseudo-color image; For any pixel in the pseudo-color image, obtaining a difference between a normalized red channel value and a normalized green channel value of the pixel as a first value; Obtain the difference between the normalized red channel value and the normalized blue channel value of the pixel as the second value; The sum of the first value and the second value is used as the red apparent value of the pixel; The result of normalizing the product of the normalized red channel value of the pixel and the red obvious value is taken as the red component degree of the pixel; Obtaining the red component degree of each pixel in the pseudo-color image, and when the red component degree is greater than a preset red component degree threshold, taking the corresponding pixel as a suspected crack pixel; The area formed by the suspected crack pixel points is used as the suspected crack area in the infrared thermal image.

3. A method for detecting cracks in roof slabs of old buildings as claimed in claim 1, characterized in that: The method for obtaining the reference crack area is: For any target area, the ratio of the number of suspected crack areas corresponding to the target area to the total number of infrared thermal images is used as the overlap degree of the target area; When the overlap degree is greater than a preset overlap degree threshold, the corresponding target area is used as a reference crack area.

4. A method for detecting cracks in roof slabs of old buildings as claimed in claim 1, characterized in that: The method for obtaining the instability degree is: For any reference crack region, obtain the area difference between the reference crack region and each corresponding suspected crack region, all of which are taken as the first difference; The mean of the first differences is taken as the instability degree of the reference crack region.

5. A method for detecting cracks in roof slabs of old buildings as claimed in claim 1, characterized in that: The method for obtaining the width of the reference crack area is: For any reference crack region, half of the number of edge pixels of the reference crack region is taken as the length of the reference crack region; The ratio of the area to the length of the reference crack region is taken as the width of the reference crack region.

6. A method for detecting cracks in roof slabs of old buildings as claimed in claim 1, characterized in that: The method for obtaining the reference category is: According to the width of each reference crack area, the reference crack area is divided by an adaptive clustering algorithm, and the divided categories are used as reference categories.

7. A method for detecting cracks in roof slabs of old buildings as claimed in claim 2, characterized in that: The method for obtaining the reasonable degree is: For any pixel point in any reference crack region within any reference category, the red component degree of each suspected crack region corresponding to the pixel point in the reference crack region is used as the reference red component degree; The mean value of the reference red component level is used as the target red component level of the pixel; Obtain the target red component level of each pixel in the reference crack area, and determine the target red component level range of the reference crack area; Obtain the overlapping length of the target red component degree range between the reference crack area and each other reference crack area in the reference category, and use them as the target length; The result of normalizing the mean value of the target length is taken as the reasonableness of the reference crack area.

8. A method for detecting cracks in roof slabs of old buildings as claimed in claim 1, characterized in that: The method for obtaining the real crack area is: According to the instability degree and rationality degree of each reference crack area, the true degree of each reference crack area is obtained; wherein the instability degree is negatively correlated with the true degree, and the rationality degree is positively correlated with the true degree; When the authenticity is greater than a preset authenticity threshold, the corresponding reference crack region is taken as the true crack region.

9. A crack detection system for roof slabs of old buildings, characterized in that: The system comprises: Infrared thermal image acquisition module, used to acquire infrared thermal images of roof panels of old houses under different lighting conditions; A suspected crack region acquisition module is used to acquire the suspected crack region in each infrared thermal image according to the color channel value of each pixel point in each infrared thermal image; The instability degree acquisition module is used to map all suspected crack areas in the infrared thermal images to the same designated image, construct the suspected crack areas that overlap in the designated image and are not in the same infrared thermal image into a set, and take the areas where the suspected crack areas in each set overlap as target areas, and screen out reference crack areas according to the number of suspected crack areas corresponding to each target area; and obtain the instability degree of each reference crack area according to the size difference between each reference crack area and each corresponding suspected crack area; A reasonable degree acquisition module is used to divide the reference crack area into reference categories based on the width of each reference crack area; and to acquire the reasonable degree of each reference crack area according to the similarity of pixel distribution between each reference crack area in each reference category and other reference crack areas; The real crack area acquisition module is used to detect the real crack area of ​​the roof slab of the old house according to the instability and rationality of each reference crack area.

10. A device for detecting cracks in the roof of an old house, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When executing the computer program, the processor implements the steps of a method for detecting cracks in the roof slab of an old house as described in any one of claims 1 to 8.

Images