Method for real-time road surface information collection, and computer program recorded on a recording medium for executing the same

Abstract

The present invention proposes a real-time road surface information collection method capable of precisely analyzing road surface conditions based on data acquired while driving on a road, rapidly and accurately detecting risk factors such as potholes, and transmitting the analyzed road surface condition information to a management server in real time. The method may include the steps of: an analysis server collecting road surface information in real time from an information collection device that collects information regarding the road surface while moving on the road; the analysis server analyzing the condition of the road surface based on the real-time collected road surface information; and the analysis server transmitting the road surface condition information according to the analyzed road surface condition to a management server in real time.

Description

Technology Field

[0001] The present invention relates to a technology for collecting real-time road surface information. More specifically, it relates to a method for collecting real-time road surface information capable of analyzing road surface conditions based on road surface information acquired while driving on a road and transmitting the analyzed results in real time, and a computer program recorded on a recording medium for executing the same. Background Technology

[0002] Recently, technologies for precisely assessing and maintaining road surface conditions are becoming increasingly important to ensure the safety and efficiency of road traffic infrastructure. Road pavements can develop damage such as cracks, unevenness, and potholes due to repetitive vehicle loads, changes in weather conditions, and rainwater infiltration. Such damage can compromise driving safety and act as a cause of traffic accidents. Therefore, technologies capable of accurately measuring road surface conditions and managing them in real time are absolutely essential.

[0003] Meanwhile, Mobile Mapping System (MMS) refers to a technology that precisely acquires three-dimensional spatial information around roads by mounting mobile surveying devices on vehicles. MMS technology can collect image and point cloud data of the road environment by combining sensors such as cameras, LiDAR, GNSS (Global Navigation Satellite System), and IMU (Inertial Measurement Unit), and produce high-precision maps based on this data. This MMS technology is utilized not only to assess the condition of road structures or manage traffic infrastructure but also to construct high-definition (HD) maps for autonomous vehicles. However, conventional MMS technology has primarily focused on precisely measuring the overall geometric shape of roads or the locations of facilities, which has limited its ability to automatically detect and analyze minute damage or real-time changes in the road surface.

[0004] In particular, potholes are depressions formed when a portion of the road pavement detaches or breaks, primarily caused by repeated vehicle loads or water infiltration. These potholes not only cause physical problems such as tire failure or suspension damage, but also trigger dangerous situations like sudden braking or abrupt lane changes while driving, which can lead to secondary traffic accidents. Therefore, it is crucial to promptly detect the occurrence of potholes, assess their location and severity, and immediately reflect this in the road management system.

[0005] Road surface analysis technology is not only used as basic data for road maintenance, but can also be utilized in various fields, such as providing data to improve the driving stability of autonomous vehicles, serving as a basis for traffic big data analysis and smart city infrastructure construction, and supporting rapid road restoration in the event of an emergency.

[0006] However, conventional road surface analysis technology suffered from a lack of real-time capability as it relied primarily on periodic field inspections or fixed equipment. Furthermore, road surface measurement devices installed in some vehicles often depended on a single sensor, such as a camera, leading to a problem where analysis accuracy was easily degraded by environmental factors such as lighting, weather, and vehicle vibration.

[0007] Furthermore, conventional road surface analysis technology lacks the capability to assess risk or determine priority for localized damage such as potholes, making it difficult to establish efficient maintenance plans. Additionally, because most of the raw data is transmitted to the server in bulk, there were problems such as high bandwidth consumption and an increased data processing burden on the server. Prior art literature

[0008] Korean Registered Patent Publication No. 10-2171827, 'Precision Road Map Construction System Capable of Producing 3D Road Maps', (Published Oct. 23, 2020) The problem to be solved

[0009] One objective of the present invention is to provide a real-time road surface information collection method capable of precisely analyzing road surface conditions based on data acquired while driving on a road and rapidly and accurately detecting risk factors such as potholes, and a computer program recorded on a recording medium for executing the same.

[0010] Another objective of the present invention is to provide a method for collecting real-time road surface information capable of transmitting analyzed road surface condition information to a management server in real time, and a computer program recorded on a recording medium for executing the same.

[0011] The technical problems of the present invention are not limited to those mentioned above, and other unmentioned technical problems will be clearly understood by those skilled in the art from the description below. means of solving the problem

[0012] To achieve the technical objectives described above, the present invention proposes a real-time road surface information collection method capable of precisely analyzing road surface conditions based on data acquired while driving on a road, rapidly and accurately detecting risk factors such as potholes, and transmitting the analyzed road surface condition information to a management server in real time. The method may include the steps of: an analysis server collecting road surface information in real time from an information collection device that collects information regarding the road surface while moving on the road; the analysis server analyzing the condition of the road surface based on the real-time collected road surface information; and the analysis server transmitting the road surface condition information according to the analyzed road surface condition to a management server in real time.

[0013] Specifically, the step of collecting the road surface information is characterized by collecting an image of the road surface captured by a camera and point cloud data obtained by laser pulses reflected from the road surface through a lidar installed together with the camera.

[0014] The above-described analysis step is characterized by identifying at least one frame containing pre-measured ground control point (GCP) coordinates within the image, estimating the exterior orientation parameters (EOP) of the identified at least one frame based on the ground control point coordinates, and performing bundle adjustment on the exterior orientation parameters.

[0015] The above-described analysis step is characterized by estimating Relative Orientation Parameters (ROP) between a first image captured by a first camera and a second image captured by a second camera, generating an epipolar image based on the estimated Relative Orientation Parameters, generating a disparity map from the generated epipolar image, and self-calibrating Exterior Orientation Parameters (EOP) through bundle adjustment based on the generated disparity map.

[0016] The above-described analysis step is characterized by generating an elevation orthophoto based on the above-described image and point cloud data, calculating the height difference of the road surface from the generated elevation orthophoto, and detecting an area where the calculated height difference is greater than or equal to a preset value as a pothole area.

[0017] The above-mentioned transmitting step is characterized by selectively extracting at least one frame in which the pothole area is detected, and transmitting the location information and metadata of the corresponding pothole area together with the extracted at least one frame to the management server.

[0018] The above-described transmitting step is characterized by calculating a risk index based on the shape of the detected pothole area, determining the priority of the pothole area according to the calculated risk index, and transmitting a frame containing the pothole area and related data to the management server according to the determined priority.

[0019] The above-mentioned transmitting step is characterized by calculating a risk index based on the area of ​​the pothole region and the average depth value relative to the area.

[0020] The above-described transmitting step is characterized by dividing the road into multiple areas according to risk level, identifying the location of the pothole area within the divided area, and assigning weights to a risk index according to a preset risk level for the identified location.

[0021] The above-mentioned transmitting step is characterized by first transmitting meta-information regarding the detected pothole area to a management server, and, when a data request signal regarding the pothole area is received from the management server in response to the meta-information, additionally transmitting at least one extracted frame.

[0022] The above-described transmitting step is characterized by calculating a reliability index of the analysis result for the pothole area, determining a transmission level based on the calculated reliability index, and selectively transmitting at least one extracted frame, location information of the corresponding pothole area, and metadata to the management server based on the determined transmission level.

[0023] The above-described transmitting step is characterized by transmitting only location information and summary metadata of the pothole area when the reliability index is greater than or equal to a first reference value, transmitting an image of a Region of Interest (ROI) containing the pothole area and point cloud data corresponding to the image when the reliability index is less than or equal to a second reference value, and transmitting at least one extracted frame and point cloud data corresponding to the frame when the reliability index is less than the second reference value.

[0024] The above computer program may be combined with a computing device comprising a transceiver, memory, and a processor that processes instructions residing in said memory. Furthermore, the above computer program may be a computer program recorded on a recording medium to enable the processor to execute the steps of: collecting road surface information in real time from an information collection device that moves along a road and collects information regarding the road surface; analyzing the condition of the road surface based on the real-time collected road surface information; and transmitting road surface condition information according to the analyzed road surface condition to a management server in real time.

[0025] Specific details of other embodiments are included in the detailed description and drawings. Effects of the invention

[0026] According to various embodiments of the present invention, by precisely analyzing the road surface condition based on data acquired while driving on the road, risk factors such as potholes can be detected quickly and accurately, and early response to road damage is possible, thereby significantly improving the efficiency of road management and maintenance.

[0027] In addition, according to various embodiments of the present invention, analyzed road surface condition information is transmitted to a management server in real time, thereby enabling immediate acquisition of the latest road condition information, more effective management of traffic safety, and rapid maintenance decision-making.

[0028] The effects of the present invention are not limited to those mentioned above, and other unmentioned effects will be clearly understood by a person skilled in the art to which the present invention pertains from the description in the claims. Brief explanation of the drawing

[0029] FIG. 1 is a configuration diagram of a road information collection system according to one embodiment of the present invention. FIG. 2 is a logical configuration diagram of an analysis server according to one embodiment of the present invention. FIGS. 3 to 7 are illustrative diagrams for explaining a road surface information analysis process according to an embodiment of the present invention. FIG. 8 is a hardware configuration diagram of an analysis server according to one embodiment of the present invention. FIG. 9 is a flowchart illustrating a method for collecting real-time road surface information according to an embodiment of the present invention. Specific details for implementing the invention

[0030] It should be noted that technical terms used in this specification are used merely to describe specific embodiments and are not intended to limit the invention. Furthermore, unless specifically defined otherwise in this specification, technical terms used in this specification should be interpreted in the sense generally understood by those skilled in the art to which the invention pertains, and should not be interpreted in an overly broad or overly narrow sense. Additionally, if a technical term used in this specification is an incorrect technical term that fails to accurately express the spirit of the invention, it should be understood as being replaced by a technical term that can be correctly understood by those skilled in the art. Moreover, general terms used in this invention should be interpreted according to their prior definitions or the context, and should not be interpreted in an overly narrow sense.

[0031] Additionally, singular expressions used in this specification include plural expressions unless the context clearly indicates otherwise. In this application, terms such as "composed of" or "have" should not be interpreted as necessarily including all of the various components or steps described in the specification, and should be interpreted as potentially including some of the components or steps, or including additional components or steps.

[0032] Additionally, terms including ordinal numbers, such as first, second, etc., used herein may be used to describe various components, but said components shall not be limited by said terms. Such terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be named the second component, and similarly, the second component may be named the first component.

[0033] When it is stated that one component is "connected" or "connected" to another component, it may be directly connected or connected to that other component, or there may be other components in between. On the other hand, when it is stated that one component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between.

[0034] Hereinafter, preferred embodiments according to the present invention will be described in detail with reference to the attached drawings. Identical or similar components regardless of drawing symbols are given the same reference number, and redundant descriptions thereof will be omitted. Furthermore, in describing the present invention, if it is determined that a detailed description of related known technology may obscure the essence of the present invention, such detailed description will be omitted. Additionally, it should be noted that the attached drawings are intended only to facilitate an easy understanding of the concept of the present invention and should not be interpreted as limiting the concept of the present invention. The concept of the present invention should be interpreted as extending to all modifications, equivalents, and substitutions other than those shown in the attached drawings.

[0036] Meanwhile, conventional road surface analysis technology suffered from a lack of real-time capability as it relied primarily on periodic field inspections or fixed equipment. Additionally, road surface measurement devices installed in some vehicles often depended on a single sensor, such as a camera, leading to a problem where analysis accuracy was easily degraded by environmental factors such as lighting, weather, and vehicle vibration.

[0037] Furthermore, conventional road surface analysis technology lacks the capability to assess risk or determine priority for localized damage such as potholes, making it difficult to establish efficient maintenance plans. Additionally, because most of the raw data is transmitted to the server in bulk, there were problems such as high bandwidth consumption and an increased data processing burden on the server.

[0038] To overcome these limitations, the present invention proposes various means to precisely analyze road surface conditions based on data acquired while driving on a road, to quickly and accurately detect risk factors such as potholes, and to transmit the analyzed road surface condition information to a management server in real time.

[0040] FIG. 1 is a configuration diagram of a road information collection system according to one embodiment of the present invention.

[0041] Referring to FIG. 1, a road information collection system according to one embodiment of the present invention may be configured to include an information collection device (100), an analysis server (200), and a management server (300).

[0042] As such, since the components of the road information collection system according to one embodiment of the present invention merely represent functionally distinct elements, two or more components may be implemented as an integrated unit in an actual physical environment, or a single component may be implemented as a separate unit in an actual physical environment.

[0043] To describe each component, the information collection device (100) is mounted on a vehicle and can perform the function of capturing road surface images or acquiring point cloud data while moving along the road.

[0044] Specifically, the information collection device (100) may include various sensors such as a camera, lidar, GPS (Global Positioning System), and IMU (Inertial Measurement Unit). The information collection device (100) can transmit data acquired through the sensors to an analysis server (200) in real time via a network.

[0045] Furthermore, the information collection device (100) can collect metadata such as location coordinates, time information, and vehicle speed together with image and point cloud data and transmit it to the analysis server (200).

[0046] The information gathering device (100) is shown as being installed in a vehicle, but is not limited thereto and can be installed in various means of transportation such as a drone, train, or bike.

[0047] Next, the analysis server (200) can perform the function of analyzing the road surface condition based on the road surface information received from the information collection device (100).

[0048] Specifically, the analysis server (200) can identify ground control points (GCP) in the received image, correct exterior orientation parameters (EOP), generate a disparity map using multiple camera images, and produce an orthophoto based on the same.

[0049] Additionally, the analysis server (200) can fuse elevation orthophotos and point cloud data to calculate height differences and changes in surface patterns of the road surface, and based on this, detect various road surface damage areas such as potholes, cracks, spalling, and unevenness.

[0050] Furthermore, the analysis server (200) can analyze the characteristics of the detected road surface damage area, such as area, depth, length, location, and shape, calculate a risk index, determine the priority of the risk, and transmit it to the management server (300).

[0051] Such an analysis server (200) can be implemented as a local server, a cloud server, an edge server, etc., and embodiments of the present invention are not limited to a specific server type.

[0052] Meanwhile, a detailed description of the analysis server (200) will be provided below with reference to the drawings.

[0053] In the following configuration, the management server (300) may be a central management device that receives road surface condition analysis results transmitted from the analysis server (200), updates the road condition database based on this, and provides real-time road information services.

[0054] The management server (300) can collect and store various road surface damage detection results, such as potholes, cracks, unevenness, and road subsidence, provided by the analysis server (200), and update the precise road map in real time based on the information.

[0055] In addition, the management server (300) can support tracking road damage history and establishing maintenance plans by accumulating and managing road condition change history over a long period.

[0056] The management server (300) may include a function to distribute updated road information to external related organizations and service providers. For example, the management server (300) may provide road information to various private businesses, such as automobile manufacturers, portal sites, navigation operators, game companies, and real estate companies, via an Application Programming Interface (API) or a database linkage method.

[0057] In addition, the management server (300) can be linked with a road management agency to propose repair priorities for dangerous sections such as potholes and cracks, and provide real-time notifications of locations where urgent repair requests are needed.

[0058] In this way, the management server (300) is not limited to simply storing analysis results, but functions as an integrated management and service hub for road condition information, and can support road users and related organizations in utilizing the latest road condition information.

[0059] Furthermore, the management server (300) can be implemented as a local server, a cloud server, an edge server, etc., and the embodiments of the present invention are not limited to a specific server type.

[0060] The information collection device (100), analysis server (200), and management server (300) described above can transmit and receive data using a network that combines one or more of a secure line, a public wired communication network, or a mobile communication network that directly connects the devices.

[0061] For example, public wired communication networks may include Ethernet, Digital Subscriber Line (xDSL), Hybrid Fiber Coax (HFC), and Fiber To The Home (FTTH), but are not limited thereto. Additionally, mobile communication networks may include Code Division Multiple Access (CDMA), Wideband CDMA (WCDMA), High Speed ​​Packet Access (HSPA), Long Term Evolution (LTE), and 5th generation mobile telecommunication, but are not limited thereto.

[0063] Hereinafter, the logical configuration of an analysis server according to one embodiment of the present invention will be described in detail.

[0064] FIG. 2 is a logical configuration diagram of an analysis server according to an embodiment of the present invention, and FIGS. 3 to 7 are illustrative diagrams for explaining a road surface information analysis process according to an embodiment of the present invention.

[0065] Referring to FIG. 2, an analysis server (200) according to one embodiment of the present invention may be configured to include a communication unit (205), an input / output unit (210), a road surface information analysis unit (215), a transmission plan establishment unit (220), and a storage unit (225).

[0066] To explain each configuration, the communication unit (205) can perform the role of transmitting and receiving data with the information collection device (100) and the management server (300).

[0067] Specifically, the communication unit (205) can receive metadata such as road surface images, LiDAR point cloud data, location coordinates, time information, and vehicle speed in real time from an information collection device (100) mounted on various mobile bodies such as vehicles and drones, and transmit this to the road surface information analysis unit (215) so that subsequent analysis can be performed.

[0068] Additionally, the communication unit (205) can also perform the function of transmitting the analysis results processed by the road surface information analysis unit (215) to the management server (300) according to the control signal of the transmission plan establishment unit (220). For example, the communication unit (205) can transmit data to the management server (300) that includes the detection results of damaged areas such as potholes, cracks, and unevenness, as well as detailed information such as the location, area, and depth of the area.

[0069] Furthermore, the communication unit (205) can receive control commands, such as data request signals or maintenance priority notifications, from the management server (300) through bidirectional communication between the analysis server (200) and the management server (300), and transmit them to an internal module to support the entire system to operate in conjunction.

[0070] With the following configuration, the input / output unit (210) can receive control commands executed by an operator or manager of the analysis server (200) or perform the function of visually outputting analysis results.

[0071] Specifically, the input / output unit (210) can be used to control the operation of the road surface information analysis unit (215) and the transmission plan establishment unit (220) by receiving control commands, setting information, policy rules, etc., input by a system administrator or operator. For example, the input / output unit (210) can adjust the analysis criteria for image and point cloud data, the threshold setting for pothole and crack detection, the weighting of the risk calculation method, and the data transmission priority policy according to the administrator's input.

[0072] Additionally, the input / output unit (210) can visually output the results performed by the analysis server (200) so that an administrator can verify them. For example, the input / output unit (210) may output pothole detection results for a specific road section, the location and length of cracks, the severity of road subsidence areas, and the calculation results of risk indices through a map-based interface or a dashboard screen.

[0073] Furthermore, the input / output unit (210) may be linked with the management server (300) to perform the function of converting and outputting analysis results according to the data format requested by an external organization or system.

[0074] With the following configuration, the road surface information analysis unit (215) can perform a core function of precisely analyzing the condition of the road surface based on image and point cloud data received from the information collection device (100). In particular, the road surface information analysis unit (215) may be configured to include a series of procedures including data preprocessing and synchronization, geometric correction and self-correction, fusion of image and point cloud data and generation of orthophotos, detection of damaged areas, selection of frames of interest, and addition of metadata.

[0075] Specifically, the road surface information analysis unit (215) performs the function of synchronizing the image frames and LiDAR point cloud data provided by the information collection device (100) along the time axis and converting them into the same coordinate system. To this end, the time points between the data are aligned based on the timestamps included in the image and point cloud data, and the process of aligning them to a single ground reference coordinate system is performed by applying external parameters between the LiDAR and the camera. This preprocessing process contributes to increasing the accuracy of geometric correction and elevation difference calculation performed in subsequent steps.

[0076] The road surface information analysis unit (215) can identify frames containing previously surveyed ground control point coordinates (GCP) in order to accurately estimate the spatial location of the road surface based on collected image data. Here, the ground control point coordinates (GCP) may refer to the coordinate values ​​of a specific point accurately measured in a ground reference coordinate system, such as latitude, longitude, and altitude. For example, the road surface information analysis unit (215) may use road signs, lane intersections, artificially installed markers, etc., around the road as ground control points.

[0077] The road surface information analysis unit (215) can search for an image frame containing these reference point coordinates and then estimate the Exterior Orientation Parameters (EOP) representing the position and orientation of the camera within the frame. Here, the Exterior Orientation Parameters are values ​​that include the center coordinates of the camera at the time of shooting and the rotation angle of the camera, and can be essential parameters for aligning the image data with the actual ground coordinate system.

[0078] The road surface information analysis unit (215) can correct the estimated external orientation elements through a bundle adjustment process. That is, the road surface information analysis unit (215) can optimize the position and orientation of the camera by considering multiple video frames simultaneously. In other words, when the same object in the video appears across multiple frames, the road surface information analysis unit (215) can minimize the error by integrating the observations. Through this, the road surface information analysis unit (215) can reduce the estimation error of the external orientation elements calculated on an individual frame basis and ensure consistent geometric alignment throughout the entire video sequence.

[0079] Additionally, the road surface information analysis unit (215) can receive images simultaneously captured by multiple cameras as input and estimate the geometric relationship between the two images. That is, the road surface information analysis unit (215) can estimate the relative orientation parameters (ROP) between the first image captured by the first camera and the second image captured by the second camera. Here, the relative orientation parameters may include rotation (roll, pitch, yaw) and translation (x, y, z displacement) between the reference coordinate systems of the two cameras.

[0080] To this end, the road surface information analysis unit (215) can extract feature points representing the same object in two images, match the correspondence between the feature points, and then calculate an essential matrix or a fundamental matrix. For example, the road surface information analysis unit (215) can detect distinct feature points such as corners, contours, and lanes in two images using feature point detection algorithms such as SIFT (Scale-Invariant Feature Transform), SURF (Speeded Up Robust Features), or ORB (Oriented FAST and Rotated BRIEF). Subsequently, the road surface information analysis unit (215) can apply a RANSAC (Random Sample Consensus) algorithm to remove incorrectly matched corresponding points, and calculate a relative rotation matrix and a movement vector between the two cameras based on the remaining valid corresponding points to estimate relative position elements.

[0081] The road surface information analysis unit (215) can generate an epipolar image based on the estimated relative position element. Here, the epipolar image may refer to an image in which the two image coordinate systems are transformed and aligned so that it is easy to find the corresponding position of the same object point on the two camera images. To this end, the road surface information analysis unit (215) can normalize the coordinate systems of the two images by realigning them through linear projection transformation so that the same object point is located on the same horizontal line in both images. Through this, the road surface information analysis unit (215) can drastically reduce the amount of computation in the parallax calculation process and increase the correspondence accuracy through epipolar geometry.

[0082] Subsequently, the road surface information analysis unit (215) generates a disparity map by calculating the difference in pixel positions between the two images based on the generated epipolar images. Specifically, when the road surface information analysis unit (215) searches for a corresponding position in the second image for a specific pixel of the first image, it can find the corresponding point with the highest similarity by applying block matching, cost function minimization, or deep learning stereo matching techniques. Here, the difference in horizontal coordinates between corresponding points is the disparity, and the road surface information analysis unit (215) can construct a disparity map by accumulating this on a pixel-by-pixel basis. Subsequently, the road surface information analysis unit (215) can restore depth information of the road surface using the disparity value, the baseline distance between the two cameras, and the focal length.

[0083] Additionally, the road surface information analysis unit (215) can precisely correct the external orientation elements (EOP) using the generated parallax map. Specifically, the road surface information analysis unit (215) can compare the point cloud data calculated based on the parallax map with the initial external orientation elements (EOP), and then re-estimate the camera position and attitude parameters to minimize residual error through a bundle adjustment technique. In this way, the road surface information analysis unit (215) can perform self-calibration to correct internal errors of the camera or deviations caused by the mounting environment.

[0084] Next, the road surface information analysis unit (215) can fuse image and point cloud data received from the information collection device (100) to generate an elevation orthophoto, calculate the elevation difference of the road surface based on this, and perform the function of detecting an area where the calculated elevation difference is greater than or equal to a predefined threshold value as a pothole area.

[0085] Specifically, the road surface information analysis unit (215) can first convert the image captured by the camera and the point cloud data obtained through the lidar into the same ground coordinate system, and then perform the process of aligning the image coordinates and the point cloud coordinates. To this end, the road surface information analysis unit (215) can apply Interior Orientation Parameters (IOP) and Exterior Orientation Parameters (EOP) to correct the image pixel coordinates and the actual ground point cloud coordinates so that they correspond accurately.

[0086] Subsequently, the road surface information analysis unit (215) can generate an elevation orthophoto by fusing the corrected image and point cloud data. Here, the elevation orthophoto is an image corrected to look as if it were taken vertically from above by removing camera tilt, perspective distortion, etc., and can accurately reflect the actual shape and location of the road surface.

[0087] In particular, the road surface information analysis unit (215) can perform various preprocessing steps to utilize the generated elevation orthophoto as training data. For example, noise may occur in images acquired from a road environment due to the influence of lighting, weather, shadows, etc., which can be a factor that hinders the learning performance of an artificial intelligence model. To address this, the road surface information analysis unit (215) can remove noise within the image by applying a Gaussian filter and a bidirectional filter. The Gaussian filter calculates the average by applying the same weight to both pixels close to and far from the filtering target location, thereby effectively mitigating random noise throughout the image.

[0088] Figure 3 is an example illustrating images before and after the application of a Gaussian filter. Figure 3 (A) is an image before filtering, which has a problem where the actual crack pattern appears blurry due to the presence of a large amount of noise on the road surface. On the other hand, Figure 3 (B) is an image with a Gaussian filter applied, in which unnecessary noise components are removed, revealing the actual damage pattern of the road surface (e.g., crack boundaries, fine irregularities) more clearly. By applying a Gaussian filter in this way, the quality of the training data is improved, which can increase the accuracy of height difference analysis and damage area detection performed in subsequent steps.

[0089] Additionally, the road surface information analysis unit (215) may apply a bidirectional filter to maintain the boundaries and crack patterns of the road image more clearly. The bidirectional filter may operate by adjusting the degree of smoothing by considering not only the distance between a reference pixel and neighboring pixels but also the difference in pixel values. Through this, the road surface information analysis unit (215) can suppress overall noise in the image while preserving boundary characteristics such as cracks. Figure 4 (A) is an image before the application of the bidirectional filter, which contains unnecessary reflected light and point noise around the crack boundaries. On the other hand, Figure 4 (B) is an image with the bidirectional filter applied, in which unnecessary noise on the surface is removed while the boundaries of the cracks remain distinct, allowing the actual damage pattern to be clearly identified. In this way, the road surface information analysis unit (215) can apply a bidirectional filter to preserve the structural features of the image during the preprocessing process, thereby increasing the precision of pothole detection and crack recognition performed in subsequent stages.

[0090] Additionally, the road surface information analysis unit (215) can perform a process of correcting the brightness and contrast of the image to ensure homogeneity of the training data. At this time, the road surface information analysis unit (215) can adjust the overall brightness of the image by adding or subtracting a certain value to each pixel value, thereby preventing a decrease in learning performance due to shadows or excessive differences in illumination. (A) of FIG. 5 is an image that has become excessively bright due to the addition of pixel values, in which the boundary of the crack is faintly expressed, and (C) is an image that has become dark due to the subtraction of pixel values, in which the detailed texture is lost. On the other hand, (B) is an image that has been appropriately corrected through a contrast adjustment algorithm, in which the crack boundary is the most distinct and the texture of the background road surface is also stably maintained. In this way, the road surface information analysis unit (215) can reduce the deviation between images captured in various lighting environments by performing preprocessing by optimizing the contrast, and improve the reliability of pothole and crack detection in subsequent stages.

[0091] In addition, the road surface information analysis unit (215) may apply a sharpening algorithm to improve the resolution of the elevation orthophoto image. Here, sharpening can be a technique that maximizes the difference between the target pixel and surrounding pixels to increase the contrast of the boundary line and emphasize edges and structural information within the image. Figure 6 (A) is an image before sharpening is applied, in which the boundary of the crack is expressed somewhat blurry, but (B) is an image with sharpening applied, in which the same crack boundary is clearly emphasized and detailed textures are clearly revealed. Thus, the road surface information analysis unit (215) can achieve the effect of increasing the accuracy of feature extraction during the model training process and improving pothole and crack detection performance by more clearly highlighting fine patterns such as cracks, damage, and irregularities of the road surface included in the training data through the sharpening process.

[0092] Additionally, the road surface information analysis unit (215) may apply a shadow removal preprocessing technique to prevent shadows within the elevation orthophoto from negatively affecting the quality of the training data. Here, shadows are generated by various light sources such as sunlight, streetlights, and vehicles in the road environment, and cause significant differences in brightness even with the same road surface structure, thereby increasing the likelihood of the artificial intelligence model misrecognizing them. For example, (A) and (B) in FIG. 7 are images before shadow removal, and it can be seen that there is a significant difference in brightness between the top and bottom of the same road surface area. In contrast, (C) and (D) are images after shadow removal, and it can be seen that the brightness of the entire image has become uniform as the shadows have been effectively removed. In particular, as can be seen in the red circle marked areas of (B) and (D), damaged parts such as cracks may remain intact even after the shadow removal process has been performed. The road surface information analysis unit (215) can exclude unnecessary optical noise by removing shadows while preserving road surface damage information intact, thereby enabling the acquisition of high-quality elevation orthophotos that can be used as training data.

[0093] The road surface information analysis unit (215) can calculate the height difference of the road surface by comparing and analyzing the generated elevation orthophoto and LiDAR point cloud data. Specifically, after estimating the reference plane of the road surface, the difference between the actual elevation value at each point and the reference plane can be calculated. At this time, there is almost no height difference in flat asphalt sections, but a large local height difference may occur in sections where potholes or cracks exist.

[0094] For example, the road surface information analysis unit (215) can detect a sunken area with a diameter of 30 cm and a depth of 5 cm by comparing the elevation of a point cloud in a nearby area based on a flat surface in the center of the lane. Conversely, if the height difference is less than a threshold value, the road surface information analysis unit (215) can determine that it is simple road surface wear or surface roughness.

[0095] Additionally, the road surface information analysis unit (215) can finally identify an area where the calculated height difference is greater than or equal to a preset value as a pothole area. Here, the threshold value can be dynamically adjusted by considering the policy of the road management agency, traffic safety standards, vehicle driving stability, etc.

[0096] In the following configuration, the transmission plan establishment unit (220) can perform the function of establishing a plan to efficiently select and transmit data related to the area when a pothole area is detected by the road surface information analysis unit (215).

[0097] Specifically, the transmission plan establishment unit (220) can preferentially and selectively extract frames containing a pothole area from among all video frames. In this process, the transmission plan establishment unit (220) can select only the core frames related to the detected damaged area without transmitting the entire unnecessary frame by referring to analysis results such as the location coordinates, size, shape, and frequency of occurrence of the pothole. For example, if the same pothole is included in multiple consecutive frames, the transmission plan establishment unit (220) can reduce data redundancy and increase transmission efficiency by selecting the most clearly identifiable representative frame and transmitting it to the management server (300).

[0098] Additionally, the transmission plan establishment unit (220) may add location information and metadata of the corresponding pothole area along with the selected frame. Here, the location information may include GPS (Global Positioning System) coordinates, geographic administrative district codes, road section identifiers, etc., and the metadata may include the time of occurrence, estimated depth, area, shape classification, and analysis reliability index of the detected pothole. For example, if a pothole with a diameter of approximately 25 cm and a depth of 5 cm or more is detected in a specific road section, the transmission plan establishment unit (220) may package metadata such as "Occurrence location: Latitude 37.1234, Longitude 127.5678 / Time of occurrence: 2025.08.20 10:00 / Estimated depth: 5 cm / Area: 25X30 cm / Reliability: 92%" along with the image frame in which the pothole exists and transmit it to the management server (300).

[0099] That is, since communication delays or transmission instability may occur when transmitting large amounts of data in real time, the transmission planning unit (220) may assign priorities to pothole detection results and transmit data with high importance first. For example, the transmission planning unit (220) may set large potholes that cause a direct risk to road safety or damage to major roads as urgent transmission targets, and transmit relatively minor damage to the management server using a batch transmission method.

[0100] Through this, the transmission planning unit (220) can support the management server (300) in quickly securing key information directly related to road safety even if it does not receive the entire video data. This can provide the effect of maximizing road management efficiency by enabling the rapid identification of sections requiring emergency repair work while minimizing network load.

[0101] Additionally, the transmission plan establishment unit (220) can perform the function of calculating a risk index based on the shape of the detected pothole area and determining the priority of the pothole area according to the risk index.

[0102] Specifically, the transmission plan establishment unit (220) can receive data on the external characteristics of a pothole provided by the road surface information analysis unit (215) and calculate a risk index. Here, the shape of the pothole can be defined by indicators such as area, depth, perimeter length, circularity, aspect ratio, and boundary irregularity. For example, since a pothole with a large diameter, deep depth, and irregular shape poses a greater risk to vehicle driving stability, the transmission plan establishment unit (220) can assign a high risk index accordingly.

[0103] Additionally, the transmission planning unit (220) can determine the transmission priority of each pothole area according to the calculated risk index. For example, if the risk index is above a certain threshold value, the transmission planning unit (220) can classify the pothole as a high-risk group and set it to be transmitted immediately to the management server (300). On the other hand, for potholes with low risk, the transmission planning unit (220) can transmit in batches after a certain period of time or transmit them in a bundled form together with other damage data. In this way, by applying a risk-based transmission policy, the transmission planning unit (220) can simultaneously support the efficient use of network bandwidth and the emergency response of the road management agency.

[0104] The transmission plan establishment unit (220) can transmit not only the location coordinates and metadata of the pothole as transmission target data, but also video frames containing the pothole area, point cloud data, and a summary report of analysis results. In particular, in the case of high-risk potholes, the transmission plan establishment unit (220) can transmit the original video frames and precise point cloud data together so that the management server (300) can quickly identify the situation.

[0105] For example, if a pothole with a diameter of about 50 cm and a depth of 10 cm or more is detected in an urban highway section, the transmission planning unit (220) can calculate the risk index as ‘High Risk’ and transmit video frames containing the pothole and related data to the management server (300) in real time. On the other hand, if a shallow pothole with a diameter of 10 cm or less is detected in a small residential road, it can be classified as ‘Low Risk’ to delay transmission or transmit it in batches along with other data.

[0106] In this way, the transmission plan establishment unit (220) can quantitatively calculate the risk level based on the shape of the pothole and dynamically adjust the data transmission priority according to the risk level index, thereby enabling the road management server to selectively receive damage information that is important for actual traffic safety.

[0107] Additionally, the transmission plan establishment unit (220) can quantitatively calculate a risk index based on the average values ​​of the area and depth of the pothole area. For example, even for potholes with the same area, the deeper the depth, the greater the impact on the vehicle tires and suspension, so the average depth value is reflected as an important factor in the risk calculation. At this time, the transmission plan establishment unit (220) can calculate a weighted risk by combining the two-dimensional area and the average depth value of the pothole, and the calculated risk can be directly used to determine the priority transmitted to the management server (300).

[0108] Furthermore, the transmission planning unit (220) can divide the entire road into multiple areas according to risk levels and then evaluate the location of a specific pothole in conjunction with the corresponding area. For example, the transmission planning unit (220) can assign a relatively higher risk level to potholes of the same size and depth in sections with high traffic volume and high driving speeds, such as highways, national roads, and urban arterial roads. In this case, the transmission planning unit (220) assigns weights to the risk index by reflecting the pre-set risk coefficient of the area to which the pothole is located, thereby allowing damage to sections critical to traffic safety to be reported to the management server (300) first.

[0109] For example, if potholes of the same size are found on a residential side road and a highway, respectively, the transmission planning unit (220) may assign a higher risk weight to the pothole in the highway section and increase the priority to transmit the data immediately. In this way, the transmission planning unit (220) can perform a precise risk assessment that reflects the actual traffic environment and locational importance, going beyond a simple risk assessment based on pothole size.

[0110] Additionally, when a pothole area is detected, the transmission plan establishment unit (220) may transmit summarized meta information to the management server (300) first, instead of transmitting all data related to the area in bulk. Here, the meta information may consist of summary data such as the location coordinates of the pothole, time of occurrence, area, depth range, frame number, and reliability index. In this way, the transmission plan establishment unit (220) can reduce network load and support the management server in efficiently allocating data processing resources by prioritizing the transmission of only key information without immediately transmitting large volumes of image and point cloud data.

[0111] Here, the management server (300) may transmit a data request signal to the analysis server (200) if it determines that detailed data regarding a specific pothole is needed based on the received metadata. At this time, the management server (300) may decide whether to request additional image and point cloud data by comprehensively considering the pothole risk index, traffic volume at the location, recent maintenance records, etc.

[0112] When the transmission plan establishment unit (220) receives a data request signal from the management server (300), it may additionally transmit at least one original frame containing a pothole or LiDAR point cloud data that matches the frame. For example, if summary data such as "occurrence of a pothole with a diameter of about 30 cm and a depth of about 6 cm at location latitude 37.1234, longitude 127.5678" is transmitted in the initial metadata, and the management server requests "original data because the section is a major arterial road," the transmission plan establishment unit (220) may retransmit the original image and precise point cloud data of the location so that the management server (300) can perform a more precise damage determination.

[0113] In this way, the transmission planning unit (220) can adjust the amount of transmission data in stages to prevent unnecessary bandwidth waste and support the management server (300) in obtaining additional detailed data only for emergency situations or high-importance ports.

[0114] Additionally, the transmission plan establishment unit (220) may calculate a reliability index to quantitatively evaluate the reliability of the analysis results for the pothole area. The reliability index may be calculated by combining the degree of agreement between the image analysis results and the point cloud-based depth estimation results, the accuracy of feature point matching, the noise level of the disparity map, and image quality indicators after shadow removal and filtering. For example, if a specific pothole is detected in the image but no distinct height difference appears in the point cloud data, the reliability index may be evaluated as low; conversely, if a consistent damage pattern is confirmed at the same location in both the image and the point cloud data, the reliability index may be calculated as high.

[0115] Based on the calculated reliability index, the transmission plan establishment unit (220) can determine the transmission level. When the reliability index is high, all data, such as original video frames, point cloud data, location information, and metadata, can be transmitted to the management server to support rapid repair decisions. For example, when the reliability index is at an intermediate level, the transmission plan establishment unit (220) may first transmit only representative frames and summarized metadata, and transmit additional data upon request from the management server (300) if necessary. When the reliability index is low, the transmission plan establishment unit (220) may transmit only minimal metadata so that the management server can perform additional verification of the same location or cross-examination with the results of other equipment.

[0116] In this way, the transmission planning unit (220) can reduce unnecessary large-capacity data transmission, reduce incorrect repair instructions due to analysis errors, and increase overall data processing efficiency.

[0117] Additionally, the transmission plan establishment unit (220) can perform a multi-stage transmission strategy based on the calculated reliability index. First, if the reliability index is calculated to be higher than the first reference value, the transmission plan establishment unit (220) determines that the analysis result is sufficiently reliable and can transmit only the location information of the pothole area and summarized metadata to the management server (300). For example, the transmission plan establishment unit (220) can support the management server (300) in establishing a repair plan while minimizing network load by transmitting only summary data such as the coordinates, time of occurrence, and area size of the pothole.

[0118] Alternatively, the transmission plan establishment unit (220) can provide more specific data when the reliability index is evaluated to be lower than the first standard value but higher than the second standard value. That is, the transmission plan establishment unit (220) can selectively extract images of a Region of Interest (ROI) that includes a pothole area and transmit point cloud data corresponding to the images together. Through this, the transmission plan establishment unit (220) can support additional verification by enabling the management server (300) to check the environment around the pothole, and at the same time, reduce the burden of large-capacity data resulting from the transmission of the entire original frame.

[0119] Additionally, the transmission plan establishment unit (220) may determine that the uncertainty of the analysis result is high when the reliability index is calculated to be lower than the second reference value. In this case, the transmission plan establishment unit (220) may transmit at least one extracted whole frame and the point cloud data corresponding to that frame directly to the management server (300). Through this, the transmission plan establishment unit (220) can support the management server in directly verifying the original image and the 3D point cloud data, and cross-verifying with the results of other analysis algorithms or auxiliary devices.

[0120] In this way, the transmission plan establishment unit (220) can efficiently utilize the communication bandwidth and improve the judgment accuracy of the management server (300) by differentiating the level of data transmitted according to the value of the reliability index.

[0121] In the following configuration, the storage unit (225) can store raw images and point cloud data received from the information collection device (100), elevation orthophotos processed by the road surface information analysis unit (215), parallax maps, and analysis results such as the location, shape, and risk index of the detected pothole area.

[0123] Below, hardware for implementing the logical components of the analysis server (200) as described above will be explained in more detail.

[0124] FIG. 8 is a hardware configuration diagram of an analysis server according to one embodiment of the present invention.

[0125] As illustrated in FIG. 8, the analysis server (200) may be configured to include a processor (250), memory (255), transceiver (260), input / output device (265), data bus (270) and storage (275).

[0126] Specifically, the processor (250) can implement the operation and function of the analysis server (200) based on instructions according to the software (280a) which implements a real-time road surface information collection method residing in the memory (255).

[0127] Software (280a) in which a real-time road surface information collection method according to embodiments of the present invention is implemented may be loaded in the memory (255).

[0128] The transceiver (260) can transmit and receive data with the information collection device (100) and the management server (300).

[0129] The input / output device (265) can receive signals necessary for the operation of the analysis server (200) or output calculation results to the outside according to the command of the processor (250).

[0130] The data bus (270) is connected to the processor (250), memory (255), transceiver (260), input / output device (265), and storage (275), and can serve as a passage for transferring data between each component.

[0131] Storage (275) can store an Application Programming Interface (API), library files, resource files, etc., necessary for the execution of software (280a) in which a real-time road surface information collection method according to embodiments of the present invention is implemented. Additionally, storage (275) can store software (280b) and a database (285) in which a method according to embodiments of the present invention is implemented.

[0132] According to one embodiment of the present invention, software (280a, 280b) for implementing a real-time road surface information collection method that resides in memory (255) or is stored in storage (275) may be a computer program recorded on a recording medium to execute the steps of: a processor (250) collecting road surface information in real time from an information collection device that moves along a road and collects information about the road surface; a processor (250) analyzing the condition of the road surface based on the real-time collected road surface information; and a processor (250) transmitting road surface condition information according to the analyzed road surface condition to a management server in real time.

[0133] More specifically, the processor (250) may include an Application-Specific Integrated Circuit (ASIC), other chipsets, logic circuits, and / or data processing devices. The memory (255) may include Read-Only Memory (ROM), Random Access Memory (RAM), flash memory, memory cards, storage media, and / or other storage devices. The transceiver (260) may include a baseband circuit for processing wired and wireless signals. The input / output device (265) may include input devices such as a keyboard, mouse, and / or joystick, image output devices such as a Liquid Crystal Display (LCD), Organic Light Emitting Diode (OLED), and / or Active Matrix OLED (AMOLED), and printing devices such as a printer and plotter.

[0134] When the embodiments included in this specification are implemented in software, the above-described method may be implemented as a module (process, function, etc.) that performs the above-described function. The module may reside in memory (255) and be executed by a processor (250). Memory (255) may be located inside or outside the processor (250) and may be connected to the processor (250) by various widely known means.

[0135] Each component illustrated in FIG. 8 may be implemented by various means, for example, hardware, firmware, software, or a combination thereof. In the case of implementation by hardware, one embodiment of the present invention may be implemented by one or more ASICs (Application Specific Integrated Circuits), DSPs (Digital Signal Processors), DSPDs (Digital Signal Processing Devices), PLDs (Programmable Logic Devices), FPGAs (Field Programmable Gate Arrays), processors, controllers, microcontrollers, microprocessors, etc.

[0136] In addition, in the case of implementation by firmware or software, an embodiment of the present invention may be implemented in the form of a module, procedure, function, etc., that performs the functions or operations described above, and may be recorded on a recording medium readable through various computer means. Here, the recording medium may include program instructions, data files, data structures, etc., either alone or in combination. The program instructions recorded on the recording medium may be those specifically designed and configured for the present invention, or they may be those known and available to those skilled in the art of computer software. For example, the recording medium includes magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs (Compact Disk Read Only Memory) and DVDs (Digital Video Disks); magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of program instructions may include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc. Such hardware devices may be configured to operate as one or more software to perform the operation of the present invention, and vice versa.

[0138] FIG. 9 is a flowchart illustrating a method for collecting real-time road surface information according to an embodiment of the present invention.

[0139] Referring to FIG. 9, first, in step S100, the analysis server can collect road surface information in real time, including images captured by a camera and point cloud data acquired through a lidar from an information collection device moving on the road. In addition, the analysis server can receive metadata such as location coordinates, time information, and vehicle speed along with the images and point cloud data and use them for subsequent analysis.

[0140] Next, in step S200, the analysis server can analyze the condition of the road surface based on the collected road surface information.

[0141] Specifically, the analysis server can identify frames containing ground point coordinates (GCPs) within the image, use them to estimate external orientation elements (EOPs) representing the camera's position and attitude, and then perform bundle correction to ensure geometric consistency.

[0142] In addition, the analysis server can estimate relative position factors (ROP) from multiple camera images, generate an epipolar image based on this, and then calculate a pixel-level parallax map to restore depth information and self-correct external position factors.

[0143] Next, the analysis server can generate an elevation orthophoto by fusing image and point cloud data, and calculate the height difference of the road surface from the generated image to detect areas exceeding a preset threshold as potholes. The analysis server can selectively extract frames containing the detected pothole areas and prepare the location coordinates of the potholes and related metadata along with the frames as data to be transmitted to the management server.

[0144] In addition, at the S300 stage, the analysis server can transmit the analyzed road surface condition information to the management server in real time.

[0145] In this process, the analysis server can calculate a risk index based on the shape of the detected potholes and determine the priority of risk to prioritize the transmission of frames and data with high importance. Here, the risk index can be calculated based on the average values ​​of the pothole's area and depth, and the analysis server can perform a more precise risk assessment by dividing the road into multiple regions according to risk and applying weights based on the importance of each region.

[0146] In addition, the analysis server can first transmit metadata regarding potholes, and then additionally transmit relevant frames and point cloud data upon receiving a data request signal from the management server. Furthermore, the analysis server can determine the transmission level by calculating the reliability index of the analysis results; by transmitting only summary information when the reliability is high, transmitting region of interest images and point cloud data when the reliability is medium, and transmitting the entire frame and point cloud data when the reliability is low, it can simultaneously ensure both efficiency and reliability of data transmission.

[0148] As described above, preferred embodiments of the present invention have been disclosed in this specification and drawings; however, it is obvious to those skilled in the art that other variations based on the technical spirit of the present invention are possible in addition to the embodiments disclosed herein. Furthermore, although specific terms have been used in this specification and drawings, they are used merely in a general sense to facilitate the explanation of the technical content of the present invention and to aid in understanding the invention, and are not intended to limit the scope of the present invention. Accordingly, the detailed description above should not be interpreted restrictively in any respect and should be considered illustrative. The scope of the present invention should be determined by a reasonable interpretation of the appended claims, and all modifications within the equivalent scope of the present invention are included within the scope of the present invention. Explanation of the symbols

[0149] 100: Information gathering device 200: Analysis server 205 : Communication unit 210 : Input / Output unit 215: Road Surface Information Analysis Unit 220: Transmission Plan Establishment Unit 225 : Storage 300 : Management Server

Claims

Claim 1 A step in which an analysis server collects road surface information in real time from an information collection device that moves along a road and collects information regarding the road surface; a step in which the analysis server analyzes the condition of the road surface based on the real-time collected road surface information; and a step in which the analysis server transmits road surface condition information according to the analyzed road surface condition to a management server in real time. The method is characterized by including the following steps: the step of collecting road surface information is characterized by collecting an image of a road surface captured by a camera and point cloud data acquired by laser pulses reflected from the road surface through a lidar installed together with the camera; the step of analyzing is characterized by aligning the time points between the data based on timestamps included in the image and the point cloud data, and performing a preprocessing process that aligns to a single ground reference coordinate system by applying external parameters between the lidar and the camera; converting the image and the point cloud data to the same ground coordinate system, aligning the image coordinates and the point cloud coordinates, and correcting so that the image pixel coordinates correspond to the actual ground point cloud coordinates by applying internal orientation parameters (IOP) and external orientation parameters (EOP); the step of analyzing is characterized by identifying at least one frame containing pre-surveyed ground control point (GCP) coordinates within the image, estimating the external orientation parameters (EOP) of the identified at least one frame based on the ground control point coordinates, and the A real-time road surface information collection method characterized by performing bundle adjustment to optimize the position and attitude of the camera by simultaneously considering multiple video frames of external expression elements. Claim 2 A method for collecting real-time road surface information according to claim 1, wherein the analyzing step comprises estimating relative orientation parameters (ROP) between a first image captured by a first camera and a second image captured by a second camera, generating an epipolar image based on the estimated relative orientation parameters, generating a disparity map from the generated epipolar image, and self-calibrating external orientation parameters (EOP) through bundle adjustment based on the generated disparity map. Claim 3 A method for collecting real-time road surface information according to claim 1, wherein the analyzing step comprises generating an elevation orthophoto based on the image and point cloud data, calculating the height difference of the road surface from the generated elevation orthophoto, and detecting an area where the calculated height difference is greater than or equal to a preset value as a pothole area. Claim 4 A method for collecting real-time road surface information according to claim 3, wherein the transmitting step is characterized by selectively extracting at least one frame in which the pothole area is detected, and transmitting the location information and metadata of the pothole area together with the extracted at least one frame to the management server. Claim 5 A method for collecting real-time road surface information according to claim 4, wherein the transmitting step calculates a risk index based on the shape of the detected pothole area, determines the priority of the pothole area according to the calculated risk index, and transmits a frame containing the pothole area and related data to the management server according to the determined priority. Claim 6 A method for collecting real-time road surface information according to claim 5, wherein the transmitting step comprises calculating a risk index based on the area of ​​the pothole area and the average depth value for the area, dividing the road into multiple areas according to the risk level, identifying the location of the pothole area within the divided areas, and assigning weights to the risk index according to a preset risk level for the identified location. Claim 7 A method for collecting real-time road surface information according to claim 4, wherein the transmitting step is characterized by first transmitting meta-information regarding the detected pothole area to a management server, and, when a data request signal regarding the pothole area is received from the management server in response to the meta-information, additionally transmitting at least one extracted frame. Claim 8 In claim 4, the transmitting step is characterized by calculating a reliability index of the analysis result for the pothole area, determining a transmission level according to the calculated reliability index, and selectively transmitting at least one extracted frame, location information of the pothole area, and metadata to the management server according to the determined transmission level, wherein if the reliability index is greater than or equal to a first reference value, only the location information of the pothole area and summary metadata are transmitted, if the reliability index is less than the first reference value and greater than or equal to a second reference value, an image of a Region of Interest (ROI) containing the pothole area and point cloud data corresponding to the image are transmitted, and if the reliability index is less than the second reference value, at least one extracted frame and point cloud data corresponding to the frame are transmitted. Claim 9 delete Claim 10 delete

Images