A road surface disease detection method, device and related medium based on pan-tilt camera

Abstract

The present invention discloses a pavement disease detection method, device and related media based on a pan-tilt camera. The method includes: using a pan-tilt camera to shoot a road to be detected to obtain a pavement overview map; in the pavement overview map, delineating a pavement area to be detected, and calculating the rotation range of the pan-tilt camera according to the pavement area to be detected; based on the pavement area to be detected, automatically planning an inspection path for the pan-tilt camera, and controlling the pan-tilt camera to perform inspection and shooting according to the inspection path to obtain a target image; detecting and identifying the location and type of the disease in the target image by image processing technology; and summarizing the detection and identification results in the pavement overview map. The present invention controls the pan-tilt camera to automatically inspect to detect the diseases existing on the pavement, and detects pavement diseases without requiring a large amount of repeated manpower operations and without requiring on-site field inspections, thereby improving the detection efficiency and accuracy of pavement diseases.

Description

Technical Field

[0001] The present invention relates to the technical field of computer vision and image processing, and in particular to a pavement disease detection method and device based on a pan-tilt camera and related media. Background Art

[0002] In recent years, the scale of urban roads and the mileage of highways have increased rapidly. With the large number of roads put into use and the continuous increase in transportation volume, the problem of pavement damage has become increasingly serious, directly affecting the performance and life of the road and reducing the safety of transportation. Timely and accurate pavement damage detection has become an important research topic in the maintenance and management of my country's transportation infrastructure.

[0003] However, taking crack detection as an example, the traditional manual detection cycle of pavement cracks often reaches several months, which seriously restricts road maintenance decisions and has lost timeliness to a certain extent; at the same time, the use of manual pavement crack detection is affected by factors such as insufficient automation of detection technology and high subjective judgment of detection personnel, and accuracy and stability are often affected. With the development of computer imaging technology and deep learning technology, image-based pavement disease detection technology has developed rapidly, but there are problems such as limited detection range and poor detection effect on small target diseases. Summary of the invention

[0004] The embodiments of the present invention provide a pavement defect detection method, device, computer equipment and storage medium based on a pan-tilt camera, aiming to improve the efficiency and accuracy of pavement defect detection.

[0005] In a first aspect, an embodiment of the present invention provides a pavement disease detection method based on a pan-tilt camera, comprising:

[0006] Use the PTZ camera to shoot the road to be inspected to obtain an overview of the road surface;

[0007] In the road overview map, delineate a road area to be detected, and calculate a rotation range of the pan / tilt camera according to the road area to be detected;

[0008] Based on the road area to be inspected, automatically planning an inspection path for the pan-tilt camera, and controlling the pan-tilt camera to perform inspection and shooting along the inspection path to obtain a target image;

[0009] Detecting and identifying the location and type of the disease in the target image by using image processing technology;

[0010] The detection and identification results are summarized in the road surface overview map.

[0011] In a second aspect, an embodiment of the present invention provides a road surface disease detection device based on a pan-tilt camera, comprising:

[0012] A road shooting unit, used to shoot the road to be inspected using a pan-tilt camera to obtain an overview of the road surface;

[0013] An area demarcation unit, used to demarcate a road surface area to be detected in the road surface overview map, and calculate a rotation range of the pan / tilt camera according to the road surface area to be detected;

[0014] An inspection and shooting unit, used to automatically plan an inspection path for the pan-tilt camera based on the road area to be inspected, and control the pan-tilt camera to perform inspection and shooting along the inspection path to obtain a target image;

[0015] A detection and identification unit, used to detect and identify the location and type of the disease in the target image by using image processing technology;

[0016] The result summarizing unit is used to summarize the detection and identification results into the road surface overview map.

[0017] In a third aspect, an embodiment of the present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the pavement defect detection method based on a gimbal camera as described in the first aspect is implemented.

[0018] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the pavement defect detection method based on a gimbal camera as described in the first aspect is implemented.

[0019] The embodiment of the present invention provides a road surface disease detection method, device, computer equipment and storage medium based on a pan-tilt camera, the method comprising: using a pan-tilt camera to shoot a road to be detected to obtain a road surface overview map; in the road surface overview map, delineating a road surface area to be detected, and calculating the rotation range of the pan-tilt camera according to the road surface area to be detected; based on the road surface area to be detected, automatically planning an inspection path for the pan-tilt camera, and controlling the pan-tilt camera to perform inspection and shooting according to the inspection path to obtain a target image; detecting and identifying the location and type of the disease in the target image by image processing technology; and summarizing the detection and identification results in the road surface overview map. The embodiment of the present invention controls the pan-tilt camera to automatically inspect to detect the diseases existing on the road surface, thereby detecting the relative road surface diseases without the need for a large amount of repeated manpower operations and without the need for on-site field inspections, thereby improving the detection efficiency and accuracy of road surface diseases and ensuring the efficiency and quality of road maintenance. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0021] Figure 1 A schematic flow chart of a pavement damage detection method based on a pan-tilt camera provided in an embodiment of the present invention;

[0022] Figure 2 Schematic diagram of the horizontal field of view angle and vertical field of view angle of the gimbal camera;

[0023] Figure 3 A schematic diagram of establishing an inspection path in a pavement disease detection method based on a pan-tilt camera provided in an embodiment of the present invention;

[0024] Figure 4 A schematic block diagram of a pavement defect detection device based on a pan-tilt camera provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0027] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0028] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0029] See below Figure 1 , Figure 1A schematic flow chart of a pavement defect detection method based on a pan-tilt camera provided in an embodiment of the present invention specifically includes steps S101 to S105.

[0030] S101, using a pan-tilt camera to photograph the road to be inspected to obtain an overview of the road surface;

[0031] S102, demarcating a road area to be detected in the road overview map, and calculating a rotation range of the pan / tilt camera according to the road area to be detected;

[0032] S103, based on the road area to be inspected, automatically planning an inspection path for the pan-tilt camera, and controlling the pan-tilt camera to perform inspection and shooting along the inspection path to obtain a target image;

[0033] S104, detecting and identifying the location and type of the disease in the target image by using image processing technology;

[0034] S105: Summarize the detection and identification results into the road overview map.

[0035] In this embodiment, the road to be inspected is first photographed using a PTZ camera, and the road area to be inspected is demarcated in the obtained road overview map, and then the rotation range of the PTZ camera in the road area to be inspected is calculated. Then, an inspection path is formulated for the PTZ camera in the road area to be inspected, so that the PTZ camera performs inspection and photography according to the inspection path, and at the same time, the target image obtained by the inspection photography is inspected for defects by combining image processing technology, and the types and locations of the detected defects are summarized in the road overview map for easy query.

[0036] This embodiment controls the pan-tilt camera to automatically patrol to detect road defects, thereby detecting relative road defects without the need for a large amount of repetitive manpower operations and without the need for on-site field inspections, thereby improving the detection efficiency and accuracy of road defects and ensuring the efficiency and quality of road maintenance.

[0037] In a specific embodiment, when the pan-tilt camera is used to shoot the road to be inspected and obtain an overview of the road surface, the pan-tilt camera is adjusted to a smaller zoom factor f. 0, and make the road surface as central as possible, and then shoot the road to be detected. At the same time, the influence of moving vehicles, pedestrians, etc. can be eliminated by video background accumulation technology to obtain an overview of the road surface. The video background accumulation technology is to shoot a video of the same road section for a period of time (for example, a few seconds). During this period of video, even if pedestrians and / or vehicles walk and / or drive on the road section, the images of the road section at different times can be combined for accumulation and superposition (i.e. pedestrians and vehicles are in a dynamic state, and their positions are different at each moment). In this way, an overview of the road surface with no influence or less influence can be obtained, thereby improving the final road surface disease detection accuracy.

[0038] In one embodiment, the step S102 includes:

[0039] Automatically extract the road surface area to be detected by using a computer vision algorithm or a deep learning semantic segmentation method;

[0040] Establishing an image coordinate system based on the road surface area to be detected;

[0041] The rotation angle of the pan-tilt camera is calculated according to the pixel positions of the road surface area to be detected in the image coordinate system.

[0042] In this embodiment, for the road overview map, the road area to be detected can be manually selected, or the road area to be detected can be automatically extracted using computer vision algorithms, deep learning semantic segmentation and other methods, and the road area to be detected is composed of boundary points. Here, when the road area to be detected is manually selected, a labeling tool can be used to manually label the road edge and boundary points; when a computer vision method is used, a road mask can be obtained by color threshold filtering, and the road boundary can be obtained by edge detection or connected area detection, and then the boundary points can be extracted; when a deep learning semantic segmentation method is used, a semantic segmentation model can be trained using labeled data, and the road overview map can be input into the trained deep learning semantic segmentation model, and the road mask can be extracted from the output category image, and then the road boundary can be obtained by edge detection or connected area detection, and then the boundary points can be extracted.

[0043] At the same time, combined Figure 2 In order to obtain the relationship between the rotation angle of the gimbal camera and the demarcated road area to be detected, an image coordinate system is established. Specifically: take the upper left corner of the image as the origin, the image rightward ray as the camera coordinate system u-axis, and the image downward ray as the camera coordinate system v-axis. It can be understood that by querying the camera description parameters, camera calibration, etc., the horizontal field of view angle size hfov and the vertical field of view angle size vfov of the gimbal camera can be obtained. In addition, if the image pixel height is h and the image pixel width is w, then the rotation range of the gimbal camera can be obtained by calculation.

[0044] Specifically, in one embodiment, the calculation of the rotation angle of the pan-tilt camera according to the pixel position of the road surface area to be detected in the image coordinate system includes:

[0045] The horizontal rotation angle θ and the vertical rotation angle β of the cloud camera are calculated respectively according to the following formula:

[0046]

[0047]

[0048] Wherein, hfov is the first horizontal field of view angle of the gimbal camera, vfov is the first vertical field of view angle of the gimbal camera, h is the image pixel height, w is the image pixel width, and x and y are the horizontal and vertical coordinates of the pixel point, respectively.

[0049] In this embodiment, for a pixel point (x, y) on the image, assuming that the camera needs to be aimed at the three-dimensional space object corresponding to the point, the horizontal rotation angle θ and the vertical rotation angle β required by the pan-tilt camera can be obtained according to the above calculation.

[0050] Furthermore, if the yaw angle of the gimbal camera is yaw 0 , the pitch angle is pitch 0 , then the gimbal camera needs to be rotated to the pixel point (x, y):

[0051] yaw=yaw 0 +θ

[0052] pitch=pitch 0 +β

[0053] For each point in the road surface area to be detected extracted above, the horizontal rotation and vertical rotation of the camera corresponding to the point can be calculated according to the calculation formula of the horizontal rotation angle θ and the vertical rotation angle β, thereby obtaining the rotation range M of the gimbal camera in the road surface area to be detected.

[0054] In one embodiment, the step S103 includes:

[0055] In the road area to be detected, adjusting the magnification of the pan-tilt camera, and acquiring a second horizontal field of view angle and a second vertical field of view angle corresponding to the magnification;

[0056] The inspection path is formulated in a sequence from top to bottom and from left to right, and the pan-tilt camera is made to perform inspection and shooting along the inspection path.

[0057] In this embodiment, in the road area to be detected, the camera inspection path is automatically planned and inspection shooting is performed. Specifically: in the road area to be detected, the magnification of the pan-tilt camera is adjusted and the magnification fi is recorded, and at the same time, the horizontal field of view angle hfov' (i.e., the second horizontal field of view angle) and the vertical field of view angle vfov' (i.e., the second vertical field of view angle) corresponding to the pan-tilt camera under the current magnification are queried and recorded, so as to plan the inspection path and perform image shooting. Here, this embodiment adopts an S-shaped route from top to bottom and from left to right for inspection.

[0058] Specifically, in one embodiment, the inspection path is formulated in a sequence from top to bottom and from left to right, and the pan / tilt camera is made to perform inspection and shooting according to the inspection path, including:

[0059] Within the rotation range of the pan-tilt camera, a point at the upper left corner of the road surface area to be inspected is used as the starting point of the inspection path, the starting point of the inspection path is photographed by the pan-tilt camera, and the corresponding first rotation angle is recorded;

[0060] Control the pan / tilt camera to rotate rightwards by an angle of a second horizontal field of view angle, and use the corresponding point as the next shooting point on the inspection path;

[0061] The same process is repeated until the rightward rotation angle of the pan-tilt camera exceeds the rotation range, and then the pan-tilt camera is controlled to rotate downward by an angle of the second vertical field of view angle, and the corresponding point is used as the next shooting point of the inspection path;

[0062] Control the pan / tilt camera to rotate leftward by an angle of a second horizontal field of view angle, and use the corresponding point as the next shooting point on the inspection path;

[0063] And so on, until the angle of the gimbal camera rotating to the right exceeds the rotation range, and then the gimbal camera is controlled to rotate downward to an angle of the second vertical field of view angle, and the corresponding point is used as the next inspection path shooting point, until the horizontal rotation angle and the vertical rotation angle of the gimbal camera both exceed the rotation range.

[0064] In this embodiment, combined with Figure 3 According to the following steps, the inspection path is formulated in the order from top to bottom and from left to right, and the pan-tilt camera is made to perform inspection and shooting according to the inspection path.

[0065] ① Take a point at the upper left of the rotation range M as the starting point of the inspection path, take an image of that point and record the camera rotation angle θ at that point i , β i .

[0066] ② Rotate right from the previous point by hfov' degrees, and take the next shooting point on the inspection path. Take the image of this point and record the camera rotation angle θ of this point. i , β i .

[0067] ③Repeat ② until the rotation range M is exceeded.

[0068] ④ Rotate vfov' degrees downward from the previous point to take the next shooting point on the inspection path, take an image of the point and record the camera rotation angle θ of the point i , β i .

[0069] ⑤ Rotate hfov' degrees to the left from the previous point, as the next shooting point on the inspection path, take an image of this point and record the camera rotation angle θ of this point i , β i .

[0070] ⑥ Repeat ⑤ until the rotation range M is exceeded.

[0071] ⑦ Rotate vfov' degrees downward from the previous point, as the next shooting point on the inspection path, take an image of this point and record the camera rotation angle θ of this point i , β i .

[0072] ⑧Repeat ② to ⑦ until the camera pitch angle exceeds the rotation range M.

[0073] It is worth noting that the inspection path can also be formulated according to other factors such as the shape of the area to be inspected and the rotation angle of the camera. At the same time, to ensure the inspection effect, the distance between the two shooting points in the distant area can be appropriately farther than the distance between the two shooting points in the near area. For example, in step ②, it can be rotated 2×hfov', etc. At the same time, every time the PTZ camera reaches a shooting point, in order to prevent camera motion blur and ensure sufficient focusing time, the camera can be stopped for a certain period of time before shooting the road surface.

[0074] In one embodiment, the step S104 includes:

[0075] Performing grayscale processing and binarization processing on the target image;

[0076] For the processed target image, any one or more of a morphological processing method, a filtering method and / or an image processing method based on deep learning is used to perform disease identification to obtain the location and type of the disease in the target image.

[0077] In this embodiment, the target image is first preprocessed, including but not limited to grayscale and binarization of the image, and then image processing techniques such as morphological processing, filtering and other methods or deep learning-based image classification, target detection, semantic segmentation, instance segmentation and other algorithms are used to identify road defects and obtain the type and location of the defect.

[0078] In one embodiment, the step S105 includes:

[0079] The coordinates (U, V) of the defect location in the pavement overview map are calculated according to the following formula:

[0080]

[0081] In the formula, u i and v i are the pixel coordinates of the disease location, θ i and β i are the horizontal rotation angle and vertical rotation angle of the gimbal camera, respectively, i is the magnification of the gimbal camera, f 0 is the zoom factor corresponding to the road overview image taken by the PTZ camera. hfov' and vfov' are the second horizontal field of view and the second vertical field of view of the PTZ camera respectively.

[0082] In this embodiment, according to the horizontal rotation angle and vertical rotation angle of the gimbal camera, the center point position coordinates of the gimbal camera can be calculated by the following formula:

[0083]

[0084] If the coordinates of the shooting point in the overview map are (x i ,y i ), the camera magnification is f i The pixel coordinates of a detected road defect in the image of the shooting point are (u i ,v i ), then the coordinates of the defect in the pavement overview map are:

[0085]

[0086] According to the position of each shooting point on the inspection route in the overview map, the position of the road disease in the image of each shooting point in the overview map is calculated respectively, and the position and type of the disease are marked in the road surface overview map.

[0087] Figure 4 A schematic block diagram of a pavement disease detection device 400 based on a pan-tilt camera provided in an embodiment of the present invention, the device 400 includes:

[0088] A road shooting unit 401 is used to shoot the road to be inspected using a PTZ camera to obtain an overview of the road surface;

[0089] An area demarcation unit 402 is used to demarcate a road area to be detected in the road overview map, and calculate a rotation range of the pan / tilt camera according to the road area to be detected;

[0090] The inspection and shooting unit 403 is used to automatically plan an inspection path for the pan-tilt camera based on the road surface area to be inspected, and control the pan-tilt camera to perform inspection and shooting along the inspection path to obtain a target image;

[0091] A detection and identification unit 404 is used to detect and identify the location and type of the disease in the target image by using image processing technology;

[0092] The result summarizing unit 405 is used to summarize the detection and identification results into the road overview map.

[0093] In one embodiment, the area demarcation unit 402 includes:

[0094] A region extraction unit, used to automatically extract the road surface region to be detected by using a computer vision algorithm or a deep learning semantic segmentation method;

[0095] A coordinate system establishing unit, used for establishing an image coordinate system based on the road surface area to be detected;

[0096] The rotation angle calculation unit is used to calculate the rotation angle of the pan-tilt camera according to the pixel point position of the road surface area to be detected in the image coordinate system.

[0097] In one embodiment, the rotation angle calculation unit includes:

[0098] The horizontal and vertical angle calculation unit is used to calculate the horizontal rotation angle θ and the vertical rotation angle β of the cloud-sky camera according to the following formula:

[0099]

[0100]

[0101] Wherein, hfov is the first horizontal field of view angle of the gimbal camera, vfov is the first vertical field of view angle of the gimbal camera, h is the image pixel height, w is the image pixel width, and x and y are the horizontal and vertical coordinates of the pixel point, respectively.

[0102] In one embodiment, the inspection and shooting unit 403 includes:

[0103] A magnification adjustment unit, used to adjust the magnification of the pan / tilt camera in the road area to be detected, and obtain a second horizontal field of view angle and a second vertical field of view angle corresponding to the magnification;

[0104] The path formulation unit is used to formulate the inspection path in a sequence from top to bottom and from left to right, and to enable the pan-tilt camera to perform inspection and shooting according to the inspection path.

[0105] In one embodiment, the path formulation unit includes:

[0106] A starting point setting unit, used to take a point in the upper left corner of the road surface area to be inspected as the starting point of the inspection path within the rotation range of the pan-tilt camera, use the pan-tilt camera to shoot the starting point of the inspection path, and record the corresponding first rotation angle;

[0107] A first angle rotation unit is used to control the pan / tilt camera to rotate rightwards to an angle of a second horizontal field of view angle, and use the corresponding point as a shooting point for the next inspection path;

[0108] A second angle rotation unit is used to control the pan-tilt camera to rotate downward by an angle of a second vertical field of view angle, and use the corresponding point as the next shooting point on the inspection path;

[0109] A third angle rotation unit is used to control the pan / tilt camera to rotate leftward by an angle of a second horizontal field of view angle, and use the corresponding point as a shooting point for the next inspection path;

[0110] The fourth angle rotation unit is used to control the pan-tilt camera to rotate downward by an angle of a second vertical field of view angle, and use the corresponding point as the next inspection path shooting point, until the horizontal rotation angle and the vertical rotation angle of the pan-tilt camera both exceed the rotation range.

[0111] In one embodiment, the detection and identification unit 404 includes:

[0112] An image processing unit, used for performing grayscale processing and binarization processing on the target image;

[0113] The disease recognition unit is used to perform disease recognition on the processed target image by using any one or more of a morphological processing method, a filtering method and / or an image processing method based on deep learning to obtain the location and type of the disease in the target image.

[0114] In one embodiment, the result aggregation unit 405 includes:

[0115] The coordinate calculation unit is used to calculate the coordinates (U, V) of the defect location in the road surface overview map according to the following formula:

[0116]

[0117] In the formula, u i and v i are the pixel coordinates of the disease location, θ i and β i are the horizontal rotation angle and vertical rotation angle of the gimbal camera, respectively, i is the magnification of the gimbal camera, f 0 is the zoom factor corresponding to the road overview image taken by the PTZ camera. hfov' and vfov' are the second horizontal field of view and the second vertical field of view of the PTZ camera respectively.

[0118] Since the embodiments of the apparatus part correspond to the embodiments of the method part, please refer to the description of the embodiments of the method part for the embodiments of the apparatus part, which will not be repeated here.

[0119] The embodiment of the present invention further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed, the steps provided in the above embodiment can be implemented. The storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0120] The embodiment of the present invention also provides a computer device, which may include a memory and a processor, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, the steps provided in the above embodiment may be implemented. Of course, the computer device may also include various network interfaces, power supplies and other components.

[0121] The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of this application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the claims of this application.

[0122] It should also be noted that, in this specification, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprising a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

Claims

1. A road surface disease detection method based on a pan-tilt camera, characterized in that: include: Use the PTZ camera to shoot the road to be inspected to obtain an overview of the road surface; In the road overview map, delineate a road area to be detected, and calculate a rotation range of the pan / tilt camera according to the road area to be detected; Based on the road area to be inspected, automatically planning an inspection path for the pan-tilt camera, and controlling the pan-tilt camera to perform inspection and shooting along the inspection path to obtain a target image; Detecting and identifying the location and type of the disease in the target image by using image processing technology; Summarizing the detection and identification results into the road surface overview map; The method of automatically planning an inspection path for the pan-tilt camera based on the road surface area to be inspected, and controlling the pan-tilt camera to perform inspection and shooting along the inspection path to obtain a target image includes: In the road area to be detected, adjusting the magnification of the pan-tilt camera, and acquiring a second horizontal field of view angle and a second vertical field of view angle corresponding to the magnification; The inspection path is formulated in a sequence from top to bottom and from left to right, and the pan / tilt camera is made to perform inspection and shooting according to the inspection path; The step of formulating the inspection path in a sequence from top to bottom and from left to right, and causing the pan-tilt camera to perform inspection and shooting according to the inspection path, includes: Within the rotation range of the pan-tilt camera, a point at the upper left corner of the road surface area to be inspected is used as the starting point of the inspection path, the starting point of the inspection path is photographed by the pan-tilt camera, and the corresponding first rotation angle is recorded; Control the pan / tilt camera to rotate rightwards by an angle of a second horizontal field of view angle, and use the corresponding point as the next shooting point on the inspection path; The same process is repeated until the rightward rotation angle of the pan-tilt camera exceeds the rotation range, and then the pan-tilt camera is controlled to rotate downward by an angle of the second vertical field of view angle, and the corresponding point is used as the next shooting point of the inspection path; Control the pan / tilt camera to rotate leftward by an angle of a second horizontal field of view angle, and use the corresponding point as the next shooting point on the inspection path; And so on, until the angle of the gimbal camera rotating to the right exceeds the rotation range, and then the gimbal camera is controlled to rotate downward to an angle of the second vertical field of view angle, and the corresponding point is used as the next inspection path shooting point, until the horizontal rotation angle and the vertical rotation angle of the gimbal camera both exceed the rotation range.

2. The road surface disease detection method using a PTZ camera according to claim 1, characterized in that: Delineating a road surface area to be detected in the road surface overview map, and calculating a rotation range of the pan / tilt camera according to the road surface area to be detected, includes: Automatically extract the road surface area to be detected by using a computer vision algorithm or a deep learning semantic segmentation method; Establishing an image coordinate system based on the road surface area to be detected; The rotation angle of the pan-tilt camera is calculated according to the pixel positions of the road surface area to be detected in the image coordinate system.

3. The road surface disease detection method using a PTZ camera according to claim 2, characterized in that: The calculating the rotation angle of the pan-tilt camera according to the pixel position of the road surface area to be detected in the image coordinate system includes: The horizontal rotation angle θ and the vertical rotation angle β of the gimbal camera are calculated respectively according to the following formula: Wherein, hfov is the first horizontal field of view angle of the gimbal camera, vfov is the first vertical field of view angle of the gimbal camera, h is the image pixel height, w is the image pixel width, and x and y are the horizontal and vertical coordinates of the pixel point, respectively.

4. The method for detecting road defects using a PTZ camera according to claim 1, characterized in that: The detecting and identifying the location and type of the disease in the target image by using image processing technology includes: Performing grayscale processing and binarization processing on the target image; For the processed target image, any one or more of a morphological processing method, a filtering method and / or an image processing method based on deep learning is used to perform disease identification to obtain the location and type of the disease in the target image.

5. The road surface disease detection method using a PTZ camera according to claim 4, characterized in that: The step of aggregating the detection and recognition results into the road overview map includes: The coordinates (U, V) of the defect location in the pavement overview map are calculated according to the following formula: In the formula, u i and v i are the pixel coordinates of the disease location, θ i and β i are the horizontal rotation angle and vertical rotation angle of the pan-tilt camera respectively, fi is the magnification of the pan-tilt camera, f0 is the zoom factor corresponding to the road overview map taken by the pan-tilt camera, hfov' and vfov' are the second horizontal field of view angle and the second vertical field of view angle of the pan-tilt camera respectively.

6. A road surface disease detection device based on a pan-tilt camera, characterized in that: include: A road shooting unit, used to shoot the road to be inspected using a pan-tilt camera to obtain an overview of the road surface; An area demarcation unit, used to demarcate a road surface area to be detected in the road surface overview map, and calculate a rotation range of the pan / tilt camera according to the road surface area to be detected; An inspection and shooting unit, used to automatically plan an inspection path for the pan-tilt camera based on the road surface area to be inspected, and control the pan-tilt camera to perform inspection and shooting along the inspection path to obtain a target image; A detection and identification unit, used to detect and identify the location and type of the disease in the target image by using image processing technology; A result summarizing unit, used for summarizing the detection and identification results into the road surface overview map; The inspection and shooting unit comprises: A magnification adjustment unit, used to adjust the magnification of the pan / tilt camera in the road area to be detected, and obtain a second horizontal field of view angle and a second vertical field of view angle corresponding to the magnification; A path formulation unit, used to formulate the inspection path in a sequence from top to bottom and from left to right, and to enable the pan-tilt camera to perform inspection and shooting according to the inspection path; The path formulation unit comprises: A starting point setting unit, used to take a point in the upper left corner of the road surface area to be inspected as the starting point of the inspection path within the rotation range of the pan-tilt camera, use the pan-tilt camera to shoot the starting point of the inspection path, and record the corresponding first rotation angle; A first angle rotation unit is used to control the pan / tilt camera to rotate rightwards to an angle of a second horizontal field of view angle, and use the corresponding point as a shooting point for the next inspection path; A second angle rotation unit is used to control the pan-tilt camera to rotate downward by an angle of a second vertical field of view angle, and use the corresponding point as the next shooting point on the inspection path; A third angle rotation unit is used to control the pan / tilt camera to rotate leftward by an angle of a second horizontal field of view angle, and use the corresponding point as a shooting point for the next inspection path; The fourth angle rotation unit is used to control the pan-tilt camera to rotate downward by an angle of a second vertical field of view angle, and use the corresponding point as the next inspection path shooting point, until the horizontal rotation angle and the vertical rotation angle of the pan-tilt camera both exceed the rotation range.

7. A computer device, characterized in that: It comprises 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, it implements the pavement damage detection method based on a pan-tilt camera as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the pavement damage detection method based on a gimbal camera as described in any one of claims 1 to 5 is implemented.

Images