Highway pavement quality detection method and system based on machine vision

Abstract

The invention relates to the technical field of multi-image analysis, in particular to a highway pavement quality detection method and system based on machine vision. The method comprises the following steps: acquiring a brightness crack probability according to the gray difference of the same suspected crack area in road surface gray images before and after watering, and acquiring a temperature crack probability according to the pixel value difference and environmental data of the same suspected crack area in road surface infrared images before and after watering; and adjusting the brightness crack probability and the temperature crack probability according to the significance degree of the pixel value difference of the same suspected crack region in the pavement feature image before and after watering, obtaining the final crack probability of the suspected crack region, and performing pavement quality detection on the to-be-detected road by using the final crack probability. According to the method, the brightness change characteristics and the temperature change characteristics of the cracks before and after watering are combined, the degree of dependence on illumination is reduced, and the accuracy of pavement crack detection is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-image analysis, and particularly to a method and system for detecting the quality of highway pavement based on machine vision. Background Art

[0002] Pavement cracks may cause steel bar corrosion and base softening, and ultimately lead to pavement collapse or bearing failure. Highway pavement quality detection is a core technical means to ensure the safety, durability and economic operation of roads. Existing methods detect pavement cracks from visible light images of the pavement based on features such as crack trends and widths. However, in low-light conditions and pavements with complex texture structures, factors such as insufficient lighting and pavement texture structures are likely to result in low accuracy in identifying crack features in visible light images, thereby reducing the accuracy of highway pavement quality detection. Summary of the Invention

[0003] In order to solve the technical problem of low accuracy in crack detection in visible light images caused by insufficient lighting and pavement texture structures, etc., the purpose of the present invention is to provide a method and system for detecting the quality of highway pavement based on machine vision, and the specific technical solutions adopted are as follows: In a first aspect, an embodiment of the present invention provides a method for detecting the quality of highway pavement based on machine vision, and the method includes: Obtain pavement feature images of the highway to be measured before and after watering, as well as environmental data at the location of the highway to be measured; the pavement feature images include pavement grayscale images and pavement infrared images; Obtain suspected crack regions in the pavement feature images; according to the grayscale difference of the same suspected crack region in the pavement grayscale images before and after watering, obtain the brightness crack probability of each suspected crack region; According to the pixel value difference of the same suspected crack region in the pavement infrared images before and after watering and the environmental data, obtain the temperature crack probability of each suspected crack region; Adjust the brightness crack probability and the temperature crack probability according to the significance degree of the pixel value difference of the same suspected crack region in the pavement feature images before and after watering, obtain the final crack probability of each suspected crack region; use the final crack probability to detect the pavement quality of the highway to be measured.

[0004] Further, the obtaining of the brightness crack probability of each suspected crack region includes: Take the average pixel value of all pixel points in the suspected crack region as the overall pixel value; Calculate the absolute value of the difference between the overall pixel values of the same suspected crack region in the pavement feature images before and after watering, and obtain the pixel difference of each suspected crack region in the pavement feature images; Normalize the pixel differences of each suspected crack region in the road surface grayscale image to obtain the brightness crack probability of each suspected crack region.

[0005] Further, the obtaining of the temperature crack probability of each suspected crack region includes: The environmental data includes temperature data and wind speed data; according to the temperature data and wind speed data, obtain the environmental influence value; Use the environmental influence value to adjust the pixel differences of each suspected crack region in the road surface infrared image to obtain the actual temperature difference of each suspected crack region; normalize the actual temperature difference to obtain the temperature crack probability of each suspected crack region.

[0006] Further, the obtaining of the final crack probability of each suspected crack region includes: Obtain the feature weight coefficient according to the significance degree of the pixel value difference of the same suspected crack region in the road surface feature images before and after watering; Respectively use the feature weight coefficient, the difference between the constant 1 and the feature weight coefficient as the weights of the temperature crack probability and the brightness crack probability of each suspected crack region, and perform weighted summation on the temperature crack probability and the brightness crack probability of each suspected crack region to obtain the final crack probability of each suspected crack region.

[0007] Further, the method for obtaining the feature weight coefficient includes: Take the ratio of the sum of the mean, standard deviation and a preset positive number of the pixel differences of all suspected crack regions in the road surface feature image as the feature significance; respectively record the corresponding feature significances of the road surface grayscale image and the road surface infrared image as the grayscale feature significance and the infrared feature significance; Take the ratio with the infrared feature significance as the numerator and the sum value of the infrared feature significance and the grayscale feature significance as the denominator as the feature weight coefficient.

[0008] Further, the obtaining of the suspected crack regions of the road surface feature image includes: Use the maximum inter-class variance method to obtain the segmentation threshold for the grayscale values of the pixel points in the road surface grayscale image before watering, and take the connected domain composed of the pixel points with grayscale values less than the segmentation threshold as the suspected crack regions of the road surface grayscale image before watering; Record the road surface infrared images before and after watering and the road surface grayscale image after watering as the analysis images; take the connected domain composed of the pixel points at the same coordinate positions in the analysis images for the pixel points in each suspected crack region of the road surface grayscale image before watering as the suspected crack regions of the analysis images.

[0009] Further, the obtaining of the actual temperature difference of each suspected crack region includes: Perform negative correlation and normalization processing on the environmental impact value, and use the processing result to weight the pixel difference of each suspected crack area in the road surface infrared image to obtain the actual temperature difference of each suspected crack area.

[0010] Furthermore, both the temperature data and the wind speed data are positively correlated with the environmental impact value.

[0011] Furthermore, the pixel points in every two road surface feature images correspond one by one.

[0012] In a second aspect, another embodiment of the present invention provides a highway pavement quality detection system based on machine vision, and the system includes: A data acquisition module, configured to respectively acquire road surface feature images of a to-be-detected highway before and after watering, and environmental data of the position of the to-be-detected highway; the road surface feature images include a road surface gray-scale image and a road surface infrared image; A brightness feature analysis module, configured to acquire suspected crack areas in the road surface feature images; and acquire the brightness crack probability of each suspected crack area according to the gray-scale difference of the same suspected crack area in the road surface gray-scale images before and after watering; A temperature feature analysis module, configured to acquire the temperature crack probability of each suspected crack area according to the pixel value difference of the same suspected crack area in the road surface infrared images before and after watering and the environmental data; A road surface quality detection module, configured to adjust the brightness crack probability and the temperature crack probability according to the significance degree of the pixel value difference of the same suspected crack area in the road surface feature images before and after watering, and acquire the final crack probability of each suspected crack area; and perform road surface quality detection on the to-be-detected highway by using the final crack probability.

[0013] The present invention has the following beneficial effects: In the first aspect: Compared with the traditional method of detecting cracks from the visible light image of the road surface based on features such as crack trend and width, this solution is based on the significant differences in the brightness change and temperature change between the road surface and the cracks after watering, combines the brightness change feature and the temperature change feature, and improves the accuracy of road surface crack detection through multi-dimensional data cross-validation.

[0014] In the second aspect: This solution combines the brightness change and the temperature change for crack detection. Compared with the traditional method of directly detecting cracks according to the visible light image, this solution additionally considers the temperature change, has a lower dependence on the lighting conditions, and can still accurately capture the road surface crack features under low lighting conditions and complex road surface texture structures, thereby effectively avoiding the occurrence of missed detection phenomena and improving the accuracy of road surface crack detection.

[0015] Third aspect: This solution uses thermal imaging technology to analyze the temperature changes at the same location before and after watering. Thermal imaging technology can penetrate surface pollutants such as dust and oil stains, directly reflect the thermal characteristic changes of road cracks caused by water evaporation, obtain more real road condition information, and improve the accuracy of road crack detection.

[0016] Fourth aspect: When analyzing the thermal characteristic changes of road cracks caused by water evaporation, the influence of environmental factors on the thermal characteristic changes caused by water evaporation is considered simultaneously, so as to improve the authenticity of the road condition. Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0018] Figure 1 It is a flowchart of the steps of a method for detecting the quality of highway pavement based on machine vision provided by an embodiment of the present invention; Figure 2 It is a flowchart of a method for obtaining the final crack probability provided by an embodiment of the present invention; Figure 3 It is a system structure diagram of a system for detecting the quality of highway pavement based on machine vision provided by an embodiment of the present invention; Figure 4 It is a schematic diagram of a computer device of a device for detecting the quality of highway pavement based on machine vision provided by an embodiment of the present invention. Detailed Embodiments

[0019] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following will, in conjunction with the accompanying drawings and preferred embodiments, describe in detail the specific embodiments, structures, features and effects of a method and system for detecting the quality of highway pavement based on machine vision proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0020] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0021] Specific scenarios targeted by the present invention: applicable to the pavement quality inspection of traffic infrastructure such as highways, urban arterial roads, bridges, tunnels, etc., especially for the rapid detection of pavement damages such as pavement aging and crack propagation caused by long-term exposure to complex environments, such as large temperature differences, high traffic flow, and frequent rain and snow erosion.

[0022] The following specifically describes the specific solution of a highway pavement quality inspection method and system based on machine vision provided by the present invention with reference to the accompanying drawings.

[0023] Embodiment 1

[0024] The present invention proposes a highway pavement quality inspection method based on machine vision. Please refer to Figure 1 , which shows the step flowchart of a highway pavement quality inspection method based on machine vision provided by an embodiment of the present invention. The method includes: Step S1: Obtain the pavement feature images of the highway to be measured before and after watering, as well as the environmental data at the position of the highway to be measured; the pavement feature images include pavement grayscale images and pavement infrared images.

[0025] Integrate an industrial camera and an infrared camera on the unmanned aerial vehicle (UAV), control the UAV to fly above the highway to be measured and keep the position of the UAV unchanged. Before the road sweeping and watering vehicle passes by, use the industrial camera and the infrared camera to collect pavement images and pavement infrared images in sequence. During the process of the watering vehicle passing by the highway to be measured and spraying water on the pavement, after one minute when the watering vehicle has passed, use the industrial camera and the infrared camera again to collect pavement images and pavement infrared images. Perform grayscale conversion and denoising processing on the pavement images before and after watering to obtain the corresponding pavement grayscale images. At the same time, perform denoising processing on the pavement infrared images before and after watering. The pavement grayscale images and pavement infrared images are collectively referred to as pavement feature images. Herein, one minute can also be two minutes, and the implementer can set it according to the specific situation, and it is necessary to collect the pavement feature images after watering before the pavement moisture has completely evaporated.

[0026] During the time period from the end moment when the watering vehicle finishes spraying water on the pavement of the highway to be measured to the moment of collecting the pavement feature images after watering, use a handheld portable weather station to collect the ambient temperature and ambient wind speed at each moment at the position of the highway to be measured, and use the average value of the ambient temperature and the average value of the ambient wind speed at all moments during this time period as temperature data and wind speed data respectively, which are collectively referred to as environmental data. Among them, the data collection frequency of the handheld portable weather station is set to once every 10 seconds.

[0027] It should be noted that the UAV is in the same position when collecting the road surface feature images before and after watering, so as to ensure that the pixel points in the road surface feature images before and after watering correspond one by one. At the same time, it is necessary to ensure that the pixel points in the road surface grayscale image before or after watering correspond one by one with the pixel points in the road surface infrared image, so that the pixel points in every two road surface feature images correspond one by one. In the embodiment of the present invention, the weighted average graying algorithm is selected for graying processing, and Gaussian filtering is used for denoising processing. The specific methods are not introduced here and are all technical means well-known to those skilled in the art.

[0028] Step S2: Obtain the suspected crack areas in the road surface feature images; according to the gray difference of the same suspected crack area in the road surface grayscale images before and after watering, obtain the brightness crack probability of each suspected crack area.

[0029] Although road cracks can be shown in visible light images, due to the complexity of the road surface texture structure, it is difficult to accurately judge cracks in visible light images. The suspected crack areas can be initially extracted from the road surface feature images first to reduce the subsequent calculation amount. Before the road surface is watered, the brightness difference between the cracks and the road surface in the visible light image may be small; but after the road surface is watered, the water remaining in the cracks forms a relatively closed environment, the propagation path of light in the cracks becomes longer, and the absorption and scattering increase, resulting in a decrease in the brightness of the cracks, while the road surface brightness remains basically unchanged due to the rapid drainage and rapid evaporation of water. Therefore, the brightness difference between the cracks and the road surface before and after watering is greater. According to the gray difference of the same suspected crack area in the road surface grayscale images before and after watering, analyze the possibility that the suspected crack area is an actual crack through the brightness change of the suspected crack area before and after watering, and obtain the brightness crack probability.

[0030] Step S3: According to the pixel value difference and environmental data of the same suspected crack area in the road surface infrared images before and after watering, obtain the temperature crack probability of each suspected crack area.

[0031] After the road surface is watered, the water quickly penetrates into the cracks and stays inside, while the water on the road surface quickly flows away due to the action of gravity. Since the evaporation process of the water inside the cracks requires a large amount of heat absorption, and the evaporation process of the small amount of water remaining on the road surface absorbs less heat, the temperature difference between the cracks and the road surface before and after watering is greater. Considering that external environmental changes such as environmental temperature and wind speed will affect the water evaporation process, resulting in errors in the temperature difference at the same position before and after watering and affecting the accuracy of crack judgment. Therefore, for the pixel value difference of the same suspected crack area in the road surface infrared images before and after watering, analyze the possibility that the suspected crack area is an actual crack through the temperature change of the suspected crack area before and after watering, and use the environmental data for adjustment to obtain the temperature crack probability.

[0032] Step S4: Adjust the brightness crack probability and the temperature crack probability according to the significant degree of the pixel value difference of the same suspected crack area in the road surface characteristic images before and after watering, and obtain the final crack probability of each suspected crack area; use the final crack probability to perform road surface quality detection on the road to be measured.

[0033] There are various types of road surface cracks. Different types of cracks have different performances on brightness information and temperature information. Brightness information is sensitive to surface texture, and temperature information reflects the internal thermal characteristics of cracks. Cracks have greater brightness differences and temperature differences before and after watering compared to the road surface. Using the significant degree of the pixel value difference of the same suspected crack area in the road surface characteristic images before and after watering, that is, the significant degree of temperature change and the significant degree of brightness change, to determine the importance of the road surface gray-scale image and the road surface infrared image in the crack detection process, dynamically adjust the ratio of the brightness crack probability and the temperature crack probability, realize the dynamic fusion of the temperature difference information and the brightness information before and after watering, and obtain the final crack probability of the suspected crack area, so as to improve the accuracy of road surface crack detection.

[0034] Preferably, in some possible implementation manners of the embodiments of the present invention, the method for obtaining the suspected crack area includes: obtaining a segmentation threshold for the gray-scale values of the pixel points in the road surface gray-scale image before watering by using the maximum inter-class variance method, and taking the connected domain composed of the pixel points with gray-scale values less than the segmentation threshold as the suspected crack area of the road surface gray-scale image before watering; recording the road surface infrared images before and after watering and the road surface gray-scale image after watering as analysis images; taking the connected domain composed of the pixel points at the same coordinate positions in the analysis images of the pixel points in each suspected crack area of the road surface gray-scale image before watering as the suspected crack area of the analysis images. Among them, the maximum inter-class variance method is a well-known technology to those skilled in the art and will not be introduced here.

[0035] It should be noted that after watering, the moisture seeps into the cracks and changes the optical characteristics of the crack area, and the infrared image is easily affected by external factors such as environmental temperature and wind speed. Since the brightness information of the gray-scale image directly reflects the road surface texture change, the suspected crack area is determined by using the road surface gray-scale image before watering. Cracks form micro-depressions due to structural fractures, the incident light is absorbed or scattered in other directions, and the light intensity reflected back to the camera is weakened, resulting in a lower gray-scale of the cracks than that of the road surface in the road surface gray-scale image. Then, the pixel points with gray-scale values less than the segmentation threshold are selected to form the suspected crack area. At the same time, the pixel points in the road surface gray-scale image before watering correspond one by one to the pixel points in the other three road surface characteristic images, so the suspected crack area in the road surface gray-scale image before watering can be mapped to the other three road surface characteristic images, thereby obtaining the suspected crack areas of the three road surface characteristic images.

[0036] It should be noted that since the pixel points in every two pavement feature images correspond one by one, each suspected crack area in the pavement feature image refers to the suspected crack area at the same position in the pavement gray-scale image before watering.

[0037] Preferably, in some possible implementation manners of the embodiments of the present invention, the method for obtaining the brightness crack probability includes: taking the average pixel value of all pixel points in the suspected crack area as the overall pixel value; calculating the absolute value of the difference between the overall pixel values of the same suspected crack area in the pavement feature images before and after watering to obtain the pixel difference of each suspected crack area in the pavement feature image; normalizing the pixel difference of each suspected crack area in the pavement gray-scale image to obtain the brightness crack probability of each suspected crack area.

[0038] It should be noted that it is known that the brightness difference of cracks is greater than that of the pavement before and after watering. The overall pixel value reflects the overall gray level of the suspected crack area in the pavement gray-scale image or the overall temperature level of the suspected crack area in the pavement infrared image; if the pixel difference of the suspected crack area in the pavement feature image is larger, it indicates that the brightness difference of the suspected crack area before and after watering is greater, then the possibility that the suspected crack area is identified as an actual crack through the brightness change characteristics before and after watering is greater, and the brightness crack probability is greater. In the embodiments of the present invention, the Norm function is used for normalization processing, and normalization methods such as function transformation and maximum-minimum normalization can also be selected, which are not limited herein.

[0039] Preferably, in some possible implementation manners of the embodiments of the present invention, the method for obtaining the temperature crack probability includes: the environmental data includes temperature data and wind speed data; obtaining an environmental influence value according to the temperature data and the wind speed data; adjusting the pixel difference of each suspected crack area in the pavement infrared image by using the environmental influence value to obtain the actual temperature difference of each suspected crack area; normalizing the actual temperature difference to obtain the temperature crack probability of each suspected crack area.

[0040] When the temperature and wind speed are higher, the water evaporation rate accelerates, the suspected crack area absorbs more heat per unit time, the cooling effect is more significant, and the environmental influence on water evaporation is greater. Therefore, both the temperature data and the wind speed data are positively correlated with the environmental influence value. In the embodiments of the present invention, the product of the temperature data and the wind speed data is used as the environmental influence value. If the environmental influence value is larger, the cooling effect of the suspected crack area per unit time is more significant, making the actual temperature difference of the suspected crack area before and after watering on the high side. It is necessary to reduce the pixel difference of the suspected crack area in the pavement infrared image, so that the environmental influence value and the actual temperature difference are negatively correlated. In the embodiments of the present invention, the environmental influence value is negatively correlated and normalized, and the processing result is used to weight the pixel difference of each suspected crack area in the pavement infrared image to obtain the actual temperature difference of each suspected crack area.

[0041] It is known that the temperature difference of cracks is greater than that of the road surface before and after watering. If the actual temperature difference is larger, it indicates that the temperature difference of the suspected crack area before and after watering is larger. Then, the greater the possibility that the suspected crack area is identified as an actual crack through the temperature change characteristics before and after watering, and the greater the temperature crack probability.

[0042] It should be noted that in the embodiments of the present invention, the environmental impact value is used as the exponent of the exponential function with the natural constant as the base to achieve negative correlation and normalization processing; the Norm function is used for normalization processing, and other normalization methods such as maximum-minimum normalization can also be selected, which are not limited herein.

[0043] Preferably, in some possible implementation manners of the embodiments of the present invention, for the method of obtaining the final crack probability, please refer to Figure 2 , which shows a flowchart of a method for obtaining the final crack probability provided by an embodiment of the present invention. The method includes: Step S410: Obtain the feature weight coefficient according to the significance degree of the pixel value difference of the same suspected crack area in the road surface feature images before and after watering.

[0044] Preferably, in some possible implementation manners of the embodiments of the present invention, the method for obtaining the feature weight coefficient includes: taking the ratio of the sum of the mean value, standard deviation and a preset positive number of the pixel differences of all suspected crack areas in the road surface feature image as the feature significance; respectively recording the corresponding feature significances of the road surface gray image and the road surface infrared image as the gray feature significance and the infrared feature significance; taking the ratio with the infrared feature significance as the numerator and the sum value of the infrared feature significance and the gray feature significance as the denominator as the feature weight coefficient.

[0045] In a specific implementation manner of the embodiments of the present invention, the feature weight coefficient is expressed by the formula: In the formula, is the infrared feature significance; is the gray feature significance; is the mean value of the pixel differences of all suspected crack areas in the road surface infrared image; is the standard deviation of the pixel differences of all suspected crack areas in the road surface infrared image; is the mean value of the pixel differences of all suspected crack areas in the road surface gray image; is the standard deviation of the pixel differences of all suspected crack areas in the road surface gray image; is a preset positive number, which is used to prevent the denominator from being 0 and causing the fraction to be meaningless. In this embodiment, is set to 0.1, and the implementer can set it according to specific circumstances.

[0046] It should be noted that the mean and standard deviation of the pixel differences in all suspected crack regions in the road surface feature image respectively reflect the average amplitude and dispersion degree of the brightness changes in all suspected crack regions before and after watering. If the feature significance is greater, the temperature change or brightness change in the suspected crack region before and after watering is more significant and consistent, and the authenticity of the crack is higher, which is suitable for identifying deep cracks; on the contrary, the temperature change or brightness change in the suspected crack region before and after watering is not significant, which is suitable for surface cracks. When is greater than , the temperature change characteristics of the crack are more significant, and the possibility that the suspected crack region is an interlayer delamination crack is greater. approaches 1, and determining whether the suspected crack region is an actual crack mainly depends on the temperature changes in the road surface infrared images before and after watering; when is less than , the brightness change characteristics of the crack are more significant, and the possibility that the suspected crack region is a surface crack is greater. approaches 0, and determining whether the suspected crack region is an actual crack mainly depends on the brightness changes in the road surface gray-scale images before and after watering.

[0047] Step S420: Respectively use the feature weight coefficient, the difference between the constant 1 and the feature weight coefficient as the weights of the temperature crack probability and the brightness crack probability of each suspected crack region, and perform weighted summation on the temperature crack probability and the brightness crack probability of each suspected crack region to obtain the final crack probability of each suspected crack region.

[0048] In a specific implementation manner of the embodiment of the present invention, the final crack probability is expressed by the formula: In the formula, ZP is the final crack probability of each suspected crack region; is the feature weight coefficient; is the temperature crack probability of each suspected crack region; is the brightness crack probability of each suspected crack region. It should be noted that when the feature weight coefficient approaches 1, determining whether the suspected crack region is an actual crack mainly depends on the temperature changes before and after watering. The weight of the temperature crack probability should be greater, and the weight of the brightness crack probability should be smaller; when the feature weight coefficient When approaching 0, to determine whether a suspected crack area is an actual crack mainly depends on the brightness change before and after watering. The weight of the temperature crack probability should be smaller, and the weight of the brightness crack probability should be larger. If the final crack probability is greater, the crack characteristics in the suspected crack area are more obvious, that is, the crack is more serious, then the possibility that the suspected crack area is an actual crack is greater.

[0049] In an embodiment of the present invention, when the final crack probability of the suspected crack area is, the suspected crack area is a normal road surface area; when the final crack probability of the suspected crack area is, the suspected crack area is a minor crack, and it is recommended to observe; when the final crack probability of the suspected crack area is, the suspected crack area is a potentially expanding crack, and the staff needs to observe regularly to prevent the crack from expanding into a serious crack; when the final crack probability of the suspected crack area is, the suspected crack area is a serious crack, and the staff needs to repair it immediately. Other embodiments can also divide the suspected crack area into cracks of different severity levels based on the final crack probability, which will not be limited here.

[0050] Embodiment 2

[0051] The present invention provides a highway pavement quality detection system based on machine vision. Please refer to Figure 3 , which shows the system structure diagram of a highway pavement quality detection system provided by an embodiment of the present invention. The system includes: A data acquisition module 510, configured to respectively acquire the pavement characteristic images of the highway to be measured before and after watering, as well as the environmental data of the position of the highway to be measured; the pavement characteristic images include pavement grayscale images and pavement infrared images; A brightness characteristic analysis module 520, configured to acquire the suspected crack areas in the pavement characteristic images; and obtain the brightness crack probability of each suspected crack area according to the grayscale difference of the same suspected crack area in the pavement grayscale images before and after watering; A temperature characteristic analysis module 530, configured to obtain the temperature crack probability of each suspected crack area according to the pixel value difference of the same suspected crack area in the pavement infrared images before and after watering and the environmental data; A pavement quality detection module 540, configured to adjust the brightness crack probability and the temperature crack probability according to the significant degree of the pixel value difference of the same suspected crack area in the pavement characteristic images before and after watering, and obtain the final crack probability of each suspected crack area; and perform pavement quality detection on the highway to be measured by using the final crack probability.

[0052] It should be noted that: For the device provided in the above embodiment, only the division of the above functional modules is used for illustration. In actual applications, the above functions can be allocated 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 embodiments of a highway pavement quality detection system based on machine vision and a highway pavement quality detection method based on machine vision provided in the above embodiments belong to the same concept. The specific implementation process can be seen in the method embodiments and will not be elaborated here.

[0053] Embodiment 3

[0054] Figure 4 The following is a schematic diagram of a computer device of a highway pavement quality detection device based on machine vision provided by an embodiment of the present invention. Exemplarily, as Figure 4 shown, the computer device includes: a memory 601, a processor 602, and a computer program 603 stored in the memory 601 and running on the processor 602. When the processor 602 executes the computer program 603, the computer device can execute any one of the above-described highway pavement quality detection methods based on machine vision.

[0055] In addition, the embodiments of the present application also protect a device, which may include a memory and a processor. Among them, an executable program code is stored in the memory, and the processor is used to call and execute the executable program code to execute a highway pavement quality detection method provided by the embodiments of the present application.

[0056] In this embodiment, the device can be divided into functional modules according to the above method example. For example, it can correspond to each functional module, or two or more functions can be integrated into one processing module. The above integrated module can be implemented in the form of hardware. It should be noted that the division of modules in this embodiment is illustrative, only a logical function division, and there may be other division methods in actual implementation.

[0057] It should be understood that the device provided in this embodiment is used to execute the above-mentioned highway pavement quality detection method based on machine vision, so the same effect as the above implementation method can be achieved.

[0058] In the case of adopting an integrated unit, the device may include a processing module and a storage module. Among them, when the device is applied to a device, the processing module can be used to control and manage the actions of the device. The storage module can be used to support the device to execute mutual program codes, etc.

[0059] Among them, the processing module can be a processor or a controller, which can implement or execute various exemplary logic blocks, modules, and circuits included in the disclosure of the present application. The processor can also be a combination that realizes computing functions, such as including a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc. The storage module can be a memory.

[0060] It should be noted that the above order of the embodiments of the present invention is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or consecutive order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0061] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments.

[0062] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for detecting the quality of highway pavement based on machine vision, characterized in that, The method includes: Obtaining the pavement characteristic images of the highway to be measured before and after watering, and the environmental data at the location of the highway to be measured; the pavement characteristic images include pavement gray-scale images and pavement infrared images; Obtaining the suspected crack areas in the pavement characteristic images; obtaining the brightness crack probability of each suspected crack area according to the gray-scale difference of the same suspected crack area in the pavement gray-scale images before and after watering; Obtaining the temperature crack probability of each suspected crack area according to the pixel value difference of the same suspected crack area in the pavement infrared images before and after watering and the environmental data; Adjusting the brightness crack probability and the temperature crack probability according to the degree of significance of the pixel value difference of the same suspected crack area in the pavement characteristic images before and after watering, obtaining the final crack probability of each suspected crack area; using the final crack probability to perform pavement quality detection on the highway to be measured.

2. The method for detecting the quality of a highway pavement based on machine vision according to claim 1, wherein, The obtaining of the brightness crack probability of each suspected crack area includes: Taking the average pixel value of all pixel points in the suspected crack area as the overall pixel value; Calculating the absolute value of the difference between the overall pixel values of the same suspected crack area in the pavement characteristic images before and after watering, and obtaining the pixel difference of each suspected crack area in the pavement characteristic images; Normalizing the pixel difference of each suspected crack area in the pavement gray-scale image to obtain the brightness crack probability of each suspected crack area.

3. A method for detecting the quality of a highway pavement based on machine vision according to claim 2, characterized in that, The obtaining of the temperature crack probability of each suspected crack area includes: The environmental data includes temperature data and wind speed data; obtaining an environmental influence value according to the temperature data and the wind speed data; Adjusting the pixel difference of each suspected crack area in the pavement infrared image by using the environmental influence value to obtain the actual temperature difference of each suspected crack area; normalizing the actual temperature difference to obtain the temperature crack probability of each suspected crack area.

4. A method for detecting the quality of highway pavement based on machine vision according to claim 2, characterized in that, The obtaining of the final crack probability of each suspected crack area includes: Obtaining a feature weight coefficient according to the degree of significance of the pixel value difference of the same suspected crack area in the pavement characteristic images before and after watering; Taking the feature weight coefficient, the difference between the constant 1 and the feature weight coefficient as the weights of the temperature crack probability and the brightness crack probability of each suspected crack area in turn, and performing weighted summation on the temperature crack probability and the brightness crack probability of each suspected crack area to obtain the final crack probability of each suspected crack area.

5. The method for detecting the quality of highway pavement based on machine vision according to claim 4, characterized in that, The method for obtaining the feature weight coefficient includes: Taking the ratio of the sum of the average value, standard deviation and a preset positive number of the pixel differences of all suspected crack areas in the pavement characteristic images as the feature significance; respectively recording the corresponding feature significances of the pavement gray-scale image and the pavement infrared image as the gray-scale feature significance and the infrared feature significance; Taking the ratio obtained by using the infrared feature significance as the numerator and the sum value of the infrared feature significance and the gray-scale feature significance as the denominator as the feature weight coefficient.

6. The method for detecting the quality of highway pavement based on machine vision according to claim 1, characterized in that, The obtaining of the suspected crack areas in the pavement characteristic images includes: The segmentation threshold is obtained by using the maximum inter-class variance method for the gray values of the pixel points in the gray-scale image of the road surface before sprinkling, and the connected domain composed of the pixel points with gray values less than the segmentation threshold is used as the suspected crack area of the gray-scale image of the road surface before sprinkling; The infrared images of the road surface before and after sprinkling and the gray-scale image of the road surface after sprinkling are denoted as analysis images; the connected domain composed of the pixel points at the same coordinate positions in the analysis images for the pixel points in each suspected crack area of the gray-scale image of the road surface before sprinkling is used as the suspected crack area of the analysis images.

7. A method for detecting the quality of highway pavement based on machine vision according to claim 3, characterized in that, The obtaining of the actual temperature difference for each suspected crack area includes: Performing negative correlation and normalization processing on the environmental influence value, and weighting the pixel difference of each suspected crack area in the infrared image of the road surface by using the processing result to obtain the actual temperature difference of each suspected crack area.

8. A method for detecting the quality of a highway pavement based on machine vision according to claim 3, characterized in that, Both the temperature data and the wind speed data are positively correlated with the environmental influence value.

9. The method for detecting the quality of highway pavement based on machine vision according to claim 1, wherein, The pixel points in every two road surface feature images correspond one by one.

10. A highway pavement quality detection system based on machine vision, characterized in that, The system includes: A data acquisition module, configured to respectively obtain the road surface feature images of the to-be-detected highway before and after sprinkling, and the environmental data at the position of the to-be-detected highway; the road surface feature images include the gray-scale image of the road surface and the infrared image of the road surface; A brightness feature analysis module, configured to obtain the suspected crack areas in the road surface feature images; and obtain the brightness crack probability of each suspected crack area according to the gray-scale difference of the same suspected crack area in the gray-scale images of the road surface before and after sprinkling; A temperature feature analysis module, configured to obtain the temperature crack probability of each suspected crack area according to the pixel value difference of the same suspected crack area in the infrared images of the road surface before and after sprinkling and the environmental data; A road surface quality detection module, configured to adjust the brightness crack probability and the temperature crack probability according to the significance degree of the pixel value difference of the same suspected crack area in the road surface feature images before and after sprinkling, and obtain the final crack probability of each suspected crack area; and perform road surface quality detection on the to-be-detected highway by using the final crack probability.

Images