Intelligent plane footprint and 3D laser automatic scanning system

Abstract

This invention relates to an intelligent planar footprint and 3D laser automatic scanning system, belonging to the field of trace investigation and intelligent scanning technology. The system includes: a scanning synthesis unit for controlling a mobile scanning device to simultaneously acquire local footprint images on the ground, automatically stitching and fusing the images to generate a full-area scene image; an intelligent comparison unit for establishing a spatial distribution coordinate system for footprints, completing intelligent footprint recognition, coordinate matching, grouping and classification, and prioritization of investigation, generating a list of key investigation targets; a 3D modeling unit for driving the device to complete full-angle laser scanning of key targets and constructing a 3D digital model; and a report generation unit for integrating multi-source investigation data and automatically generating a standardized investigation report and a visualized investigation atlas. This system achieves fully automated and non-destructive operation of footprint investigation at criminal scenes, significantly improving footprint collection efficiency and the standardization of evidence processing.

Description

Technical Field

[0001] This invention belongs to the field of trace investigation and intelligent scanning technology, specifically relating to an intelligent planar footprint and 3D laser automatic scanning system. Background Technology

[0002] Footprints are invaluable physical evidence in criminal scene investigations, directly reflecting the perpetrator's walking characteristics, physical characteristics, and crime scene path. They are the core basis for case investigation, suspect screening, and scene reconstruction. The completeness of footprint collection, the accuracy of analysis, and the authenticity of modeling directly affect the validity of physical evidence and the efficiency of investigation. Traditional footprint investigation is mainly based on manual operation and adopts two main implementation modes, both of which have unavoidable technical defects.

[0003] In the traditional manual investigation model, which involves taking photos on-site and processing them in the laboratory, investigators need to manually adjust the lighting angle and take multiple photos of the footprints. This only yields two-dimensional footprint information and is easily affected by ambient lighting, ground roughness, and obstructions from debris, resulting in blurry footprint images and loss of details. After taking the photos, the images need to be taken back to the laboratory and scanned and compared using specialized equipment. The process is cumbersome and data processing is slow, making it impossible to obtain footprint analysis results in real time on-site, which seriously delays the opportunity for case investigation.

[0004] Another type of manual fixed-point scanning mode requires investigators to manually move a special footprint scanner and place the entire device over a single footprint to conduct a static scan. The operation process is complex, and the original footprints and surrounding related evidence are easily trampled and damaged during the equipment relocation process, which violates the basic principle of non-destructive investigation of criminal scenes. This mode only supports fixed-point collection of single footprints and cannot achieve synchronous scanning of long-distance, continuous footprints. For large-area investigation scenes, the equipment position needs to be repeatedly adjusted, which is time-consuming, has limited coverage, and has extremely low overall investigation efficiency.

[0005] While existing footprint survey equipment and intelligent systems attempt to optimize workflows, significant technical shortcomings remain. Most devices only possess the single function of planar footprint scanning, lacking integrated 3D laser scanning capabilities. This prevents the creation of 3D models and data reproduction for key footprints and critical objects at the scene. Planar data acquisition and 3D modeling data are fragmented, failing to form a complete system of on-site evidence data. Furthermore, existing systems cannot achieve simultaneous footprint scanning during movement, relying on manual visual observation for location, resulting in poor positioning accuracy. Additionally, locally acquired footprint images cannot be automatically stitched and merged into a comprehensive scene image.

[0006] In addition, the existing system lacks the ability to intelligently analyze footprints, and cannot automatically compare features, group and classify the collected footprints, or determine the exploration priority. It still requires manual screening of key exploration targets, resulting in a low level of intelligence. At the same time, the system does not support the automatic integration of exploration data and the automatic generation of exploration reports. It requires manual summarization of various data and manual compilation of reports, which further increases the workload of field operations. Summary of the Invention

[0007] To address the aforementioned problems in the existing technology, this invention provides an intelligent planar footprint and 3D laser automatic scanning system. The objective of this invention can be achieved through the following technical solutions: include: The scanning synthesis unit controls the mobile scanning device to move within the survey area, acquiring local footprint images on the ground during the movement; the local footprint images are automatically stitched and fused to generate a full-area scene image; The intelligent comparison unit establishes a spatial distribution coordinate system of footprints in the exploration area based on the full-area scene image; extracts various feature information of all suspected footprint targets, matches corresponding spatial coordinates for each suspected footprint target; automatically compares and performs similarity analysis on the various feature information, automatically groups and classifies footprints according to the comparison analysis results, determines the exploration priority of each footprint target, and generates a list of key exploration targets. The 3D modeling unit, based on the list of key exploration targets, drives the mobile scanning device to the corresponding target position and adjusts its working posture; it performs a full-angle surround scan of the key exploration targets, acquires 3D spatial data, performs preprocessing, and constructs the corresponding 3D digital model; The report generation unit, through the aforementioned three-dimensional digital model, integrates and correlates the full-area scene images with the corresponding footprints, generating a full-scale survey information; simultaneously and automatically generating a field footprint survey report and a visualized survey map.

[0008] Specifically, the process of acquiring local footprint images on the ground is as follows: The mobile scanning device is controlled to move steadily along the planned route of the survey area, and uniform light is applied to the ground area covered by the scan. Ground footprint images are captured at preset fixed intervals, ensuring that adjacent images have a fixed overlapping area; Continuously acquire ground footprint images of the entire exploration area, and collect all ground footprint images to form a complete set of local footprint images.

[0009] Specifically, the process of generating the full-area scene image is as follows: Lens distortion correction is performed sequentially on all local footprint images, and grayscale normalization is performed on each image; Extract feature points from overlapping regions of adjacent images, and perform spatial registration of adjacent images based on the feature points; All registered images are seamlessly stitched together, and smooth blending is performed on the stitching edges. All stitched content is integrated according to the actual spatial layout of the survey area.

[0010] Specifically, the process of establishing the spatial distribution coordinate system of footprints in the exploration area is as follows: Select a fixed point within the exploration area as the origin of the coordinate system, and set the coordinate boundaries, coordinate axis extension directions, and spatial correspondence rules. Multiple fixed reference points were marked in the exploration area, and a spatial coordinate system was built based on the origin and reference points to cover all locations where footprints appeared.

[0011] Specifically, the process of matching the corresponding spatial coordinates for each suspected footprint target is as follows: Locate the specific position of each suspected footprint target in the entire scene image; Map the target location information to the footprint spatial distribution coordinate system, and calculate the coordinate parameters of the suspected footprint target in the coordinate system; Correct the spatial bias caused by the mapping, assign spatial coordinate identifiers to each suspected footprint target, and bind all targets to their corresponding coordinates.

[0012] Specifically, the process of determining the exploration priority of each footprint target is as follows: Extract the feature completeness, spatial distribution and correlation information of each footprint target, and comprehensively quantify the information according to the preset fixed rules for on-site investigation; Calculate the weight value of each footprint target based on the analysis results, and arrange all targets in descending order of weight value; Based on the ranking results, each footprint target is assigned an exploration priority level.

[0013] Specifically, the process of generating the list of key exploration targets is as follows: Based on the defined footprint exploration priority, select footprint targets that meet the preset key exploration criteria; Extract the core features, spatial coordinates, and priority hierarchy information of the footprint target, and organize all filtered information from high to low priority. The information was compiled and structured in a unified format to form a list of key exploration targets containing complete information.

[0014] Specifically, the process of driving the moving scanning device to the corresponding target position and adjusting its working posture is as follows: Read the spatial coordinate information of each target in the list of key exploration targets; Plan the travel route of the mobile scanning device and control the mobile scanning device to travel along the planned route to the vicinity of the target; Adjust the device's orientation, scanning angle, and relative distance to the target based on the target's location to ensure the scanning area covers the target's footprint range.

[0015] Specifically, the preprocessing process after acquiring the three-dimensional spatial data is as follows: Based on the acquired three-dimensional spatial data, invalid noise and redundant interference information in the data are identified and removed, while retaining the effective data that reflects the true spatial characteristics of the target. The effective data is formatted and standardized, and the data acquired from multiple perspectives are registered and aligned to eliminate spatial positional deviations between data, thus forming a regular three-dimensional spatial basic data.

[0016] Specifically, the three-dimensional digital model includes an input layer, a parsing layer, a construction layer, and an output layer, and the specific construction process is as follows: The input is preprocessed 3D spatial data, and the output is a 3D digital model. The input layer receives 3D spatial data and standardizes its format, removes invalid information, and establishes a standardized processing channel; the parsing layer extracts the spatial contour, surface undulation, and size ratio features of the target, and locks the spatial relationship of structural features; the construction layer builds a continuous and smooth three-dimensional surface to restore the complete spatial form of the target; the output layer outputs a 3D digital model containing the spatial form, surface details, size parameters, and structural relationships of the target.

[0017] Specifically, the process of associating and integrating the full-area scene images with the corresponding footprints and the complete survey information is as follows: Using a three-dimensional digital model as the core spatial anchor point, all exploration information of the entire area scene images and corresponding footprints was obtained; Establish spatial correspondences between 3D models, full-area scene images, and exploration information, and bind all data to the spatial distribution coordinate system of footprints.

[0018] Specifically, the process of automatically generating the on-site footprint survey report and the visualized survey map is as follows: The system retrieves a standard report template that conforms to the preset field investigation specifications, automatically fills the integrated investigation data into the corresponding positions in the template, and generates a complete and formatted investigation report text. Based on the spatial distribution coordinate system of footprints, the location, characteristics and priority of all footprints are marked on the scene image of the whole area, and the on-site footprint survey report and visualized survey map are output simultaneously.

[0019] The beneficial effects of this invention are as follows: (1) By setting up a structure with mobile synchronous scanning, full-domain scene stitching, intelligent footprint analysis and three-dimensional laser reconstruction, it can move autonomously and continuously collect ground footprint information within the exploration area, automatically complete the stitching and fusion of local footprint images to form a complete scene picture, accurately identify and locate suspected footprints and complete grouping, classification and exploration priority delineation, and implement full-angle three-dimensional scanning modeling for key targets, replacing the traditional manual fixed-point scanning and post-processing mode, avoiding disturbance and destruction of the original footprints on site, and greatly improving the integrity and continuity of footprint collection and the overall operational efficiency of on-site exploration; (2) By setting up a structure for the automatic output of multi-source exploration data association and fusion and standardized results, it is possible to deeply associate and integrate the scene images of the whole area, footprint spatial positioning information, exploration priority division results and three-dimensional digital models, and automatically generate professional exploration reports and intuitive visualization maps according to the field exploration specifications. This eliminates the tedious process of manual data collection, sorting and report preparation, ensures the standardization and uniformity of exploration results, and significantly improves the management efficiency and practical application value of field physical evidence data. Attached Figure Description

[0020] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to the accompanying drawings.

[0021] Figure 1 This is a system architecture diagram of an intelligent planar footprint and 3D laser automatic scanning system according to the present invention; Figure 2 This is a data flow diagram of an intelligent planar footprint and 3D laser automatic scanning system according to the present invention. Detailed Implementation

[0022] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided.

[0023] Please see Figure 1-2 An intelligent planar footprint and 3D laser automatic scanning system; include: The scanning synthesis unit controls the mobile scanning device to move within the survey area, acquiring local footprint images on the ground during the movement; the local footprint images are automatically stitched and fused to generate a full-area scene image; The intelligent comparison unit establishes a spatial distribution coordinate system of footprints in the exploration area based on the full-area scene image; extracts various feature information of all suspected footprint targets, matches corresponding spatial coordinates for each suspected footprint target; automatically compares and performs similarity analysis on the various feature information, automatically groups and classifies footprints according to the comparison analysis results, determines the exploration priority of each footprint target, and generates a list of key exploration targets. The 3D modeling unit, based on the list of key exploration targets, drives the mobile scanning device to the corresponding target position and adjusts its working posture; performs full-angle laser scanning on the key exploration targets, acquires 3D spatial data, performs preprocessing, and constructs the corresponding 3D digital model; The report generation unit, through the aforementioned three-dimensional digital model, integrates and correlates the full-area scene images with the corresponding footprints, generating a full-scale survey information; simultaneously and automatically generating a field footprint survey report and a visualized survey map.

[0024] In this embodiment, a local footprint image refers to a single segment of ground footprint image collected by the mobile scanning device during its movement in the survey area. Specifically, it includes the texture features of the ground footprint, the outline shape of the footprint, the ground environment information around the footprint, and clear image data after supplementary lighting and correction within a single image.

[0025] In this embodiment, the full-area scene image refers to the complete picture of the exploration area formed by registering, stitching and fusing multiple local footprint images. Specifically, it includes the overall appearance of the ground in the entire exploration area, the spatial distribution of all footprints, the relative distance and orientation relationship between each footprint, and the environmental features of the ground objects in the exploration area. It is the basic picture of the entire area for footprint recognition and positioning.

[0026] In this embodiment, suspected footprint targets refer to potential footprint objects initially screened out in the entire scene image through morphological features. Specifically, these include ground traces with human footprint outlines, surface textures, and trampling morphological features, as well as suspicious trace targets that have not yet completed feature verification. These are the preliminary objects for subsequent accurate identification. Footprint targets refer to valid footprints that have been confirmed to exist after feature comparison and verification. Specifically, these include trampling traces with complete ridge features, partial ridge features, or outline features, and real footprint carriers identified after excluding interference information such as debris shadows and ground unevenness. Key exploration targets refer to high-value footprint targets screened according to exploration priority. Specifically, these include footprint targets with complete and clear footprint features, close association with the exploration scene, key spatial location, and high physical evidence value. These are the core objects for the system to perform 3D scanning and modeling.

[0027] In this embodiment, the total exploration information refers to the complete data set generated by the present invention throughout the entire footprint exploration process. Specifically, it includes local footprint images, full-area scene images, footprint spatial coordinates, footprint feature information, classification and grouping results, exploration priorities, three-dimensional digital models, and spatial correlation information between various types of data. The standard report template for on-site exploration refers to a fixed-format document framework that conforms to industry on-site exploration requirements. Specifically, it includes a footprint basic information module, a spatial positioning information module, a feature and classification module, an exploration priority module, a three-dimensional model description module, and an exploration conclusion module, which are used to automatically fill in data and generate standardized reports.

[0028] Specifically, the process of acquiring local footprint images on the ground is as follows: The mobile scanning device is controlled to move steadily along the planned route of the survey area, and uniform light is applied to the ground area covered by the scan. Ground footprint images are captured at preset fixed intervals, ensuring that adjacent images have a fixed overlapping area; Continuously acquire ground footprint images of the entire exploration area, and collect all ground footprint images to form a complete set of local footprint images.

[0029] Specifically, the process of generating the full-area scene image is as follows: Lens distortion correction is performed sequentially on all local footprint images, and grayscale normalization is performed on each image; Extract feature points from overlapping regions of adjacent images, and perform spatial registration of adjacent images based on the feature points; All registered images are seamlessly stitched together, and smooth blending is performed on the stitching edges. All stitched content is integrated according to the actual spatial layout of the survey area.

[0030] Specifically, the process of establishing the spatial distribution coordinate system of footprints in the exploration area is as follows: Select a fixed point within the exploration area as the origin of the coordinate system, and set the coordinate boundaries, coordinate axis extension directions, and spatial correspondence rules. Multiple fixed reference points were marked in the exploration area, and a spatial coordinate system was built based on the origin and reference points to cover all locations where footprints appeared.

[0031] Specifically, the process of matching the corresponding spatial coordinates for each suspected footprint target is as follows: Locate the specific position of each suspected footprint target in the entire scene image; Map the target location information to the footprint spatial distribution coordinate system, and calculate the coordinate parameters of the suspected footprint target in the coordinate system; Correct the spatial bias caused by the mapping, assign spatial coordinate identifiers to each suspected footprint target, and bind all targets to their corresponding coordinates.

[0032] Specifically, the process of determining the exploration priority of each footprint target is as follows: Extract the feature completeness, spatial distribution and correlation information of each footprint target, and comprehensively quantify the information according to the preset fixed rules for on-site investigation; Calculate the weight value of each footprint target based on the analysis results, and arrange all targets in descending order of weight value; Based on the ranking results, each footprint target is assigned an exploration priority level.

[0033] Specifically, the process of generating the list of key exploration targets is as follows: Based on the defined footprint exploration priority, select footprint targets that meet the preset key exploration criteria; Extract the core features, spatial coordinates, and priority hierarchy information of the footprint target, and organize all filtered information from high to low priority. The information was compiled and structured in a unified format to form a list of key exploration targets containing complete information.

[0034] Specifically, the process of driving the moving scanning device to the corresponding target position and adjusting its working posture is as follows: Read the spatial coordinate information of each target in the list of key exploration targets; Plan the travel route of the mobile scanning device and control the mobile scanning device to travel along the planned route to the vicinity of the target; Adjust the device's orientation, scanning angle, and relative distance to the target based on the target's location to ensure the scanning area covers the target's footprint range.

[0035] Specifically, the preprocessing process after acquiring the three-dimensional spatial data is as follows: Based on the acquired three-dimensional spatial data, invalid noise and redundant interference information in the data are identified and removed, while retaining the effective data that reflects the true spatial characteristics of the target. The effective data is formatted and standardized, and the data acquired from multiple perspectives are registered and aligned to eliminate spatial positional deviations between data, thus forming a regular three-dimensional spatial basic data.

[0036] Specifically, the three-dimensional digital model includes an input layer, a parsing layer, a construction layer, and an output layer, and the specific construction process is as follows: The input is preprocessed 3D spatial data, and the output is a 3D digital model. The input layer receives 3D spatial data and standardizes its format, removes invalid information, and establishes a standardized processing channel; the parsing layer extracts the spatial contour, surface undulation, and size ratio features of the target, and locks the spatial relationship of structural features; the construction layer builds a continuous and smooth three-dimensional surface to restore the complete spatial form of the target; the output layer outputs a 3D digital model containing the spatial form, surface details, size parameters, and structural relationships of the target.

[0037] Specifically, the process of associating and integrating the full-area scene images with the corresponding footprints and the complete survey information is as follows: Using a three-dimensional digital model as the core spatial anchor point, all exploration information of the entire area scene images and corresponding footprints was obtained; Establish spatial correspondences between 3D models, full-area scene images, and exploration information, and bind all data to the spatial distribution coordinate system of footprints.

[0038] Specifically, the process of automatically generating the on-site footprint survey report and the visualized survey map is as follows: The system retrieves a standard report template that conforms to the preset field investigation specifications, automatically fills the integrated investigation data into the corresponding positions in the template, and generates a complete and formatted investigation report text. Based on the spatial distribution coordinate system of footprints, the location, characteristics and priority of all footprints are marked on the scene image of the whole area, and the on-site footprint survey report and visualized survey map are output simultaneously.

[0039] In this embodiment, the specific implementation details of the scanning synthesis unit are as follows: The planned route for the exploration area adopts a bow-shaped full-coverage traversal route. First, the boundary outline of the exploration area is identified. Using the right-angle vertex of the area as the starting point, the main travel path is planned along a direction parallel to the long side. The distance between adjacent main travel paths does not exceed the lateral width of a single scan by the mobile scanning device, ensuring a 10% to 15% overlap between adjacent paths. During the journey, obstacles are identified and avoided in real time before returning to the preset route, completely covering all ground areas where footprints may exist. When acquiring local footprint images, the travel speed of the mobile scanning device and the image acquisition frequency are matched to ensure that the lateral overlap area of ​​adjacent acquired images is stabilized at 30% to 50% of the width of a single image, providing sufficient matching benchmarks for subsequent registration and stitching. Simultaneously, a surrounding ring-shaped diffused lighting structure is used, illuminating the ground with polarized soft light at an angle of 30 to 45 degrees. The brightness of the supplementary light automatically adapts to the ambient light, stabilizing the grayscale value of the acquired images between 120 and 180. The image clearly shows the texture and subtle features of the footprints within the specified range. After acquisition, the Zhang Zhengyou calibration method, which was used for pre-shipment calibration using a checkerboard pattern, was employed to perform radial and tangential distortion correction on each local footprint image. The size error of the corrected image did not exceed 0.5%. Then, the ORB feature point extraction algorithm was used to extract feature points in the overlapping areas of adjacent images. After matching using a brute-force matching algorithm and removing mismatches using the RANSAC algorithm, the homography matrix was calculated to achieve accurate registration of adjacent images. The registration error did not exceed 1 pixel. Finally, based on the registration results, all images were superimposed onto the global canvas. A weighted average fusion algorithm was used to smooth the transition of overlapping areas and eliminate stitching gaps. Then, the brightness and contrast of the global image were uniformly normalized to generate a full-area scene image without misalignment, discontinuity, or uniform brightness.

[0040] In this embodiment, the specific implementation details of the intelligent comparison unit are as follows: The footprint spatial distribution coordinate system adopts a two-dimensional Cartesian coordinate system, which can also be extended to a three-dimensional Cartesian coordinate system as needed. During the construction, a fixed and undisturbed hard ground point within the exploration area is selected as the coordinate origin. The direction of the long side of the area is taken as the positive X-axis, and the direction of the short side perpendicular to the long side is taken as the positive Y-axis. A coordinate boundary not less than 0.5 meters beyond the actual edge of the exploration area is defined. Fixed reference points are marked at the four vertices of the area. The coordinate system is calibrated and constructed based on the spatial relationship between the origin and the reference points, so that the coordinate system completely covers all possible locations of footprints, and the coordinate positioning error does not exceed 2. Millimeters; After establishing the coordinate system, the smallest bounding rectangle of each suspected footprint target is located in the entire scene image using a target detection algorithm. The center point of the rectangle is taken as the position reference point. Through the mapping transformation matrix between the entire scene image and the coordinate system, the pixel coordinates of the reference point are converted into actual spatial coordinates. After correcting system deviations through on-site reference point verification, a unique spatial coordinate identifier is assigned to each suspected footprint target, establishing a one-to-one correspondence between the target pixel position and the actual position on site. Simultaneously, a footprint feature database containing standard templates of different shoe types, patterns, and sizes is established. Multi-dimensional feature information such as contour, texture, and size of each suspected footprint target is extracted and compared pairwise with the templates and different targets. The feature cosine similarity is calculated. When the similarity exceeds a preset 85%, the feature is considered complete. When the threshold is reached, two footprints are determined to be from the same source. Based on this, the footprints are first classified into three main categories according to the type of footprints: barefoot, sock prints, and shoe prints, and then further subdivided into shoe type subcategories to complete the classification. Footprints from the same source are then grouped into the same group, and the number, distribution location, and walking direction of each group of footprints are recorded to complete the orderly sorting of the footprints. A quantitative evaluation system is preset, which includes three indicators: feature completeness, spatial location correlation, and on-site correlation. The weights of the three indicators are 40%, 30%, and 30%, respectively. Each indicator is assigned a value from 0 to 100 according to the corresponding rules. The scores of the three indicators for each footprint are multiplied by the corresponding weights and then summed to obtain a comprehensive weight value from 0 to 100. Footprints are then divided into Level 1 (80 points and above), Level 2 (60-79 points), and Level 3 (60-79 points) according to the comprehensive weight from high to low. The exploration priority is divided into three levels. The first-level priority targets are selected as key exploration targets. Their numbers, types, feature information, spatial coordinates, weight values ​​and priority information are extracted and sorted in descending order of comprehensive weight. The results are then structured and summarized to generate a list of key exploration targets that can directly guide 3D scanning operations.

[0041] In this embodiment, the specific implementation of the 3D modeling unit is as follows: The spatial coordinate information of the key survey target is read. Starting from the current position of the device and ending at a distance of 0.8-1.2 meters from the target's perimeter, the shortest unobstructed route is planned. After the device reaches the endpoint, the horizontal orientation, lens pitch angle, and relative distance to the target are adjusted sequentially to ensure the target is completely within the center of the scanning field of view, thus completing the work posture adjustment. A fixed-center, equal-radius surround scanning method is adopted, using the target's center point as the center and a preset distance as the radius, driving the device to complete a full-angle surround scan within a 0-360 degree horizontal range and a -15 to 45 degree vertical range. A set of 3D point cloud data is collected every 0.5 degrees, completely covering the target's full-view spatial information. The collected 3D spatial data undergoes sequential integrity verification, statistical filtering and noise reduction, format standardization, and ICP verification. The algorithm employs a five-step preprocessing process, including multi-view registration and voxel downsampling simplification, ensuring that the amount of data after simplification does not exceed 40% of the original data, thus forming basic 3D spatial data that can be directly used for modeling. A four-layer progressive serial processing architecture is constructed, consisting of an input layer, a parsing layer, a construction layer, and an output layer. The input layer receives the preprocessed data and performs standardization processing. The parsing layer extracts and locks the spatial correlation of the target's core structural features. The construction layer builds a 3D curved surface structure based on the features and matches it with real textures. After accuracy verification and optimization, the output layer outputs a high-precision 3D digital model with a spatial accuracy of no less than 0.1 mm, containing full-dimensional information of the target.

[0042] In this embodiment, the report generation unit is implemented as follows: Using the 3D digital model of the key target as the core physical evidence anchor point and the spatial distribution coordinate system of footprints as the unified benchmark, it retrieves the full-area scene images and corresponding footprint survey information, binds the origin of the 3D model coordinates to the spatial coordinates of the corresponding footprints, and then associates the full-area footprint survey information with the corresponding spatial coordinates to establish a one-to-one correspondence between spatial coordinates, survey information, 3D model, and panoramic image position, completing the integration and unified management of all survey data; it retrieves a preset standardized report template, which includes six mandatory modules: basic on-site information, footprint survey overview, footprint details, description of the 3D model of the key target, survey conclusion, and attachments. Each module field corresponds one-to-one with the full-area survey data, and the data is automatically filled into the corresponding positions in the template to generate a field footprint survey report that conforms to legal regulations; using the full-area scene images as the base map, it overlays coordinate system grids, footprint position annotation boxes, priority indicators, and 3D model jump links to mark the core area of ​​the scene and the direction of the crime path, generating an interactive and traceable vector-type visual survey map, and finally, simultaneously and automatically outputs the survey report and the visual map.

[0043] In this embodiment, taking the footprint investigation at a closed scene of an indoor criminal case as an example, the specific implementation process is as follows: First, complete the pre-exploration preparations, delineate the closed area to be explored, place the mobile scanning device at the preset starting point of the exploration area, complete the device power-on self-test, lens calibration and verification, and ambient light parameter acquisition, and start the system's fully automated exploration program. The system controls a mobile scanning device to travel along a bow-shaped, full-coverage route across the survey area. After identifying the area's boundary contours, it plans a main path parallel to the long side, starting from the right-angle vertices of the area. The distance between adjacent paths does not exceed the lateral width of a single scan, ensuring an overlap of a% between adjacent paths. During travel, the device identifies and avoids obstacles in real time before returning to the preset route, completely covering all possible footprint areas on the ground. During travel, the device employs a surround-type ring-shaped diffused lighting structure, using polarized soft light at a c°-d° tilt angle to uniformly illuminate the scanned area. The lighting brightness automatically adapts to ambient light, stabilizing the grayscale values ​​of the acquired images within the ef range, clearly presenting the details of the footprint texture. The system synchronously matches the device's travel speed with the image acquisition frequency, acquiring ground footprint images at fixed intervals, ensuring a stable lateral overlap of b% between adjacent images. After acquisition, a complete set of local footprint images is formed. Subsequently, the system uses a pre-calibrated Zhang Zhengyou calibration method to perform distortion correction on all images, ensuring the dimensional error after correction does not exceed g%, and then passes the ORB (Orbital Boundary) calibration. The algorithm extracts feature points from adjacent images, and after matching and eliminating mismatches, completes accurate registration with a registration error of no more than h pixels. Finally, the weighted average fusion algorithm completes image stitching and brightness normalization to generate a full-area scene image without misalignment or tortuosity. A two-dimensional Cartesian spatial distribution coordinate system for footprints is constructed based on the full-area scene image. Fixed, undisturbed hard points within the area are selected as the origin. The long side of the area is defined as the positive X-axis, and the perpendicular short side as the positive Y-axis. A coordinate boundary extending at least *i* meters beyond the area's edge is delineated. Reference points are marked at the four vertices of the area. The calibrated coordinate system covers all footprint distribution areas, with a positioning error not exceeding *j* millimeters. The system uses a target detection algorithm to locate the reference points of suspected footprint targets. A mapping transformation matrix converts the pixel coordinates to actual spatial coordinates. After verifying and correcting deviations, a unique spatial coordinate identifier is assigned to each target, establishing a one-to-one correspondence between pixel positions and actual on-site positions. Simultaneously, a pre-set footprint feature database is retrieved to extract multi-dimensional features such as the target's contour, texture, and size. Cosine similarity is calculated pairwise between the target and the database template, using *k%* as the cosine similarity. To determine footprints from the same source based on similarity thresholds, footprints are first classified according to carrying type, and then those from the same source are grouped together to complete the orderly sorting of footprints. Subsequently, the system builds a quantitative evaluation system with three core indicators: feature completeness, spatial location correlation, and on-site correlation, with corresponding weights of 1%, m%, and n, respectively. Each indicator is assigned a value between 0 and 100 according to rules. The comprehensive weight value of each footprint is calculated, and the footprints are divided into three levels of exploration priority according to weight. The highest priority target is selected as the key exploration target, and the core information is extracted and summarized in descending order of weight to generate a list of key exploration targets that can guide subsequent operations. The system reads the spatial coordinates of key survey targets, using the current location of the device as the starting point and the target's perimeter (qr meters) as the ending point, and plans the shortest unobstructed route. After the device reaches the endpoint, it adjusts the device's orientation, lens pitch angle, and relative distance to the target to ensure the target is completely centered in the scanning field of view, completing the operational posture adjustment. Subsequently, the device employs a fixed-center, equal-radius surround scanning method, using the target's center point as the center, to complete a full-angle surround scan within the horizontal range of 0° to 5° and the vertical range of -t° to u°. During the scan, a set of 3D spatial data is acquired every v° interval, fully covering the target's full-view information. After acquisition, the system performs five preprocessing steps on the 3D spatial data: integrity verification, noise reduction, format unification, registration and alignment, and data simplification. The simplified data volume does not exceed w% of the original data, forming basic 3D spatial data suitable for modeling. Finally, the system uses a four-layer progressive serial architecture (input layer, parsing layer, construction layer, and output layer), with the preprocessed data as the total input, to sequentially complete data standardization, feature parsing, solid surface construction and texture mapping, and accuracy verification and optimization, outputting spatial accuracy no less than x. High-precision 3D digital model at millimeter level; Using the 3D digital model of key targets as the core anchor point and the spatial distribution coordinate system of footprints as the unified benchmark, the system retrieves all the scene images of the entire area and the full amount of investigation information of the corresponding footprints, establishing a one-to-one correspondence between spatial coordinates, investigation information, 3D models, and panoramic image positions, and completing the integration and unified management of all investigation data. Subsequently, the system retrieves a standard report template that conforms to the legal norms for criminal scene investigation. The template has six mandatory modules, and each field corresponds one-to-one with the investigation data. The data is automatically filled into the corresponding positions in the template to generate a scene footprint investigation report that meets legal requirements. Simultaneously, using the scene images of the entire area as the base map, coordinate system grids, footprint annotations, priority indicators, and 3D model jump links are overlaid to generate an interactive and traceable visual investigation map. Finally, the investigation report and the visual map are output simultaneously, completing this fully automated footprint investigation operation.

[0044] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. An intelligent planar footprint and 3D laser automatic scanning system, characterized in that, include: The scanning synthesis unit controls the mobile scanning device to move within the exploration area and acquire local footprint images on the ground during the movement. The local footprint images are automatically stitched and fused to generate a full-area scene image; The intelligent comparison unit establishes a spatial distribution coordinate system of footprints in the exploration area based on the full-area scene image; extracts various feature information of all suspected footprint targets, matches corresponding spatial coordinates for each suspected footprint target; automatically compares and performs similarity analysis on the various feature information, automatically groups and classifies footprints according to the comparison analysis results, determines the exploration priority of each footprint target, and generates a list of key exploration targets. The 3D modeling unit, based on the list of key exploration targets, drives the mobile scanning device to the corresponding target location and adjusts its working posture; A full-angle surround scan was performed on key exploration targets to obtain three-dimensional spatial data, which was then preprocessed and a corresponding three-dimensional digital model was constructed. The report generation unit, through the aforementioned three-dimensional digital model, integrates and correlates the full-area scene images with the corresponding footprints, generating a full-scale survey information; simultaneously and automatically generating a field footprint survey report and a visualized survey map.

2. The system according to claim 1, characterized in that, The specific process for obtaining local footprint images on the ground is as follows: The mobile scanning device is controlled to move steadily along the planned route of the survey area, and uniform light is applied to the ground area covered by the scan. Ground footprint images are captured at preset fixed intervals, ensuring that adjacent images have a fixed overlapping area; Continuously acquire ground footprint images of the entire exploration area, and collect all ground footprint images to form a complete set of local footprint images.

3. The system according to claim 1, characterized in that, The specific process for generating the full-area scene image is as follows: Lens distortion correction is performed sequentially on all local footprint images, and grayscale normalization is performed on each image; Extract feature points from overlapping regions of adjacent images, and perform spatial registration of adjacent images based on the feature points; All registered images are seamlessly stitched together, and smooth blending is performed on the stitching edges. All stitched content is integrated according to the actual spatial layout of the survey area.

4. The system according to claim 1, characterized in that, The specific process for establishing the spatial distribution coordinate system of footprints in the exploration area is as follows: Select a fixed point within the exploration area as the origin of the coordinate system, and set the coordinate boundaries, coordinate axis extension directions, and spatial correspondence rules. Multiple fixed reference points were marked in the exploration area, and a spatial coordinate system was built based on the origin and reference points to cover all locations where footprints appeared.

5. The system according to claim 1, characterized in that, The specific process of matching the corresponding spatial coordinates for each suspected footprint target is as follows: Locate the specific position of each suspected footprint target in the entire scene image; Map the target location information to the footprint spatial distribution coordinate system, and calculate the coordinate parameters of the suspected footprint target in the coordinate system; Correct the spatial bias caused by the mapping, assign spatial coordinate identifiers to each suspected footprint target, and bind all targets to their corresponding coordinates.

6. The system according to claim 1, characterized in that, The specific process for determining the exploration priority of each footprint target is as follows: Extract the feature completeness, spatial distribution and correlation information of each footprint target, and comprehensively quantify the information according to the preset fixed rules for on-site investigation; Calculate the weight value of each footprint target based on the analysis results, and arrange all targets in descending order of weight value; Based on the ranking results, each footprint target is assigned an exploration priority level.

7. The system according to claim 1, characterized in that, The specific process for generating the list of key exploration targets is as follows: Based on the defined footprint exploration priority, select footprint targets that meet the preset key exploration criteria; Extract the core features, spatial coordinates, and priority hierarchy information of the footprint target, and organize all filtered information from high to low priority. The information was compiled and structured in a unified format to form a list of key exploration targets containing complete information.

8. The system according to claim 1, characterized in that, The specific process of driving the mobile scanning device to the corresponding target position and adjusting its working posture is as follows: Read the spatial coordinate information of each target in the list of key exploration targets; Plan the travel route of the mobile scanning device and control the mobile scanning device to travel along the planned route to the vicinity of the target; Adjust the device's orientation, scanning angle, and relative distance to the target based on the target's location to ensure the scanning area covers the target's footprint range.

9. The system according to claim 1, characterized in that, The specific process of preprocessing the acquired three-dimensional spatial data is as follows: Based on the acquired three-dimensional spatial data, invalid noise and redundant interference information in the data are identified and removed, while retaining the effective data that reflects the true spatial characteristics of the target. The effective data is formatted and standardized, and the data acquired from multiple perspectives are registered and aligned to eliminate spatial positional deviations between data, thus forming a regular three-dimensional spatial basic data.

10. The system according to claim 1, characterized in that, The three-dimensional digital model includes an input layer, a parsing layer, a construction layer, and an output layer. The specific construction process is as follows: The input is preprocessed 3D spatial data, and the output is a 3D digital model. The input layer receives 3D spatial data and standardizes its format, removes invalid information, and establishes a standardized processing channel; the parsing layer extracts the spatial contour, surface undulation, and size ratio features of the target, and locks the spatial correlation of structural features; the construction layer builds a continuous and smooth 3D surface to restore the complete spatial form of the target. The output layer outputs a three-dimensional digital model containing the target's spatial morphology, surface details, dimensional parameters, and structural relationships.

11. The system according to claim 1, characterized in that, The specific process of associating and integrating the full-area scene images with the corresponding footprints and the complete survey information is as follows: Using a three-dimensional digital model as the core spatial anchor point, all exploration information of the entire area scene images and corresponding footprints was obtained; Establish spatial correspondences between 3D models, full-area scene images, and exploration information, and bind all data to the spatial distribution coordinate system of footprints.

12. The system according to claim 1, characterized in that, The specific process for automatically generating the on-site footprint survey report and visualized survey map is as follows: The system retrieves a standard report template that conforms to the preset field investigation specifications, automatically fills the integrated investigation data into the corresponding positions in the template, and generates a complete and formatted investigation report text. Based on the spatial distribution coordinate system of footprints, the location, characteristics and priority of all footprints are marked on the scene image of the whole area, and the on-site footprint survey report and visualized survey map are output simultaneously.

Images