Serial footprint collection splicing method and system based on two-dimensional laser radar

Abstract

The application provides a serial footprint collection splicing method and system based on a two-dimensional laser radar, and belongs to the technical field of footprint collection and image processing. The method comprises the following steps: establishing a global reference coordinate system and obtaining a ranging calibration compensation term; realizing automatic extraction of a cooperative target and spatial anchor point positioning through intensity filtering, connected domain clustering, morphological screening and intensity weighted centroid positioning; synchronously collecting IMU attitude data and footprint images and extracting geometric centers; taking the first footprint as a reference, constructing a rotation matrix according to an IMU attitude deviation to perform rotation correction, superimposing a physical translation deviation, and after unified conversion of physical coordinates and pixel coordinates, mapping to a global splicing canvas to form a serial footprint global image. The application realizes flexible arrangement of a footprint collection device along an arbitrary path, adapts to accurate positioning and splicing of a linear, curved and irregularly distributed footprint sequence, and improves field adaptability and geometric consistency.

Description

Technical Field

[0001] This invention relates to the field of footprint acquisition and image processing technology, and more particularly to a serial footprint acquisition and stitching method and system based on two-dimensional lidar. Background Technology

[0002] In the fields of criminal investigation and biometrics, the acquisition and analysis of serial footprints is of great significance for applications such as individual identification and gait reconstruction. Traditional footprint acquisition techniques typically rely on optical imaging or pressure sensing to record static or dynamic interaction data between the sole of the foot and the contact surface, forming single footprint images or continuous gait sequences. This data can provide a foundation for footprint morphological analysis and gait frequency calculation, but it depends on a fixed spatial relationship between the device and the footprint, such as vertical imaging or planar contact.

[0003] Existing serial footprint acquisition methods mainly fall into two categories: one is based on large, floor-standing equipment, which uses a mechanical structure to drive the acquisition module to move along a straight line and scan footprints one by one; the other utilizes a gait platform, requiring the subject to walk along a fixed path to generate a continuous footprint sequence. However, both methods require strict constraints on the movement trajectory of the equipment or the human body, making them unsuitable for scenarios common at crime scenes involving curved, circuitous, or irregular footprint distributions. Furthermore, the limitation of straight paths may lead to footprint image distortion or omissions, especially in complex terrain, where the adaptability and data integrity of existing technologies have significant deficiencies. Summary of the Invention

[0004] To address the technical problem that existing serial footprint acquisition methods require equipment or human movement along preset straight lines or fixed paths, which cannot adapt to the actual situation of footprints being curved, meandering, or irregularly distributed at crime scenes, this invention provides a serial footprint acquisition and stitching method and system based on two-dimensional lidar. This invention primarily utilizes a fixed 360° two-dimensional lidar to establish a global reference coordinate system, automatically extracting and intensity-weighted centroid positioning of the retroreflective cooperative target on the top of the mobile acquisition module. It simultaneously fuses IMU attitude data with the geometric center extraction results of the footprint image, and introduces ranging calibration compensation, IMU attitude deviation rotation correction, and a unified physical coordinate-pixel coordinate conversion mechanism. This allows the footprint acquisition equipment to be flexibly arranged along any path, automatically adapting to straight, curved, and irregularly distributed footprint sequences, avoiding the tedious manual target selection operation, achieving scale uniformity between radar physical coordinates and image pixel coordinates, and significantly improving the geometric consistency and accuracy of global reconstruction of serial footprints.

[0005] The technical means employed in this invention are as follows: A serial footprint acquisition and stitching method based on two-dimensional lidar includes: S1. Establish a global reference coordinate system with the scanning center of the two-dimensional lidar as the origin, and calibrate the ranging deviation of the two-dimensional lidar to obtain the on-site ranging calibration compensation item. S2. Based on the global reference coordinate system and the ranging calibration compensation term, the spatial anchor point of the mobile acquisition module is located to obtain the anchor point position of the acquisition device. IMU attitude data is acquired synchronously, and the footprint image is segmented to obtain the geometric center of the footprint and the spatial resolution of the footprint image. S3. Using the first footprint as the stitching reference, based on the acquisition device anchor point position, IMU attitude data, footprint geometric center and footprint image spatial resolution of each footprint obtained in step S2, perform coordinate transformation and registration on each footprint in sequence, and then overlay the registered footprint images onto the global stitching canvas to form a serial footprint global image.

[0006] Further, in step S1, establishing a global reference coordinate system with the scanning center of the two-dimensional lidar as the origin includes: The clockwise angle output by the radar Convert to standard angles in the mathematical Cartesian coordinate system (from (Rotation counterclockwise in the positive direction of the axis) The conversion relationship is:

[0007] After conversion, each scan point output by the radar is represented as a rectangular coordinate, as follows:

[0008] in, This is the distance measurement value.

[0009] Further, in step S1, the ranging deviation of the two-dimensional lidar is calibrated to obtain a ranging calibration compensation term, including: Before the formal data collection, two calibration rods were set up outside the footprint area on site. Highly reflective marker strips were fixed on the two calibration rods, and the installation height of the highly reflective marker strips was consistent with the height of the cooperative target on the mobile data collection module. The actual distance between the two calibration rods is measured using a laser rangefinder, steel ruler, or tape measure, and denoted as . ; Two highly reflective marker strips are scanned and automatically identified by a two-dimensional lidar to obtain the positioning coordinates of the two calibration rods in the radar coordinate system, which are denoted as follows:

[0010] in, This represents the horizontal and vertical coordinates of the first calibration rod in the radar coordinate system; This indicates the horizontal and vertical coordinates of the second calibration rod in the radar coordinate system; Based on the obtained positioning coordinates of the two calibration rods in the radar coordinate system, the radar-measured distance is calculated using the following formula:

[0011] in, This indicates the distance between the two calibration rods as measured by the radar; Based on the actual distance and the distance measured by the radar, the single ranging deviation is calculated using the following formula:

[0012] in, Indicates the first Single distance measurement deviation in a single measurement; Indicates the first The actual distance corresponding to each measurement; Indicates the first The radar distance corresponding to this measurement; On site The repeated measurement, for the th The average distance measurement deviation from each measurement is taken as the average distance measurement deviation and used as the on-site distance measurement calibration compensation item.

[0013] in, Indicates the on-site ranging calibration compensation item; This indicates the number of times the measurement was repeated on-site.

[0014] Further, in step S2, the step of spatial anchor point positioning of the mobile acquisition module based on the global reference coordinate system and the ranging calibration compensation term to obtain the anchor point position of the acquisition device includes: For each footprint, the polar coordinate data of each scan point is obtained by scanning the cooperative target on the 2D lidar scanning mobile acquisition module. ,in, For scanning angle, This is the distance value. This represents the reflection intensity value; In actual measurements, distance values and scanning angle Due to environmental and device noise, the distance value will be affected. and scanning angle They are modeled as follows:

[0015]

[0016] in, Indicates the on-site ranging calibration compensation item; This represents random noise in distance measurements; This represents random noise in angle measurements; and These are the standard deviations of distance and angle measurements, respectively. Preset intensity threshold , retain satisfaction The scan points form a set of highly reflective candidate points, as follows:

[0017] in, Based on the reflectivity of the target objects and the on-site environment, they are pre-calibrated; in subsequent clustering processing, only their angle and distance information are used. ; Cluster the candidate point set according to angular neighborhood, and group points with angular intervals smaller than the angular resolution. The consecutive points of multiples are grouped into a cluster. ,in, Indicates the first 1 candidate cluster; Calculate each candidate cluster angular width and radial span The formula is as follows:

[0018]

[0019] in, Candidate clusters Angular width, Candidate clusters The maximum scanning angle in the middle, Candidate clusters The minimum scanning angle in; Candidate clusters radial span, Candidate clusters The maximum distance value in, Candidate clusters The minimum distance value in; The angular width of the candidate clusters is matched with the theoretical angular width of the preset high-reflectivity sign strip to select the target clusters that best match the geometric features of the sign strip. Assume the actual length of the highly reflective marking strip is... Candidate clusters The corrected average distance to the center of the lidar is Then the theoretical angle between the two ends of the marking strip and the radar center is:

[0020] exist hour, , The distance is obtained by calibrating the mean distance between each scan point within the candidate cluster with distance measurement bias. ,in, Candidate clusters The number of scan points in the data. The obtained field ranging calibration compensation item; During morphological screening, candidate clusters that meet the following conditions are selected as target clusters:

[0021] in, This indicates the preset angle tolerance, which is set based on the lidar angular resolution, the installation error of the marker strip, and the on-site noise conditions. If there are multiple candidate clusters that meet the conditions in the same frame of scan data, the target is selected according to the current working stage: in the calibration stage, the target cluster corresponding to the high reflectivity marking strips of the two calibration rods is selected; in the footprint acquisition stage, the target cluster corresponding to the cooperative target on the top of the mobile acquisition module and meeting the constraints of distance range, radial span and spatial continuity is selected; the scan points in the selected target cluster are used as the cooperative target point set, and no manual selection is required. Set target cluster Include There are scan points, and the polar coordinates of each point are . The reflection intensity is The intensity-weighted centroids are as follows:

[0022]

[0023] in, Indicates the intensity-weighted average angle; Indicates the intensity-weighted average distance; Based on the on-site ranging calibration compensation item The intensity-weighted average distance is corrected to obtain the corrected center distance, as follows:

[0024] The corrected center distance The position coordinates are converted to radar Cartesian coordinates as follows:

[0025]

[0026] The coordinates of the target center are obtained and used as the anchor point location for the data acquisition equipment. .

[0027] Further, in step S2, the synchronous acquisition of IMU attitude data includes: During each data acquisition, the two-dimensional lidar scan data, the device yaw angle data output by the IMU, and the footprint image data are recorded simultaneously. Each acquisition generates a data set, as follows:

[0028] in, Indicates the first The device yaw angle output by the IMU during footprint acquisition; Indicates the first Image data of the footprints; Only when the timestamp difference between radar data, IMU attitude data, and footprint image data is less than a preset synchronization threshold is it considered to be the same sampling event, ensuring that multi-source data are strictly aligned in time.

[0029] Further, in step S2, the segmentation processing of the footprint image to obtain the geometric center of the footprint and the spatial resolution of the footprint image includes: Let the first During the second data acquisition, the position coordinates of the cooperative target or the anchor point of the acquisition equipment on the mobile acquisition module in the radar global coordinate system are: At the same moment, the yaw angle of the IMU output acquisition device relative to the preset reference direction is... This indicates the overall deflection angle of the data acquisition equipment caused by manual placement of the equipment; For the first Footprint images obtained from the second collection Image segmentation is performed to obtain the footprint mask. , These are the pixel coordinates in the image coordinate system. According to footprint mask Extract the footprint boundary contour and calculate the footprint geometric center using the following formula:

[0030] in, Indicates the first The pixel coordinates of the geometric center of each footprint; Let the set of footprint contour points be ,in, The first character on the footprint outline The coordinates of each pixel; Find the minimum bounding rectangle of the footprint contour point set, and obtain the four vertices of the minimum bounding rectangle. The geometric center of the footprint is then defined as the intersection of the diagonals of the smallest bounding rectangle, as follows:

[0031] The intersection of the diagonals of the smallest bounding rectangle can be expressed in coordinate form as follows:

[0032] Let the first The spatial resolutions of the footprint images are as follows: horizontal and vertical. and Then any pixel in the image Relative to the geometric center of the footprint The local physical coordinates are represented as:

[0033] in, Indicates the first Pixels in a footprint image Local physical coordinate vector relative to the geometric center of the footprint; This represents the actual physical length corresponding to each pixel in the horizontal direction of the footprint image; This represents the actual physical length corresponding to each pixel in the vertical direction of the footprint image; Let the yaw angle output by the IMU during the acquisition of the first footprint be... Then the first The IMU attitude deviation of the first footprint relative to the first footprint is:

[0034] in, This indicates the device yaw angle output by the IMU. It should be consistent with the actual relative deflection direction of the acquisition device, that is, when the first When the secondary acquisition device rotates clockwise or counterclockwise relative to the first footprint acquisition device, the footprint image is rotated and corrected in the corresponding direction during the stitching process.

[0035] Further, in step S3, using the first footprint as the stitching reference, and based on the anchor point position of the acquisition device, IMU attitude data, footprint geometric center, and footprint image spatial resolution of each footprint obtained in step S2, coordinate transformation and registration are performed on each footprint sequentially, including: Establish a global mosaic canvas coordinate system, and set the horizontal and vertical spatial resolutions of the global mosaic canvas as follows: and ,in, This represents the actual physical length of each pixel in the horizontal direction of the global spliced ​​canvas. This represents the actual physical length of each pixel in the vertical direction of the global splicing canvas; Let the pixel coordinates of the geometric center of the first footprint in the global mosaic canvas be:

[0036] in, This represents the pixel coordinates (horizontal and vertical) of the geometric center of the first footprint in the global mosaic canvas. Set to a specified pixel position in the center of the canvas or after the reserved boundary, depending on the canvas size; Let the anchor point of the data acquisition device corresponding to the first footprint be... ,in, This represents the horizontal and vertical coordinates of the moving acquisition module in the global reference coordinate system when the first footprint was acquired. Let the first The location of the data acquisition device anchor point corresponding to each footprint is ,in, Indicates the first The horizontal and vertical coordinates of the moving acquisition module in the global reference coordinate system during footprint acquisition; Each subsequent footprint was... Overlaying is performed using pixel reference centers; For the first Any pixel in a footprint image With the geometric center of the footprint As a local reference center, the pixel is first converted from image pixel coordinates to local physical coordinates; According to the IMU attitude deviation of the first footprint relative to the first footprint Construct a rotation matrix for the first... The local physical coordinates in the footprint image are rotated and corrected, and the rotation matrix is ​​as follows:

[0037] Based on the anchor point position of the acquisition device obtained by the two-dimensional lidar positioning, calculate the... The physical translation deviation of the second acquisition relative to the first acquisition is calculated using the following formula:

[0038] in, This represents the physical translation deviation in the radar's global coordinate system; Calculate the first The pixel translation deviation of each footprint relative to the first footprint in the global mosaic canvas is calculated using the following formula:

[0039] in, Indicates the first The pixel translation deviation vector of each footprint relative to the first footprint in the global mosaic canvas; Indicates the horizontal pixel translation deviation; Indicates the vertical pixel translation deviation; If the radar coordinate system The axis direction relative to the image pixel coordinate system If the axis direction is opposite, the sign of the vertical pixel offset is corrected according to the system calibration relationship; The local physical coordinates and physical translation deviation after IMU attitude correction By overlaying the images, we obtain the physical stitching coordinates of the pixel with reference to the first footprint:

[0040] in, Indicates the first The points in the footprint images are stitched together in physical space with the first footprint as a reference after IMU attitude correction and radar translation compensation.

[0041] Further, in step S3, the step of superimposing the registered footprint images onto the global stitching canvas to form a serial footprint global image includes: Based on the horizontal and vertical spatial resolution of the global mosaic canvas , Convert the physical stitching coordinates to pixel coordinates in the global stitching canvas:

[0042] Then we get the first Points in a footprint image Pixel coordinates mapped to the global mosaic canvas:

[0043] If there are cases where the radar coordinate system, footprint image coordinate system and global stitching canvas coordinate system have opposite axes or different origin definitions, the direction and sign of the corresponding coordinate axes shall be corrected according to the system calibration relationship. Based on the transformed global pixel coordinates of each pixel, each footprint image, footprint outline or footprint mask is resampled and superimposed onto a unified pixel canvas to form a serial global footprint image with consistent orientation, uniform scale and true spatial position relationship. For non-integer pixel coordinates, nearest neighbor interpolation, bilinear interpolation, or mask region filling methods are used to complete pixel assignment; During global image synthesis, the rotation angle of each footprint image is calculated using the IMU pose deviation relative to the first footprint. The centerline direction of the first footprint is only used to unify the display direction of the final stitching result and does not participate in the relative rotation calculation between each footprint. Thus, the spatial reconstruction of the dispersed footprint sequence from a single acquired image to a global serial footprint image is completed.

[0044] This invention also provides a serial footprint acquisition and stitching system based on two-dimensional lidar, implemented using the aforementioned serial footprint acquisition and stitching method based on two-dimensional lidar, comprising: a field equipment calibration module, a serial footprint acquisition module, and a footprint sequence stitching module, wherein: The field equipment calibration module is used to establish a global reference coordinate system with the scanning center of the two-dimensional lidar as the origin, and to calibrate the ranging deviation of the two-dimensional lidar to obtain the field ranging calibration compensation item. The sequence footprint acquisition module, based on the global reference coordinate system and ranging calibration compensation term, performs spatial anchor point positioning of the moving acquisition module to obtain the anchor point position of the acquisition device, synchronously acquires IMU attitude data, and performs segmentation processing on the footprint image to obtain the geometric center of the footprint and the spatial resolution of the footprint image; The footprint sequence stitching module uses the first footprint as the stitching reference benchmark. Based on the obtained anchor point position of the acquisition device, IMU attitude data, footprint geometric center and footprint image spatial resolution of each footprint, it sequentially performs coordinate transformation and registration on each footprint, and then superimposes the registered footprint images onto the global stitching canvas to form a serial footprint global image.

[0045] Compared with the prior art, the present invention has the following advantages: 1. The serial footprint acquisition and stitching method based on two-dimensional lidar provided by this invention achieves automatic identification of all scanning points corresponding to highly reflective markers from the original radar scanning data through a fully automatic cooperative target extraction mechanism that combines intensity threshold filtering, connected component clustering and morphological screening. This avoids the tedious operation of manually selecting targets in traditional methods and improves acquisition efficiency and consistency.

[0046] 2. The serial footprint acquisition and stitching method based on two-dimensional lidar provided by the present invention establishes a global reference coordinate system by setting a fixed 360° lidar and dynamically positioning the mobile acquisition module. This enables the footprint acquisition device to be flexibly arranged along any path. Whether it is a straight line, a curve or an irregularly distributed footprint sequence, it can be accurately located and stitched, which significantly expands the field adaptability of footprint acquisition technology.

[0047] 3. The serial footprint acquisition and stitching method based on two-dimensional lidar provided by this invention, through the intensity-weighted centroid positioning algorithm, realizes the conversion of automatically extracted high reflectivity cooperative target point sets into stable geometric center coordinates, effectively suppresses single-point measurement errors caused by factors such as laser scattering, minor equipment vibration, and uneven ground, and achieves stable positioning of the mobile acquisition module.

[0048] 4. The serial footprint acquisition and stitching method based on two-dimensional lidar provided by the present invention records the attitude deflection of the acquisition device during manual placement by using the IMU yaw angle, and uses the first footprint as a reference during the stitching process to perform rotation correction on the footprint image by using the IMU attitude difference between each acquisition, thereby avoiding the angle superposition error caused by mistakenly taking the direction angle of the footprint centerline as the device attitude angle.

[0049] 5. The serial footprint acquisition and stitching method based on two-dimensional lidar provided by this invention introduces a footprint geometric center and a unified pixel canvas coordinate transformation mechanism during the footprint stitching process. The physical space translation amount obtained by the two-dimensional lidar is converted into pixel translation amount according to the horizontal and vertical spatial resolution of the global stitching canvas. Each footprint image is uniformly mapped to the same pixel canvas, realizing the scale uniformity between the radar physical coordinates and the image pixel coordinates, avoiding stitching deviations caused by inconsistent units, and improving the geometric consistency and on-site restoration accuracy of the global reconstruction of serial footprints.

[0050] Based on the above reasons, this invention can be widely applied in fields such as on-site investigation, footprint examination and identification, gait analysis, and public safety. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention or 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 some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0053] 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 of the present invention. 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.

[0054] It should be noted that the terms "comprising" and "having" and any variations thereof in the specification, claims and accompanying drawings of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product or device.

[0055] like Figure 1 As shown, this invention provides a serial footprint acquisition and stitching method based on two-dimensional lidar, comprising: S1. Establish a global reference coordinate system with the scanning center of the two-dimensional lidar as the origin, and calibrate the ranging deviation of the two-dimensional lidar to obtain the on-site ranging calibration compensation item. S2. Based on the global reference coordinate system and the ranging calibration compensation term, the spatial anchor point of the mobile acquisition module is located to obtain the anchor point position of the acquisition device. IMU attitude data is acquired synchronously, and the footprint image is segmented to obtain the geometric center of the footprint and the spatial resolution of the footprint image. S3. Using the first footprint as the stitching reference, based on the acquisition device anchor point position, IMU attitude data, footprint geometric center and footprint image spatial resolution of each footprint obtained in step S2, perform coordinate transformation and registration on each footprint in sequence, and then overlay the registered footprint images onto the global stitching canvas to form a serial footprint global image.

[0056] In a specific implementation, as a preferred embodiment of the present invention, step S1, which involves establishing a global reference coordinate system with the scanning center of the two-dimensional lidar as the origin, includes: The clockwise angle output by the radar Convert to standard angles in the mathematical Cartesian coordinate system (from (Rotation counterclockwise in the positive direction of the axis) The conversion relationship is:

[0057] After conversion, each scan point output by the radar is represented as a rectangular coordinate, as follows:

[0058] in, This is the distance measurement value.

[0059] In a specific implementation, as a preferred embodiment of the present invention, step S1, which involves calibrating the ranging deviation of the two-dimensional lidar to obtain a ranging calibration compensation item, includes: Before the formal data collection, two calibration rods were set up outside the footprint area on site. Highly reflective marker strips were fixed on the two calibration rods, and the installation height of the highly reflective marker strips was consistent with the height of the cooperative target on the mobile data collection module. The actual distance between the two calibration rods is measured using a laser rangefinder, steel ruler, or tape measure, and denoted as . ; Two highly reflective marker strips are scanned and automatically identified by a two-dimensional lidar to obtain the positioning coordinates of the two calibration rods in the radar coordinate system, which are denoted as follows:

[0060] in, This represents the horizontal and vertical coordinates of the first calibration rod in the radar coordinate system; This indicates the horizontal and vertical coordinates of the second calibration rod in the radar coordinate system; Based on the obtained positioning coordinates of the two calibration rods in the radar coordinate system, the radar-measured distance is calculated using the following formula:

[0061] in, This indicates the distance between the two calibration rods as measured by the radar; Based on the actual distance and the distance measured by the radar, the single ranging deviation is calculated using the following formula:

[0062] in, Indicates the first Single distance measurement deviation in a single measurement; Indicates the first The actual distance corresponding to each measurement; Indicates the first The radar distance corresponding to this measurement; On site The repeated measurement, for the th The average distance measurement deviation from each measurement is taken as the average distance measurement deviation and used as the on-site distance measurement calibration compensation item.

[0063] in, Indicates the on-site ranging calibration compensation item; This indicates the number of times the measurement was repeated on-site.

[0064] In a specific implementation, as a preferred embodiment of the present invention, step S2, which involves spatial anchor point positioning of the mobile acquisition module based on the global reference coordinate system and the ranging calibration compensation term to obtain the anchor point position of the acquisition device, includes: For each footprint, the polar coordinate data of each scan point is obtained by scanning the cooperative target on the 2D lidar scanning mobile acquisition module. ,in, For scanning angle, This is the distance value. This represents the reflection intensity value; In actual measurements, distance values and scanning angle Due to environmental and device noise, the distance value will be affected. and scanning angle They are modeled as follows:

[0065]

[0066] in, Indicates the on-site ranging calibration compensation item; This represents random noise in distance measurements; This represents random noise in angle measurements; and These are the standard deviations of distance and angle measurements, respectively. Preset intensity threshold , retain satisfaction The scanned points form a set of highly reflective candidate points, as follows:

[0067] in, Based on the reflectivity of the target objects and the on-site environment, they are pre-calibrated; in subsequent clustering processing, only their angle and distance information are used. ; Cluster the candidate point set according to angular neighborhood, and group points with angular intervals smaller than the angular resolution. The consecutive points of multiples are grouped into a cluster. ,in, Indicates the first 1 candidate cluster; Calculate each candidate cluster angular width and radial span The formula is as follows:

[0068]

[0069] in, Candidate clusters Angular width, Candidate clusters The maximum scanning angle in the middle, Candidate clusters The minimum scanning angle in; Candidate clusters radial span, Candidate clusters The maximum distance value in, Candidate clusters The minimum distance value in; The angular width of the candidate clusters is matched with the theoretical angular width of the preset high-reflectivity sign strip to select the target clusters that best match the geometric features of the sign strip. Assume the actual length of the highly reflective marking strip is... Candidate clusters The corrected average distance to the center of the lidar is Then the theoretical angle between the two ends of the marking strip and the radar center is:

[0070] exist hour, , The distance is obtained by calibrating the mean distance between each scan point within the candidate cluster with distance measurement bias. ,in, Candidate clusters The number of scan points in the data. The obtained field ranging calibration compensation item; During morphological screening, candidate clusters that meet the following conditions are selected as target clusters:

[0071] in, This indicates the preset angle tolerance, which is set based on the lidar angular resolution, marker strip installation error, and on-site noise conditions; for example, it can be set to... , This refers to the angular resolution of a two-dimensional lidar. If there are multiple candidate clusters that meet the conditions in the same frame of scan data, the target is selected according to the current working stage: in the calibration stage, the target cluster corresponding to the highly reflective marking strips of the two calibration rods is selected; in the footprint acquisition stage, the target cluster corresponding to the cooperative target on the top of the mobile acquisition module and meeting the constraints of distance range, radial span and spatial continuity is selected; the scan points in the selected target cluster are used as the cooperative target point set, and no manual selection is required. Set target cluster Include There are scan points, and the polar coordinates of each point are . The reflection intensity is The intensity-weighted centroids are as follows:

[0072]

[0073] in, Indicates the intensity-weighted average angle; Indicates the intensity-weighted average distance; Based on the on-site ranging calibration compensation item The intensity-weighted average distance is corrected to obtain the corrected center distance, as follows:

[0074] The corrected center distance The position coordinates are converted to radar Cartesian coordinates as follows:

[0075]

[0076] The coordinates of the target center are obtained and used as the anchor point location for the data acquisition equipment. In this embodiment, during the calibration phase, this coordinate serves as the position of the calibration rod; during the footprint acquisition phase, this coordinate serves as the spatial anchor point of the moving acquisition module, used for subsequent calculations of the translational deviation of each acquisition position relative to the first footprint acquisition position. This spatial anchor point characterizes the positional change of the acquisition device, but is not directly equivalent to the geometric center position of the footprint.

[0077] In a specific implementation, as a preferred embodiment of the present invention, step S2, the synchronous acquisition of IMU attitude data, includes: During each data acquisition, the two-dimensional lidar scan data, the device yaw angle data output by the IMU, and the footprint image data are recorded simultaneously. Each acquisition generates a data set, as follows:

[0078] in, Indicates the first The device yaw angle output by the IMU during footprint acquisition; Indicates the first Image data of the footprints; Only when the timestamp difference between radar data, IMU attitude data, and footprint image data is less than a preset synchronization threshold is it considered to be the same sampling event, ensuring that multi-source data are strictly aligned in time.

[0079] In a specific implementation, as a preferred embodiment of the present invention, step S2, which involves segmenting the footprint image to obtain the footprint geometric center and the footprint image spatial resolution, includes: Let the first During the second data acquisition, the position coordinates of the cooperative target or the anchor point of the acquisition equipment on the mobile acquisition module in the radar global coordinate system are: At the same moment, the yaw angle of the IMU output acquisition device relative to the preset reference direction is... This indicates the overall deflection angle of the data acquisition equipment caused by manual placement of the equipment; For the Footprint images obtained from the second collection Image segmentation is performed to obtain the footprint mask. , These are the pixel coordinates in the image coordinate system. According to footprint mask Extract the footprint boundary contour and calculate the footprint geometric center using the following formula:

[0080] in, Indicates the first The pixel coordinates of the geometric center of each footprint; in this embodiment, the geometric center of the footprint The center point can be determined by the center of the circumscribed rectangle of the footprint, the center of the minimum circumscribed rectangle, or the geometric center of the footprint outline. It is preferable to use the center of the minimum circumscribed rectangle of the footprint outline as the geometric center of the footprint to reduce the influence of local defects, dirt noise, or local adhesion areas on the calculation of the center point.

[0081] Let the set of footprint contour points be ,in, The first character on the footprint outline The coordinates of each pixel; Find the minimum bounding rectangle of the footprint contour point set, and obtain the four vertices of the minimum bounding rectangle. The geometric center of the footprint is then defined as the intersection of the diagonals of the smallest bounding rectangle, as follows:

[0082] The intersection of the diagonals of the smallest bounding rectangle can be expressed in coordinate form as follows:

[0083] Let the first The spatial resolutions of the footprint images are as follows: horizontal and vertical. and Then any pixel in the image Relative to the geometric center of the footprint The local physical coordinates are represented as:

[0084] in, Indicates the first Pixels in a footprint image Local physical coordinate vector relative to the geometric center of the footprint; This represents the actual physical length corresponding to each pixel in the horizontal direction of the footprint image; This represents the actual physical length corresponding to each pixel in the vertical direction of the footprint image; Let the yaw angle output by the IMU during the acquisition of the first footprint be... Then the first The IMU attitude deviation of the first footprint relative to the first footprint is:

[0085] in, This indicates the device yaw angle output by the IMU. It should be consistent with the actual relative deflection direction of the acquisition device, that is, when the first When the secondary acquisition device rotates clockwise or counterclockwise relative to the first footprint acquisition device, the footprint image is rotated and corrected according to the corresponding direction during the stitching process. In this embodiment, the direction angle of the footprint centerline can be obtained based on the footprint boundary contour. This direction angle of the footprint centerline can be used to determine the display reference direction of the first footprint, and can also serve as an auxiliary parameter for subsequent left and right foot recognition, footprint morphological feature extraction, gait direction analysis, or stitching result verification.

[0086] In a specific implementation, as a preferred embodiment of the present invention, step S3, using the first footprint as a stitching reference, involves sequentially performing coordinate transformation and registration on each footprint based on the anchor point position of the acquisition device, IMU attitude data, footprint geometric center, and footprint image spatial resolution obtained in step S2. This includes: Establish a global mosaic canvas coordinate system, and set the horizontal and vertical spatial resolutions of the global mosaic canvas as follows: and ,in, This represents the actual physical length of each pixel in the horizontal direction of the global spliced ​​canvas. This represents the actual physical length of each pixel in the vertical direction of the global spliced ​​canvas; Let the pixel coordinates of the geometric center of the first footprint in the global mosaic canvas be:

[0087] in, This represents the pixel coordinates (horizontal and vertical) of the geometric center of the first footprint in the global mosaic canvas. Set to a specified pixel position in the center of the canvas or after the reserved boundary, depending on the canvas size; This invention selects the first footprint as a unified stitching reference benchmark, uses the geometric center of the first footprint as the image stitching reference center, and uses the centerline direction of the first footprint as the display direction benchmark for the final stitching result. Let the anchor point of the data acquisition device corresponding to the first footprint be... ,in, This represents the horizontal and vertical coordinates of the moving acquisition module in the global reference coordinate system when the first footprint was acquired. Let the first The location of the data acquisition device anchor point corresponding to each footprint is ,in, Indicates the first The horizontal and vertical coordinates of the moving acquisition module in the global reference coordinate system during footprint acquisition; Each subsequent footprint was... Overlaying is performed using pixel reference centers; For the Any pixel in a footprint image With the geometric center of the footprint As a local reference center, the pixel is first converted from image pixel coordinates to local physical coordinates; According to the IMU attitude deviation of the first footprint relative to the first footprint Construct a rotation matrix for the first... The local physical coordinates in the footprint image are rotated and corrected, and the rotation matrix is ​​as follows:

[0088] Based on the anchor point position of the acquisition device obtained by the two-dimensional lidar positioning, calculate the... The physical translation deviation of the second acquisition relative to the first acquisition is calculated using the following formula:

[0089] in, This represents the physical translation deviation in the radar's global coordinate system; Calculate the first The pixel translation deviation of each footprint relative to the first footprint in the global mosaic canvas is calculated using the following formula:

[0090] in, Indicates the first The pixel translation deviation vector of each footprint relative to the first footprint in the global mosaic canvas; Indicates the horizontal pixel translation deviation; Indicates the vertical pixel translation deviation; If the radar coordinate system The axis direction relative to the image pixel coordinate system If the axis direction is opposite, the sign of the vertical pixel offset is corrected according to the system calibration relationship; The local physical coordinates and physical translation deviation after IMU attitude correction By overlaying the images, we obtain the physical stitching coordinates of the pixel with reference to the first footprint:

[0091] in, Indicates the first The points in the footprint images are stitched together in physical space with the first footprint as a reference after IMU attitude correction and radar translation compensation.

[0092] In a specific implementation, as a preferred embodiment of the present invention, step S3, which involves superimposing the registered footprint images onto a global stitching canvas to form a serial footprint global image, includes: Based on the horizontal and vertical spatial resolution of the global mosaic canvas , Convert the physical stitching coordinates to pixel coordinates in the global stitching canvas:

[0093] Then we get the first Points in a footprint image Pixel coordinates mapped to the global mosaic canvas:

[0094] If there are cases where the radar coordinate system, footprint image coordinate system, and global stitching canvas coordinate system have opposite axes or different origin definitions, the direction and sign of the corresponding coordinate axes are corrected according to the system calibration relationship; in this embodiment, this step makes the first Each footprint image is first normalized to coordinates using its own geometric center as a local reference center, then rotated based on IMU attitude deviation, and then superimposed with the physical translation deviation measured by the two-dimensional lidar, and finally mapped onto a unified global stitching canvas.

[0095] Through the above transformation, the first The image of each footprint is first converted into physical coordinates by taking its own geometric center as the local reference center; then it is rotated and corrected according to the IMU attitude deviation relative to the first footprint; then the physical translation deviation measured by the two-dimensional lidar is superimposed; finally, it is converted into pixel coordinates according to the horizontal and vertical spatial resolution of the global stitching canvas and placed in the unified stitching canvas.

[0096] Based on the transformed global pixel coordinates of each pixel, each footprint image, footprint outline or footprint mask is resampled and superimposed onto a unified pixel canvas to form a serial global footprint image with consistent orientation, uniform scale and true spatial position relationship. For non-integer pixel coordinates, nearest neighbor interpolation, bilinear interpolation, or mask region filling methods are used to complete pixel assignment; During global image synthesis, the rotation angle of each footprint image is calculated using the IMU pose deviation relative to the first footprint. The centerline direction of the first footprint is only used to unify the display direction of the final stitching result and does not participate in the relative rotation calculation between each footprint. Thus, the spatial reconstruction of the dispersed footprint sequence from a single acquired image to a global serial footprint image is completed.

[0097] This invention also provides a serial footprint acquisition and stitching system based on two-dimensional lidar, implemented using the aforementioned serial footprint acquisition and stitching method based on two-dimensional lidar, comprising: a field equipment calibration module, a serial footprint acquisition module, and a footprint sequence stitching module, wherein: The field equipment calibration module is used to establish a global reference coordinate system with the scanning center of the two-dimensional lidar as the origin, and to calibrate the ranging deviation of the two-dimensional lidar to obtain the field ranging calibration compensation item. The sequence footprint acquisition module, based on the global reference coordinate system and ranging calibration compensation term, performs spatial anchor point positioning of the moving acquisition module to obtain the anchor point position of the acquisition device, synchronously acquires IMU attitude data, and performs segmentation processing on the footprint image to obtain the geometric center of the footprint and the spatial resolution of the footprint image; The footprint sequence stitching module uses the first footprint as the stitching reference benchmark. Based on the obtained anchor point position of the acquisition device, IMU attitude data, footprint geometric center and footprint image spatial resolution of each footprint, it sequentially performs coordinate transformation and registration on each footprint, and then superimposes the registered footprint images onto the global stitching canvas to form a serial footprint global image.

[0098] The embodiments of the present invention are described simply because they correspond to those in the embodiments above. For any similarities, please refer to the descriptions in the embodiments above, which will not be elaborated here.

[0099] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for serial footprint acquisition and stitching based on two-dimensional lidar, characterized in that, include: S1. Establish a global reference coordinate system with the scanning center of the two-dimensional lidar as the origin, and calibrate the ranging deviation of the two-dimensional lidar to obtain the on-site ranging calibration compensation item. S2. Based on the global reference coordinate system and the ranging calibration compensation term, the spatial anchor point of the mobile acquisition module is located to obtain the anchor point position of the acquisition device. IMU attitude data is acquired synchronously, and the footprint image is segmented to obtain the geometric center of the footprint and the spatial resolution of the footprint image. S3. Using the first footprint as the stitching reference, based on the acquisition device anchor point position, IMU attitude data, footprint geometric center and footprint image spatial resolution of each footprint obtained in step S2, perform coordinate transformation and registration on each footprint in sequence, and then overlay the registered footprint images onto the global stitching canvas to form a serial footprint global image.

2. The serial footprint acquisition and stitching method based on two-dimensional lidar according to claim 1, characterized in that, In step S1, establishing a global reference coordinate system with the scanning center of the two-dimensional lidar as the origin includes: The clockwise angle output by the radar Convert to standard angles in the mathematical Cartesian coordinate system The transformation relationship is as follows: After conversion, each scan point output by the radar is represented as a rectangular coordinate, as follows: in, This is the distance measurement value.

3. The serial footprint acquisition and stitching method based on two-dimensional lidar according to claim 1, characterized in that, In step S1, the ranging deviation of the two-dimensional lidar is calibrated to obtain a ranging calibration compensation item, including: Before the formal data collection, two calibration rods were set up outside the footprint area on site. Highly reflective marker strips were fixed on the two calibration rods, and the installation height of the highly reflective marker strips was consistent with the height of the cooperative target on the mobile data collection module. The actual distance between the two calibration rods is measured using a laser rangefinder, steel ruler, or tape measure, and denoted as . ; Two highly reflective marker strips are scanned and automatically identified by a two-dimensional lidar to obtain the positioning coordinates of the two calibration rods in the radar coordinate system, which are denoted as follows: in, This represents the horizontal and vertical coordinates of the first calibration rod in the radar coordinate system; This indicates the horizontal and vertical coordinates of the second calibration rod in the radar coordinate system; Based on the obtained positioning coordinates of the two calibration rods in the radar coordinate system, the radar-measured distance is calculated using the following formula: in, This indicates the distance between the two calibration rods as measured by the radar; Based on the actual distance and the distance measured by the radar, the single ranging deviation is calculated using the following formula: in, Indicates the first Single distance measurement deviation in a single measurement; Indicates the first The actual distance corresponding to each measurement; Indicates the first The radar distance corresponding to this measurement; On site The repeated measurement, for the th The average distance measurement deviation from each measurement is taken as the average distance measurement deviation and used as the on-site distance measurement calibration compensation item. in, Indicates the on-site ranging calibration compensation item; This indicates the number of times the measurement was repeated on-site.

4. The serial footprint acquisition and stitching method based on two-dimensional lidar according to claim 1, characterized in that, In step S2, the step of spatial anchor point positioning of the mobile acquisition module based on the global reference coordinate system and the ranging calibration compensation term to obtain the anchor point position of the acquisition device includes: For each footprint, the polar coordinate data of each scan point is obtained by scanning the cooperative target on the 2D lidar scanning mobile acquisition module. ,in, For scanning angle, This is the distance value. This represents the reflection intensity value; In actual measurements, distance values and scanning angle Due to environmental and device noise, the distance value will be affected. and scanning angle They are modeled as follows: in, Indicates the on-site ranging calibration compensation item; This represents random noise in distance measurements; This represents random noise in angle measurements; and These are the standard deviations of distance and angle measurements, respectively. Preset intensity threshold , retain satisfaction The scanned points form a set of highly reflective candidate points, as follows: in, Based on the reflectivity of the target objects and the on-site environment, they are pre-calibrated; in subsequent clustering processing, only their angle and distance information are used. ; Cluster the candidate point set according to angular neighborhood, and group points with angular intervals smaller than the angular resolution. The consecutive points of multiples are grouped into a cluster. ,in, Indicates the first 1 candidate cluster; Calculate each candidate cluster angular width and radial span The formula is as follows: in, Candidate clusters Angular width, Candidate clusters The maximum scanning angle in the middle, Candidate clusters The minimum scanning angle in; Candidate clusters radial span, Candidate clusters The maximum distance value in, Candidate clusters The minimum distance value in; The angular width of the candidate clusters is matched with the theoretical angular width of the preset high-reflectivity sign strip to select the target clusters that best match the geometric features of the sign strip. Let the actual length of the highly reflective marking strip be... Candidate clusters The corrected average distance to the center of the lidar is Then the theoretical angle between the two ends of the marking strip and the radar center is: exist hour, , The distance is obtained by calibrating the mean distance between each scan point within the candidate cluster with distance measurement bias. ,in, Candidate clusters The number of scan points in the data. The obtained field ranging calibration compensation item; During morphological screening, candidate clusters that meet the following conditions are selected as target clusters: in, This indicates the preset angle tolerance, which is set based on the lidar angular resolution, the installation error of the marker strip, and the on-site noise conditions. If there are multiple candidate clusters that meet the conditions in the same frame of scan data, the target is selected according to the current working stage: in the calibration stage, the target cluster corresponding to the highly reflective marking strips of the two calibration rods is selected; in the footprint acquisition stage, the target cluster corresponding to the cooperative target on the top of the mobile acquisition module and meeting the constraints of distance range, radial span and spatial continuity is selected; the scan points in the selected target cluster are used as the cooperative target point set, and no manual selection is required. Set target cluster Include There are scan points, and the polar coordinates of each point are . The reflection intensity is The intensity-weighted centroids are as follows: in, Indicates the intensity-weighted average angle; Indicates the intensity-weighted average distance; Based on the on-site ranging calibration compensation item The intensity-weighted average distance is corrected to obtain the corrected center distance, as follows: The corrected center distance The position coordinates are converted to radar Cartesian coordinates as follows: The coordinates of the target center are obtained and used as the anchor point location for the data acquisition equipment. .

5. The serial footprint acquisition and stitching method based on two-dimensional lidar according to claim 1, characterized in that, In step S2, the synchronous acquisition of IMU attitude data includes: During each data acquisition, the two-dimensional lidar scan data, the device yaw angle data output by the IMU, and the footprint image data are recorded simultaneously. Each acquisition generates a data set, as follows: in, Indicates the first The device yaw angle output by the IMU during footprint acquisition; Indicates the first Image data of the footprints; Only when the timestamp difference between radar data, IMU attitude data, and footprint image data is less than a preset synchronization threshold is it considered to be the same sampling event, ensuring that multi-source data are strictly aligned in time.

6. The serial footprint acquisition and stitching method based on two-dimensional lidar according to claim 1, characterized in that, In step S2, the segmentation of the footprint image to obtain the geometric center of the footprint and the spatial resolution of the footprint image includes: Let the first During the second data acquisition, the position coordinates of the cooperative target or the anchor point of the acquisition equipment on the mobile acquisition module in the radar global coordinate system are: At the same moment, the yaw angle of the IMU output acquisition device relative to the preset reference direction is... This indicates the overall deflection angle of the data acquisition equipment caused by manual placement of the equipment; For the Footprint images obtained from the second collection Image segmentation is performed to obtain the footprint mask. , These are the pixel coordinates in the image coordinate system. According to footprint mask Extract the footprint boundary contour and calculate the footprint geometric center using the following formula: in, Indicates the first The pixel coordinates of the geometric center of each footprint; Let the set of footprint contour points be ,in, The first character on the footprint outline The coordinates of each pixel; Find the minimum bounding rectangle of the footprint contour point set, and obtain the four vertices of the minimum bounding rectangle. The geometric center of the footprint is then defined as the intersection of the diagonals of the smallest bounding rectangle, as follows: The intersection of the diagonals of the smallest bounding rectangle can be expressed in coordinate form as follows: Let the first The spatial resolutions of the footprint images are as follows: horizontal and vertical. and Then any pixel in the image Relative to the geometric center of the footprint The local physical coordinates are represented as: in, Indicates the first Pixels in a footprint image Local physical coordinate vector relative to the geometric center of the footprint; This represents the actual physical length corresponding to each pixel in the horizontal direction of the footprint image; This represents the actual physical length corresponding to each pixel in the vertical direction of the footprint image; Let the yaw angle output by the IMU during the acquisition of the first footprint be... Then the first The IMU attitude deviation of the first footprint relative to the first footprint is: in, This indicates the device yaw angle output by the IMU. It should be consistent with the actual relative deflection direction of the acquisition device, that is, when the first When the secondary acquisition device rotates clockwise or counterclockwise relative to the first footprint acquisition device, the footprint image is rotated and corrected in the corresponding direction during the stitching process.

7. The serial footprint acquisition and stitching method based on two-dimensional lidar according to claim 1, characterized in that, In step S3, using the first footprint as the stitching reference, and based on the anchor point position of the acquisition device, IMU attitude data, footprint geometric center, and footprint image spatial resolution of each footprint obtained in step S2, coordinate transformation and registration are performed on each footprint sequentially, including: Establish a global mosaic canvas coordinate system, and set the horizontal and vertical spatial resolutions of the global mosaic canvas as follows: and ,in, This represents the actual physical length of each pixel in the horizontal direction of the global spliced ​​canvas. This represents the actual physical length of each pixel in the vertical direction of the global splicing canvas; Let the pixel coordinates of the geometric center of the first footprint in the global mosaic canvas be: in, This represents the pixel coordinates (horizontal and vertical) of the geometric center of the first footprint in the global mosaic canvas. Set to a specified pixel position in the center of the canvas or after the reserved boundary, depending on the canvas size; Let the anchor point of the data acquisition device corresponding to the first footprint be... ,in, This represents the horizontal and vertical coordinates of the moving acquisition module in the global reference coordinate system when the first footprint was acquired. Let the first The location of the data acquisition device anchor point corresponding to each footprint is ,in, Indicates the first The horizontal and vertical coordinates of the moving acquisition module in the global reference coordinate system during footprint acquisition; Each subsequent footprint was... Overlaying is performed using pixel reference centers; For the Any pixel in a footprint image With the geometric center of the footprint As a local reference center, the pixel is first converted from image pixel coordinates to local physical coordinates; According to the IMU attitude deviation of the first footprint relative to the first footprint Construct a rotation matrix for the first... The local physical coordinates in the footprint image are rotated and corrected, and the rotation matrix is ​​as follows: Based on the anchor point position of the acquisition device obtained from the two-dimensional lidar positioning, calculate the... The physical translation deviation of the second acquisition relative to the first acquisition is calculated using the following formula: in, This represents the physical translation deviation in the radar's global coordinate system; Calculate the first The pixel translation deviation of each footprint relative to the first footprint in the global mosaic canvas is calculated using the following formula: in, Indicates the first The pixel translation deviation vector of each footprint relative to the first footprint in the global mosaic canvas; Indicates the horizontal pixel translation deviation; Indicates the vertical pixel translation deviation; If the radar coordinate system The axis direction relative to the image pixel coordinate system If the axis direction is opposite, the sign of the vertical pixel offset is corrected according to the system calibration relationship; The local physical coordinates and physical translation deviation after IMU attitude correction By overlaying the images, we obtain the physical stitching coordinates of the pixel with reference to the first footprint: in, Indicates the first The points in the footprint images are stitched together in physical space with the first footprint as a reference after IMU attitude correction and radar translation compensation.

8. The serial footprint acquisition and stitching method based on two-dimensional lidar according to claim 1, characterized in that, In step S3, the step of superimposing the registered footprint images onto the global stitching canvas to form a serial footprint global image includes: Based on the horizontal and vertical spatial resolution of the global mosaic canvas , Convert the physical stitching coordinates to pixel coordinates in the global stitching canvas: Then we get the first Points in a footprint image Pixel coordinates mapped to the global mosaic canvas: If there are cases where the radar coordinate system, footprint image coordinate system and global stitching canvas coordinate system have opposite axes or different origin definitions, the direction and sign of the corresponding coordinate axes shall be corrected according to the system calibration relationship. Based on the transformed global pixel coordinates of each pixel, each footprint image, footprint outline or footprint mask is resampled and superimposed onto a unified pixel canvas to form a serial global footprint image with consistent orientation, uniform scale and true spatial position relationship. For non-integer pixel coordinates, nearest neighbor interpolation, bilinear interpolation, or mask region filling methods are used to complete pixel assignment; During global image synthesis, the rotation angle of each footprint image is calculated using the IMU pose deviation relative to the first footprint. The centerline direction of the first footprint is only used to unify the display direction of the final stitching result and does not participate in the relative rotation calculation between each footprint. Thus, the spatial reconstruction of the dispersed footprint sequence from a single acquired image to a global serial footprint image is completed.

9. A serial footprint acquisition and stitching system based on a two-dimensional lidar, implemented using the serial footprint acquisition and stitching method based on any one of claims 1-8, characterized in that, include: The module includes an on-site equipment calibration module, a sequence footprint acquisition module, and a footprint sequence stitching module, among which: The field equipment calibration module is used to establish a global reference coordinate system with the scanning center of the two-dimensional lidar as the origin, and to calibrate the ranging deviation of the two-dimensional lidar to obtain the field ranging calibration compensation item. The sequence footprint acquisition module, based on the global reference coordinate system and ranging calibration compensation term, performs spatial anchor point positioning of the moving acquisition module to obtain the anchor point position of the acquisition device, synchronously acquires IMU attitude data, and performs segmentation processing on the footprint image to obtain the geometric center of the footprint and the spatial resolution of the footprint image; The footprint sequence stitching module uses the first footprint as the stitching reference benchmark. Based on the obtained anchor point position of the acquisition device, IMU attitude data, footprint geometric center and footprint image spatial resolution of each footprint, it sequentially performs coordinate transformation and registration on each footprint, and then superimposes the registered footprint images onto the global stitching canvas to form a serial footprint global image.

Images