A mobile phone terminal highway slope disease positioning method, system, device, medium and program product based on a curvilinear coordinate system

Abstract

The application discloses a kind of mobile phone terminal highway slope disease positioning method, system, equipment, medium and program product based on curve coordinate system, comprising: mobile phone terminal is carried out multi-source data acquisition, forms original track point set;Smooth trajectory pretreatment, obtain the smooth inspection trajectory;Fitting and carrying out line parameter registration, construct fitting path and stake number system mapping, establish Frenet curve coordinate system;Carry out key section feature point guide labeling, obtain section feature point set with feature type label;Characteristic point is projected to Frenet curve coordinate system learning and updates section threshold sequence;Disease observation point is carried out sliding window matching, obtains the optimal projection position of disease point in Frenet curve coordinate system;Determine side and slope / level;Output standardization result.The application can be in the case where lacking existing center line data, design linear or high-precision template data, accurately position, grade and standard output low cost to highway slope disease.

Description

Technical Field

[0001] This invention relates to the field of highway slope disease location technology, and in particular to a mobile-based highway slope disease location method, system, equipment, medium and program product based on a curve coordinate system. Background Technology

[0002] Existing methods for locating highway slope defects typically fall into the following categories. The first category relies on existing road centerlines, design drawings, stationing data, or a GIS base, projecting the on-site location points onto the existing linear parameter system. This method is suitable for trunk roads with complete data, but is often unsuitable for inspection scenarios lacking standard centerlines. The second category relies on RTK, total stations, or other dedicated high-precision surveying equipment. While offering high accuracy, this method is costly, complex in its operation, and not easily applicable to low-cost, high-frequency maintenance inspections. The third category relies solely on mobile phone GNSS planar coordinates or photographic locations for rough positioning. This method is usable on straight road sections, but prone to stationing errors and layer confusion in continuous curves, obstructed environments, valley environments, or multi-level platform slope scenarios.

[0003] Therefore, the existing technology has at least the following drawbacks: First, it relies heavily on existing centerlines or station templates and cannot adapt to unknown alignments or road sections with missing data; second, it is not robust enough to complex curves and weak signal environments; third, slope feature points still require a lot of manual identification, which is inefficient and highly subjective; and fourth, the output results are not uniform and it is difficult to form a standardized, machine-readable, and traceable expression.

[0004] For the reasons mentioned above, there is an urgent need in this field for a low-cost method for locating highway slope defects that does not rely on existing centerlines, can independently establish a curve coordinate system via mobile devices, and can accurately locate and output standardized information about slope defects. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a method, system, device, medium and program product for locating highway slope defects on a mobile phone based on a curve coordinate system, which at least partially solves the problems existing in the prior art.

[0006] Other features and advantages of the invention will become apparent from the following detailed description, or may be learned in part by practice of the invention.

[0007] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0008] According to a first aspect of the present invention, the present invention provides a method for locating highway slope defects on a mobile phone based on a curvilinear coordinate system, comprising:

[0009] The inspection mobile app collects multi-source data to form a set of original trajectory points.

[0010] The original trajectory point set is subjected to smooth trajectory preprocessing to obtain the smoothed inspection trajectory.

[0011] The smoothed inspection trajectory is fitted and line parameter registration is performed to construct the mapping between the fitted path and the stationing system, and a Frenet curve coordinate system is established.

[0012] Perform guided annotation of key cross-sectional feature points to obtain a set of cross-sectional feature points with feature type labels;

[0013] The feature points are projected onto the Frenet curve coordinate system to learn and update the cross-section threshold sequence;

[0014] Sliding window matching is performed on the disease observation points to obtain the optimal projection position of the disease points in the Frenet curve coordinate system;

[0015] Determine the side based on the projection location, and determine the slope / platform level based on the cross-section threshold range and level type;

[0016] Output standardized results.

[0017] Furthermore, the inspection mobile terminal performs multi-source data collection to form an original trajectory point set, including: the inspection mobile terminal slowly moves along the road section to be inspected, and collects GNSS coordinates, IMU three-axis acceleration and angular velocity, magnetic orientation, captured images and timestamp information at a unified clock to form an original trajectory point set.

[0018] Furthermore, the step of performing smoothing trajectory preprocessing on the original trajectory point set to obtain the smoothed inspection trajectory includes: performing time synchronization, outlier removal, and filtering smoothing on the original trajectory point set to obtain the smoothed inspection trajectory.

[0019] Furthermore, the process of fitting the smoothed inspection trajectory and performing line parameter registration to construct the mapping between the fitted path and the station number includes: dividing the smoothed inspection trajectory into curvature segments or mass segments, performing weighted spline fitting within each segment to obtain the fitted path; and using two or more known station number control points, known landmark points, or on-site manual confirmation points to establish a mapping relationship between the fitted path and the station number system; when the number of control points is small, piecewise linear registration is used; when the number of control points is large, least squares fitting or spline mapping is used.

[0020] Furthermore, the step of guiding the annotation of key cross-sectional feature points to obtain a set of cross-sectional feature points with feature type labels includes: the inspection mobile terminal prompts and guides inspection personnel to collect key cross-sectional feature points at preset pile sections or cross-sectional locations with large curvature changes. The key cross-sectional feature points include slope top, platform boundary, slope toe, ditch edge, and retaining wall top edge; a candidate point list is generated based on image edges, local pose changes, local elevation changes, and existing cross-sectional templates. The candidate points are displayed in the form of "feature type + confidence score", and are manually confirmed or corrected with a single click to obtain a set of cross-sectional feature points with feature type labels.

[0021] Furthermore, the step of projecting feature points onto the Frenet curve coordinate system to learn and update the cross-section threshold sequence includes: projecting the set of cross-section feature points onto the Frenet curve coordinate system, calculating the lateral reference distance of each feature point relative to the reference baseline, and maintaining the threshold sequence according to the road side and station position; the threshold sequence is used to describe the slope / platform boundary positions that appear sequentially from the road reference boundary outwards.

[0022] Furthermore, the steps of guiding the annotation of key cross-section feature points and projecting the feature points onto the Frenet curve coordinate system to learn and update the cross-section threshold sequence are performed during the first modeling, cross-section model update, or when accuracy needs to be improved; if a stable cross-section threshold library has been established for a certain road segment, the historical model is directly called in the retest scenario.

[0023] Furthermore, the step of performing sliding window matching on the disease observation points to obtain the optimal projection position of the disease points in the Frenet curve coordinate system includes: when a disease observation event is received, constructing candidate projection intervals based on the current GNSS position, IMU integrated mileage, magnetic azimuth and adjacent observations, and using a sliding window-based hidden Markov model to obtain the optimal matching position, thereby obtaining the arc length position and normal offset of the disease observation point on the fitted path in the Frenet curve coordinate system;

[0024] The determination of side based on projection position and the determination of slope / platform level based on cross-sectional threshold interval and level type include: determining whether the disease is located on the left or right side based on the sign of the normal offset; determining whether the disease belongs to a first-level slope, first-level platform, second-level slope or other special level based on the interval relationship between the lateral reference distance of the disease point and the cross-sectional threshold sequence, and in conjunction with the level mapping table.

[0025] Furthermore, the step of using a sliding window-based hidden Markov model to find the optimal matching position includes:

[0026] The observation probability is jointly given by the lateral distance residual, the heading difference, and the sensor mass. The optimal state transition probability is given by the consistency of the arc length increment between adjacent time moments, the curvature continuity, and the velocity constraint. Based on the preset sliding window length, the observation probability and the state transition probability are jointly solved to output the optimal matching position.

[0027] Furthermore, before outputting the standardized results, a confidence level is calculated, which is calculated by combining the matching residual, GNSS quality, and threshold posterior variance.

[0028] Furthermore, the standardized output results include the output of "station number + side identification + slope / platform level + confidence level".

[0029] According to a second aspect of the present invention, a mobile terminal-based highway slope defect location system based on a curvilinear coordinate system is provided, comprising:

[0030] The data acquisition module is configured to collect multi-source data using the inspection mobile app to form a set of original trajectory points.

[0031] The trajectory preprocessing module is configured to perform smooth trajectory preprocessing on the original trajectory point set to obtain the smoothed inspection trajectory.

[0032] The fitting path self-calibration and line parameter registration module is configured to fit the smoothed inspection trajectory and perform line parameter registration, construct the mapping between the fitting path and the stationing system, and establish the Frenet curve coordinate system.

[0033] The feature point guided annotation module is configured to perform guided annotation of key cross-sectional feature points, and obtain a set of cross-sectional feature points with feature type labels;

[0034] The cross-section threshold learning and update module is configured to project feature points onto the Frenet curve coordinate system to learn and update the cross-section threshold sequence.

[0035] The disease projection positioning module is configured to perform sliding window matching on the disease observation points to obtain the optimal projection position of the disease points in the Frenet curve coordinate system.

[0036] The side identification and level determination module is configured to determine the side identification based on the projection position and the slope / platform level based on the cross-section threshold range and level type.

[0037] The standardization output module is configured to output standardized results.

[0038] According to a third aspect of the present invention, a computer device is provided, the device comprising: a processor and a memory;

[0039] The memory is used to store one or more program instructions;

[0040] The processor is used to run one or more program instructions to execute the above-described method for locating highway slope defects on a mobile device based on a curve coordinate system.

[0041] According to a fourth aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, wherein when the computer program is executed by a processor, the above-described method for locating highway slope defects on a mobile device based on a curvilinear coordinate system is implemented.

[0042] According to a fifth aspect of the present invention, a computer program product is provided, the computer program product including computer program instructions, which, when executed by a processor, implement the above-described method for locating highway slope defects on a mobile device based on a curvilinear coordinate system.

[0043] By adopting the above technical solution, the present invention has at least the following beneficial effects:

[0044] This invention uses a mobile inspection terminal as the data carrier. By collecting multi-source data such as GNSS, IMU, magnetic azimuth, and images, it autonomously fits and establishes a Frenet curve coordinate system. Subsequently, it combines key cross-sectional feature points such as slope tops, platform boundaries, and ditch edges collected under the guidance of inspection personnel to learn cross-sectional threshold sequences online. Finally, it projects the disease observation points onto this curve coordinate system to complete the output of station number, side identification, slope / platform level, and confidence level. This invention can provide low-cost, accurate location, classification, and standardized output of highway slope diseases even in the absence of existing centerline data, design alignment, or high-precision template data. Attached Figure Description

[0045] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0046] Figure 1 A flowchart of a method for locating highway slope defects on a mobile device based on a curve coordinate system, provided in an embodiment of the present invention;

[0047] Figure 2 This is a schematic diagram of curve coordinate projection provided for an embodiment of the present invention. Detailed Implementation

[0048] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0049] like Figure 1 As shown, this embodiment provides a method for locating highway slope defects on a mobile phone based on a curved coordinate system, including steps S101, S102, S103, S104, S105, S106, S107 and S108.

[0050] Step S101: The inspection mobile device collects multi-source data to form the original trajectory point set.

[0051] This step mainly utilizes the mobile inspection terminal as a data collection platform to collect multi-source raw trajectory data, providing a raw data foundation for subsequent path fitting.

[0052] Specifically, the mobile inspection device moves slowly (by driving or walking) along the section of road to be inspected, and collects GNSS coordinates, IMU three-axis acceleration and angular velocity, magnetic orientation, captured images and timestamp information using a unified clock to form an original trajectory point set Pi. The original trajectory point set Pi may also include speed and accuracy factors.

[0053] Step S102: Perform smooth trajectory preprocessing on the original trajectory point set to obtain the smoothed inspection trajectory.

[0054] This step is mainly used as a foundational trajectory preprocessing step for stabilizing subsequent fitting and matching.

[0055] Specifically, the original trajectory point set undergoes time synchronization, outlier removal, and filtering smoothing to obtain a smoothed inspection trajectory. Outliers can be removed based on rules such as sudden velocity changes, sudden acceleration changes, anomalies in accuracy factors, and sudden angle jumps between adjacent points. Filtering can employ first-order or second-order Kalman filtering, complementary filtering, or a sliding window-based smoothing strategy. The output of this step is the smoothed inspection trajectory point sequence and a multi-source observation sequence under a unified time base.

[0056] Step 103: Fit the smoothed inspection trajectory and perform line parameter registration to construct the mapping between the fitted path and the station number system, and establish the Frenet curve coordinate system.

[0057] This step, based on the aforementioned steps, mainly involves performing path fitting self-calibration and line parameter registration for road sections with unknown alignment types, establishing a Frenet curve coordinate system, and solving the problem of not being able to establish line parameters for road sections without centerline data.

[0058] Specifically, the smoothed inspection trajectory is segmented by curvature or quality. For example, the trajectory points can be segmented based on GNSS quality, heading changes, and curvature changes. Weighted spline fitting is performed within each segment to obtain the fitted path γ(s). For example, a weighted spline objective function is constructed within each segment, assigning greater weight to points with higher quality and less weight to points with lower quality, resulting in a smooth fitted path γ(s) that still maintains the true curve shape. A mapping relationship K(s) between the fitted path and the stationing system is established using two or more known station control points, known landmark points, or on-site manually confirmed points. When the number of control points is small, piecewise linear registration is used; when the number of control points is large, least squares fitting or spline mapping is used to reduce local errors.

[0059] Through the above processing, even if there is no readily available centerline data in the original scene, it is still possible to establish a self-calibrating reference path and line parameter system based on the inspection trajectory.

[0060] Step S104: Perform guided annotation of key cross-sectional feature points to obtain a set of cross-sectional feature points with feature type labels.

[0061] This step is a guided feature point acquisition step, which enables the establishment of cross-sectional structure models for road sections without available data.

[0062] Specifically, the mobile inspection app guides inspectors to collect key cross-sectional feature points such as slope tops, platform boundaries, slope toes, ditch edges, and retaining wall tops at preset pile sections or sections with significant curvature changes. Collection methods include clicking to annotate, voice confirmation, photo triggering, or selecting points on an image. This step outputs a set of cross-sectional feature points Fj labeled with feature type.

[0063] Preferably, the feature point guided annotation process also includes automatic candidate feature point extraction.

[0064] When collecting key cross-sectional feature points, on the one hand, manual confirmation of feature points is allowed; on the other hand, a candidate point list is generated based on image edges, local pose changes, local elevation changes, and existing cross-sectional templates. Candidate points are displayed in the form of "feature type + confidence score," and can be confirmed or corrected manually with a single click to obtain a set of cross-sectional feature points labeled with feature type. This approach retains human interpretability while reducing the workload of purely manual point-by-point annotation.

[0065] Step S105: Project the feature points onto the Frenet curve coordinate system to learn and update the cross-section threshold sequence;

[0066] This step belongs to the section threshold learning and update step, which can solve the problem of slope level or platform stability determination.

[0067] Specifically, the set of cross-sectional feature points Fj is projected onto the Frenet curve coordinate system, the lateral reference distance xh of each feature point relative to the reference baseline is calculated, and a threshold sequence {Wk(s,c)} is maintained according to the road side c and the station position s; the threshold sequence is used to describe the slope / platform boundary positions that appear sequentially from the road reference boundary outward.

[0068] Steps S104 and S105 are preferred to be executed when modeling for the first time, updating the cross-section model, or when improving accuracy is required. If a relatively stable cross-section threshold library has been established for a certain road segment, the historical model is directly called in the retest scenario, and steps S101, S102, S103, S106, S107 and S108 are executed.

[0069] Step 106: Perform sliding window matching on the disease observation points to obtain the optimal projection position of the disease points in the Frenet curve coordinate system.

[0070] This step belongs to the trajectory matching and disease projection step, combining the sliding window with the fitted path and cross-sectional model. (Reference) Figure 2 As shown.

[0071] Specifically, when a defect observation event is received, such as when an inspection personnel discover a defect and take photos, select points, or record voice messages, a candidate projection interval is constructed based on the current GNSS position, IMU integral mileage, magnetic azimuth, and adjacent observations. The optimal matching position is obtained by using a sliding window-based hidden Markov model matching, which belongs to the sliding window shortest path algorithm. The arc length position s* and normal offset y* of the defect observation point on the fitted path in the Frenet curve coordinate system are obtained.

[0072] More specifically, robust projection localization of fault points can be achieved by using a Hidden Markov Model (HMM) based on a sliding window to find the optimal matching position. This includes: when a fault observation event is received, multiple candidate projection intervals st are generated around the current GNSS point on the fitted path γ(s). The observation probability is jointly given by the lateral distance residual, heading difference, and sensor quality. The optimal state transition probability is given by the consistency of arc length increments between adjacent time steps, curvature continuity, and velocity constraints. Based on a preset sliding window length, the shortest path approximation is solved by jointly inputting the observation probability and the state transition probability, and the optimal matching position is output.

[0073] Step S107: Determine the side based on the projection position, and determine the slope / platform level based on the cross-section threshold range and level type.

[0074] This step can provide business-meaning location results for subsequent outputs.

[0075] Specifically, the sign of the normal offset y* determines whether the defect is located on the left or right side; based on the interval relationship between the lateral reference distance xh of the defect point and the cross-sectional threshold sequence {Wk(s,c)}, and combined with the layer mapping table layer.type, it is determined whether the defect belongs to a first-level slope, a first-level platform, a second-level slope, or other special levels.

[0076] At the same time, confidence levels can be calculated by combining matching residuals, GNSS quality, and threshold posterior variance.

[0077] Step S108: Output the standardized results.

[0078] Specifically, it outputs a standardized location representation, including "station number + side + slope / platform level + confidence level", for example, "K123+456, left secondary slope, confidence level 0.92". The results are also written to a data packet for subsequent verification, statistics, and integration with third-party systems, increasing traceability and system compatibility.

[0079] The above-mentioned method for locating highway slope defects using a mobile phone based on a curvilinear coordinate system is applicable to highway slope defect location methods that can operate on road sections with unknown alignments. This method uses a mobile inspection device as the data carrier, collecting multi-source data to form an initial trajectory point set, and then autonomously fitting and establishing a Frenet curvilinear coordinate system. Subsequently, it combines key cross-sectional feature points such as slope tops, platform boundaries, and ditch edges collected under the guidance of inspection personnel to learn cross-sectional threshold sequences online. Finally, it projects the defect observation points onto this curvilinear coordinate system, completing the output of station number, side identification, slope / platform level, and confidence level.

[0080] The aforementioned mobile inspection terminal of this invention is transformed into a "self-calibrating line parameter establishment device + cross-section model learning device + disease location and classification device". It achieves the following: First, constructing a reusable fitting path γ(s) under the condition of no centerline data; second, transforming the complex hierarchical and plateau problems of slope cross-sections into a cross-section threshold learning problem along the fitting path; third, simultaneously utilizing planar position, heading, inertial constraints, and the cross-section model during positioning, thereby outputting more stable and traceable results.

[0081] In addition, this invention also provides a mobile phone-based highway slope disease location system corresponding to the above method, comprising: a data acquisition module configured to acquire multi-source data using an inspection mobile phone to form an original trajectory point set; a trajectory preprocessing module configured to perform smooth trajectory preprocessing on the original trajectory point set to obtain a smoothed inspection trajectory; a fitting path self-calibration and line parameter registration module configured to fit the smoothed inspection trajectory and perform line parameter registration, construct a mapping between the fitting path and the stationing system, and establish a Frenet curve coordinate system; and a feature point guided annotation module configured to... The system is configured to: guide the annotation of key cross-sectional feature points to obtain a set of cross-sectional feature points with feature type labels; project the feature points onto the Frenet curve coordinate system to learn and update the cross-sectional threshold sequence; perform sliding window matching on the disease observation points to obtain the optimal projection position of the disease points in the Frenet curve coordinate system; determine the lateral and level based on the projection position and the slope / platform level based on the cross-sectional threshold range and level type; and output the standardized results.

[0082] In addition, this embodiment of the invention also provides a device, which includes: a processor and a memory; the memory is used to store one or more program instructions; the processor is used to run one or more program instructions to execute the mobile terminal highway slope disease location method based on the curve coordinate system as described above.

[0083] In addition, embodiments of the present invention also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for locating highway slope defects on a mobile device based on a curvilinear coordinate system.

[0084] In addition, this embodiment of the invention also provides a computer program product, which includes computer program instructions. When the computer program instructions are executed by a processor, they implement the above-described method for locating highway slope defects on a mobile phone based on a curve coordinate system.

[0085] In this embodiment of the invention, the processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in this embodiment of the invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in this embodiment of the invention can be directly implemented by a hardware decoding processor, or implemented by a combination of hardware and software modules in the decoding processor. The software modules can be located in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The processor reads information from the storage medium and, in conjunction with its hardware, completes the steps of the above methods. The storage medium can be memory, for example, volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced Synchronous DRAM (ESDRAM), Synchlink DRAM (SLDRAM), and Direct Rambus RAM (DRRAM).The storage media described in the embodiments of this invention are intended to include, but are not limited to, these and any other suitable types of memory. Those skilled in the art will recognize that the functions described in the above examples can be implemented using a combination of hardware and software. When applied software, the corresponding functions can be stored in a computer-readable medium or transmitted as one or more instructions or code on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein communication media include any medium that facilitates the transmission of computer programs from one place to another. Storage media can be any available medium accessible to general-purpose or special-purpose computers. Although the invention has been described in detail above with general description and specific embodiments, modifications or improvements can be made to it, which will be apparent to those skilled in the art. Therefore, such modifications or improvements made without departing from the spirit of the invention are all within the scope of protection claimed by the invention.

[0086] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any simple modifications, equivalent changes, or alterations made by those skilled in the art using the disclosed technical content shall fall within the protection scope of the present invention.

Claims

1. A mobile-based method for locating highway slope defects based on a curvilinear coordinate system, characterized in that... include: The inspection mobile app collects multi-source data to form a set of original trajectory points. The original trajectory point set is subjected to smooth trajectory preprocessing to obtain the smoothed inspection trajectory. The smoothed inspection trajectory is fitted and line parameter registration is performed to construct the mapping between the fitted path and the stationing system, and a Frenet curve coordinate system is established. Perform guided annotation of key cross-sectional feature points to obtain a set of cross-sectional feature points with feature type labels; The feature points are projected onto the Frenet curve coordinate system to learn and update the cross-section threshold sequence; Sliding window matching is performed on the disease observation points to obtain the optimal projection position of the disease points in the Frenet curve coordinate system; Determine the side based on the projection location, and determine the slope / platform level based on the cross-section threshold range and level type; Output standardized results.

2. The method for locating highway slope defects on a mobile phone based on a curvilinear coordinate system according to claim 1, characterized in that, The inspection mobile terminal performs multi-source data collection to form the original trajectory point set, including: the inspection mobile terminal moves slowly along the road section to be inspected, and collects GNSS coordinates, IMU three-axis acceleration and angular velocity, magnetic orientation, captured images and timestamp information with a unified clock to form the original trajectory point set; And / or, the step of performing smoothing trajectory preprocessing on the original trajectory point set to obtain the smoothed inspection trajectory includes: performing time synchronization, outlier removal and filtering smoothing on the original trajectory point set to obtain the smoothed inspection trajectory; And / or, the process of fitting the smoothed inspection trajectory and performing line parameter registration to construct the mapping between the fitted path and the station number includes: dividing the smoothed inspection trajectory into curvature segments or quality segments, performing weighted spline fitting within each segment to obtain the fitted path; and using two or more known station number control points, known landmark points, or on-site manual confirmation points to establish a mapping relationship between the fitted path and the station number system; when the number of control points is small, piecewise linear registration is used; when the number of control points is large, least squares fitting or spline mapping is used.

3. The method for locating highway slope defects on a mobile phone based on a curvilinear coordinate system according to claim 1, characterized in that, The process of guiding the annotation of key cross-sectional feature points to obtain a set of cross-sectional feature points with feature type labels includes: The inspection mobile app prompts and guides inspection personnel to collect key cross-sectional feature points at preset pile sections or cross-sectional locations with significant curvature changes. These key cross-sectional feature points include slope tops, platform boundaries, slope toes, ditch edges, and retaining wall top edges. A candidate point list is generated based on image edges, local pose changes, local elevation changes, and existing cross-sectional templates. Candidate points are displayed in the form of "feature type + confidence score," and are manually confirmed or corrected with a single click to obtain a set of cross-sectional feature points with feature type labels. And / or, the step of projecting feature points onto the Frenet curve coordinate system to learn and update the cross-section threshold sequence includes: projecting the set of cross-section feature points onto the Frenet curve coordinate system, calculating the lateral reference distance of each feature point relative to the reference baseline, and maintaining the threshold sequence according to the road side and station position; the threshold sequence is used to describe the slope / platform boundary positions that appear sequentially from the road reference boundary outwards; The steps of guiding the annotation of key cross-section feature points and projecting the feature points onto the Frenet curve coordinate system to learn and update the cross-section threshold sequence are performed during the first modeling, cross-section model update, or when accuracy needs to be improved; if a stable cross-section threshold library has been established for a certain road segment, the historical model is directly called in the retest scenario.

4. The method for locating highway slope defects on a mobile phone based on a curvilinear coordinate system according to claim 1, characterized in that, The process of performing sliding window matching on the disease observation points to obtain the optimal projection position of the disease points in the Frenet curve coordinate system includes: when a disease observation event is received, constructing candidate projection intervals based on the current GNSS position, IMU integrated mileage, magnetic azimuth and adjacent observations, and using a sliding window-based hidden Markov model to obtain the optimal matching position, thereby obtaining the arc length position and normal offset of the disease observation point on the fitted path in the Frenet curve coordinate system. The determination of side based on projection position and the determination of slope / platform level based on cross-sectional threshold interval and level type include: determining whether the disease is located on the left or right side based on the sign of the normal offset; determining whether the disease belongs to a first-level slope, first-level platform, second-level slope or other special level based on the interval relationship between the lateral reference distance of the disease point and the cross-sectional threshold sequence, and in conjunction with the level mapping table.

5. The method for locating highway slope defects on a mobile phone based on a curvilinear coordinate system according to claim 4, characterized in that, The method of using a sliding window-based Hidden Markov Model (HMM) to find the optimal matching position includes: The observation probability is jointly given by the lateral distance residual, the heading difference, and the sensor mass. The optimal state transition probability is given by the consistency of the arc length increment between adjacent time moments, the curvature continuity, and the velocity constraint. Based on the preset sliding window length, the observation probability and the state transition probability are jointly solved to output the optimal matching position.

6. The method for locating highway slope defects on a mobile phone based on a curvilinear coordinate system according to claim 1, characterized in that, Before outputting the standardized results, the confidence level is calculated, which is calculated by combining the matching residual, GNSS quality, and threshold posterior variance. And / or, the output standardization results include the output of "station number + side identification + slope / platform level + confidence level".

7. A mobile-based highway slope defect location system based on a curvilinear coordinate system, characterized in that, include: The data acquisition module is configured to collect multi-source data using the inspection mobile app to form a set of original trajectory points. The trajectory preprocessing module is configured to perform smooth trajectory preprocessing on the original trajectory point set to obtain the smoothed inspection trajectory. The fitting path self-calibration and line parameter registration module is configured to fit the smoothed inspection trajectory and perform line parameter registration, construct the mapping between the fitting path and the stationing system, and establish the Frenet curve coordinate system. The feature point guided annotation module is configured to perform guided annotation of key cross-sectional feature points, and obtain a set of cross-sectional feature points with feature type labels; The cross-section threshold learning and update module is configured to project feature points onto the Frenet curve coordinate system to learn and update the cross-section threshold sequence. The disease projection positioning module is configured to perform sliding window matching on the disease observation points to obtain the optimal projection position of the disease points in the Frenet curve coordinate system. The side identification and level determination module is configured to determine the side identification based on the projection position and the slope / platform level based on the cross-section threshold range and level type. The standardization output module is configured to output standardized results.

8. A computer device, characterized in that, The device includes: a processor and a memory; The memory is used to store one or more program instructions; The processor is configured to run one or more program instructions to execute the mobile terminal highway slope disease location method based on a curve coordinate system as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the mobile terminal highway slope disease location method based on a curve coordinate system as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, The computer program product includes computer program instructions that, when executed by a processor, implement the mobile terminal highway slope disease location method based on a curve coordinate system as described in any one of claims 1 to 6.

Images