Cigarette three-dimensional shape reconstruction method and system based on laser three-dimensional scanning and application

Abstract

The invention belongs to the technical field of cigarette production, and provides a cigarette three-dimensional shape reconstruction method and system based on laser three-dimensional scanning and application, and the method comprises the steps: collecting three-dimensional point cloud data of the whole surface of a cigarette based on a laser triangular reflection principle; preprocessing the three-dimensional point cloud data by adopting a multi-stage filtering algorithm strategy to obtain reference data; the method comprises the following steps of: dispersing reference data into a plurality of two-dimensional sections along the axial direction of a cigarette, synchronously extracting upper and lower semicircular feature points of the cigarette in different two-dimensional sections by adopting a multi-thread technology, performing offset vector calculation and mirror image translation transformation so as to reconstruct a plurality of continuous two-dimensional reference sections in a virtual space, and performing curved surface reconstruction on the continuous two-dimensional reference sections, obtaining a three-dimensional shape model of the cigarette; based on the three-dimensional shape model, the perimeter and the length of the cigarette are extracted through a mixed measurement model. The cigarette shape detection efficiency can be improved, and the detection accuracy and the data accuracy are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of cigarette production, in particular to a method and system for reconstructing three-dimensional morphology of a cigarette based on laser three-dimensional scanning, and application thereof. BACKGROUND

[0002] Traditional methods mainly rely on manual sampling inspection to judge the appearance defects (such as wrinkled cigarettes, cloth belt lines, and contact heads) of cigarettes. Such methods have inherent limitations such as low efficiency, incomplete coverage of detection devices, and difficulty in quantitative evaluation, especially in the accurate detection of three-dimensional morphology of cigarettes (such as surface depression, end face air head, and filter shrinkage).

[0003] In the process of cigarette production, the stability control of the circumference and length of cigarettes is the core link of process quality control in cigarette enterprises. Production practice shows that in the quality control process, the phenomenon of excessive deviation of cigarette draw resistance caused by the deviation of circumference or length frequently occurs. Relevant research results in the industry have also confirmed that the mean draw resistance, ventilation rate, and draw resistance standard deviation of cigarettes have significant correlation with physical size parameters such as the circumference and length of cigarettes. The existing manual sampling inspection method for the circumference and length of cigarettes has exposed many problems in actual production: not only time-consuming and labor-intensive, but also increasing production costs, and the detection results are easily disturbed by subjective experience, fatigue degree, and other human factors, making it difficult to effectively guarantee the accuracy and consistency of the detection results. More importantly, the manual method is difficult to accurately quantify the defect indicators, and cannot meet the requirements of data refinement management in modern industrial production.

[0004] At present, two-dimensional vision detection technology has been widely applied, but two-dimensional images lack depth information, are easily affected by lighting conditions, and may have detection blind spots for specific angles such as cigarette end faces, resulting in missed detection or misjudgment. Therefore, it is of great significance to develop non-contact, high-precision, and fully automatic three-dimensional morphology detection technology and devices for cigarettes to improve the level of product quality control.

[0005] It is worth noting that existing research has begun to explore the application of line structured light scanning, three-dimensional point cloud data analysis, and vision-point cloud fusion technology in cigarette defect detection. For example, some research has obtained point cloud data of the surface of a cigarette by line structured light scanning, and combined with clustering analysis method to identify surface defects, and it is reported that the detection accuracy can reach 98.25%. However, this technology is used to analyze the circumference and length parameters of the cigarette by measuring the surface morphology of the cigarette, and its precision is insufficient. SUMMARY

[0006] In view of the above-mentioned defects of the prior art, the present application provides a cigarette three-dimensional topography reconstruction method, system and application based on laser three-dimensional scanning, which can efficiently process the complex and subtle features of cigarettes, solve the key technical problems such as slow speed response in online reconstruction process, low processing efficiency of massive point cloud data, and significant noise introduced by reconstruction algorithm. At the same time, accurate three-dimensional morphology of cigarettes is provided, which provides strong support for ensuring batch consistency and testing efficiency of key parameters (circumference, length) of cigarettes, and provides research data and direction for the influence of other parameters of cigarettes on quality.

[0007] To achieve the above object and related objects, the present application adopts the following technical solutions:

[0008] The present application provides a cigarette three-dimensional topography reconstruction method based on laser three-dimensional scanning in the first aspect, comprising the following steps:

[0009] Step S100: based on the principle of laser triangular reflection, three-dimensional point cloud data of the entire surface of the cigarette is collected;

[0010] Step S200: a multi-level filtering algorithm strategy is used to preprocess the three-dimensional point cloud data to obtain reference data;

[0011] Step S300: the reference data is discretized into several two-dimensional sections along the axial direction of the cigarette, a multi-thread technology is used to synchronously extract the upper and lower half-circle feature points in different two-dimensional sections, offset vector calculation and mirror translation transformation are performed, several continuous two-dimensional reference sections are reconstructed in the virtual space, and surface reconstruction is performed to obtain a three-dimensional topography model of the cigarette;

[0012] Step S400: based on the three-dimensional topography model, a hybrid measurement model is used to extract the circumference and length of the cigarette, wherein during the length extraction process, an adaptive Simpson integral algorithm oriented to the geometric features of the cigarette is used for length measurement, the adaptive Simpson integral algorithm includes automatically encrypting sampling points at the end turning area of the three-dimensional space curve of the three-dimensional topography model, and using sparse sampling in the flat area.

[0013] Further, in step S200, the multi-level filtering algorithm strategy includes filtering algorithm, smoothing algorithm and interpolation algorithm.

[0014] Further, in step S300, discretizing the reference data into several two-dimensional sections along the axial direction of the cigarette includes: discretizing the reference data into several equidistant thin layers along the axial direction of the cigarette, using an adaptive layering strategy to obtain several two-dimensional sections with complete geometric structure and noise removal, and the adaptive layering strategy includes intelligently expanding the axial search interval when the effective point cloud density of any thin layer is insufficient, until continuous point clusters are captured, and discrete noise clusters with a number of points less than a threshold value are discarded.

[0015] Further, in step S300, the upper and lower semicircle feature points in different two-dimensional sections are extracted, offset vector calculation and mirror translation transformation are performed, and a plurality of continuous two-dimensional reference sections are reconstructed in the virtual space, including: extracting the upper semicircle peak points and the lower semicircle valley points in different two-dimensional sections, calculating the three-dimensional offset vector between the two, and reconstructing the upper semicircle data and the lower semicircle data into a complete virtual circumference in the virtual space through a composite space transformation including mirror and translation, and a plurality of continuous two-dimensional reference sections are reconstructed.

[0016] Further, in step S400, the hybrid measurement model includes least squares method and template matching method.

[0017] Further, in step S400, the circumference of the cigarette is extracted by using the hybrid measurement model, including: dividing the three-dimensional topographic model into a plurality of single slice layers, projecting the point cloud data in the single slice layer to a two-dimensional fitting plane to establish a virtual coordinate system, and defining a template; converting the template into a local circular model with a center and a radius based on the least squares method, constructing all single slice layers into a plurality of local circular models through the hybrid measurement model, and aggregating the radius parameters of all local circular models along the axial direction of the cigarette to form a global circumference distribution feature to extract the circumference.

[0018] Further, in step S400, the length of the cigarette is extracted by using the hybrid measurement model, including: dividing the three-dimensional topographic model into a plurality of single slice layers, projecting the point cloud data in the single slice layer to a two-dimensional fitting plane to establish a virtual coordinate system, and defining a template; converting the template into a cross-section layer with a center and a radius based on the least squares method, and constructing all single slice layers into a plurality of cross-section layers through the hybrid measurement model.

[0019] Connecting the center parameters of all cross-section layers in the axial sequence to form a discrete space point sequence, and fitting it into a three-dimensional space curve, and measuring the length by using an adaptive Simpson integral algorithm.

[0020] The second aspect of the present application provides a cigarette three-dimensional topographic reconstruction system based on laser three-dimensional scanning, including:

[0021] The data acquisition module is used for acquiring three-dimensional point cloud data of the entire surface of the cigarette based on the principle of laser triangular reflection;

[0022] The data processing module is used for pre-processing the three-dimensional point cloud data by using a multi-level filtering algorithm strategy to obtain reference data.

[0023] The three-dimensional appearance model reconstruction module is used for discretizing the reference data into a plurality of two-dimensional sections along the axial direction of the cigarette, synchronously extracting feature points of upper and lower half circles on the cigarette in different two-dimensional sections by using a multi-thread technology, performing offset vector calculation and mirror translation transformation, reconstructing a plurality of continuous two-dimensional reference sections in a virtual space, and performing surface reconstruction to obtain a three-dimensional appearance model of the cigarette.

[0024] The cigarette parameter extraction module is used for extracting the circumference and length of the cigarette based on the three-dimensional appearance model by using a hybrid measurement model.

[0025] The third aspect of the present application provides a computer readable storage medium having computer readable instructions stored thereon, when the computer readable instructions are executed by a processor of a computer, the computer executes the above-mentioned cigarette three-dimensional appearance reconstruction method based on laser three-dimensional scanning.

[0026] The fourth aspect of the present application provides a computer device, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, when the processor executes the computer program, the steps of the above-mentioned cigarette three-dimensional appearance reconstruction method based on laser three-dimensional scanning are implemented.

[0027] The beneficial technical effects of the present application are that:

[0028] The present application is based on the laser triangular reflection principle, uses active optical three-dimensional scanning technology to collect three-dimensional point cloud data, can ensure the accuracy and integrity of the original data, through multi-level filtering algorithm strategy for data preprocessing, can obtain high-quality reference data, so as to ensure the accuracy of the subsequent model, through layered processing, the complex three-dimensional problem can be simplified, through the double half circle space splicing technology, the problem of incomplete circumference of curved cigarette can be solved, based on the reconstructed complete two-dimensional reference section, the surface reconstruction can be carried out, the accurate three-dimensional appearance model can be obtained, the hybrid measurement model is adopted, the optimal algorithm is selected for different cigarette parameters, the layered modeling is adopted for the circumference, and the adaptive Simpson integral algorithm is adopted for the length, so that the measurement accuracy and efficiency can be ensured.

[0029] The present application can accurately reconstruct the three-dimensional geometric appearance of the cigarette, and quantitatively represent the key size parameters and surface morphology characteristics. Combined with advanced image processing algorithm and machine learning technology, intelligent detection and classification evaluation of cigarette quality parameters can be realized. This provides the possibility for realizing comprehensive and accurate detection of key parameters of the cigarette, and through intuitive digital representation means, complex data can also be converted into an easy-to-understand and analyze graphical interface.

[0030] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application, as claimed. BRIEF DESCRIPTION OF DRAWINGS

[0031] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and serve to explain the principles of the application. It is readily apparent to one skilled in the art that the following figures are merely illustrative of some embodiments of the application and that other figures can be obtained from these figures without resorting to inventive skill. In the drawings:

[0032] Figure 1 Flow chart of the method of the application for reconstructing the three-dimensional appearance of a cigarette based on laser three-dimensional scanning;

[0033] Figure 2 Laser triangulation principle of the application;

[0034] Figure 3 Cigarette three-dimensional point cloud simulation diagram of the application;

[0035] Figure 4 Point cloud schematic diagram after noise processing of the application;

[0036] Figure 5 Framework diagram of the system of the application for reconstructing the three-dimensional appearance of a cigarette based on laser three-dimensional scanning;

[0037] Figure 6 The structural schematic diagram of a computer system of a computer device suitable for the embodiments of the application is shown. DETAILED DESCRIPTION

[0038] Unless otherwise defined, all technical and / or scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the application pertains. It is to be understood that certain features that are, for clarity, described in the context of separate embodiments can also be provided in combination in a single embodiment. Conversely, various features that are, for brevity, described in the context of a single embodiment can also be provided separately or in any suitable combination. The use of numbering in the claims is intended to supplement, not supersede, a classification that is determined according to the patent classification art and may, in some embodiments, be used merely for numbering purposes. Certain features that are, for brevity, described in the context of several embodiments, can also be provided in combination in a single embodiment. To the extent that specific men tal steps are described in this detailed description, it is understood that such descriptions are merely for illustrative purposes and are not intended to limit the scope of the application. Embodiments of the present application are further defined by the following written description and claims.

[0039] Referring to Figure 1 , the flowchart of the method for reconstructing the three-dimensional appearance of a cigarette based on laser three-dimensional scanning of the present application is described in detail as follows:

[0040] Step S100, based on the principle of laser triangulation reflection, three-dimensional point cloud data of the entire surface of the cigarette is collected.

[0041] Specifically, the present application uses active optical three-dimensional scanning technology to collect power data on the surface of the cigarette. Based on the principle of laser triangulation reflection, a line laser or a multi-point scanning sensor obtains distance information through a moving scanning point or line and generates dense point clouds (i.e., a set of three-dimensional spatial coordinate points (X, Y, Z) representing the surface morphology of the cigarette) in combination with position data. Further, as shown in Figure 2 , the laser beam of the present application is perpendicular to the reference plane and is incident on the surface of the measured sample. The partially diffused light of the incident laser is imaged on the photosensitive surface of the line array CCD through the lens. When the depth of the surface of the measured sample changes, the imaging points on the photosensitive surface of the line array CCD will also change accordingly. The point cloud data difference of any point on the surface of the measured sample can be obtained by measuring the imaging position of the spot on the photosensitive surface of the line array CCD. Based on the morphological characteristics of the cigarette, the present application configures a high-precision line array laser scanning sensor to collect three-dimensional point clouds on the surface of the cigarette by transverse translation scanning combined with axial rotation of the cigarette. The complete point cloud data on the surface of the cigarette (the point cloud data simulation diagram of the surface of the cigarette is shown in Figure 3 ) is obtained by combining the automatic rotation mechanism of the cigarette. Image synthesis and calculation are performed through processing of the point cloud data.

[0042] Step S200, a multi-level filtering algorithm strategy is used to pre-process the three-dimensional point cloud data to obtain reference data.

[0043] Specifically, the pre-processing of the three-dimensional point cloud data on the entire surface of the cigarette is related to the accuracy of modeling, the stability and reliability of detection data. The original three-dimensional data of the cigarette contains noise, void, redundant information or inconsistency, etc. Therefore, the multi-level filtering algorithm strategy process of the present application mainly includes filtering algorithm, smoothing algorithm and interpolation algorithm. Through the point cloud noise filtering of multiple algorithms, the curved surface morphology of the cigarette is accurately analyzed. The present application uses the multi-level filtering algorithm strategy to filter and remove outliers, removes isolated noise points by radius filtering, uses the smoothing algorithm to retain key edge features while suppressing high-frequency noise and optimizing surface continuity, and uses the interpolation algorithm to intelligently fill the point cloud void area and restore the missing surface geometric information.

[0044] Specifically, the filtering algorithm is the basis of three-dimensional data preprocessing, which improves the signal-to-noise ratio by removing high-frequency noise and retaining low-frequency information; the smoothing algorithm reduces data mutations and promotes continuous uniformity, which is beneficial to subsequent analysis and modeling; the interpolation algorithm fills in data gaps to ensure completeness, and the methods are diverse, such as linear, polynomial, and radial basis function interpolation, and the selection is based on the characteristics of the gap and the distribution of the surrounding data. More specifically, the filtering algorithm is used to efficiently identify and remove outlier noise points and redundant data. Statistical filtering is based on the distance distribution characteristics of the local neighborhood, and its calculation model satisfies the following relationship 1:

[0045] (Relationship 1);

[0046] When the distance of the neighborhood points exceeds the threshold range , it is determined as an outlier.

[0047] The radius filtering adopts a spatial density constraint condition to remove isolated points, and the radius filtering calculation model satisfies the following relationship 2:

[0048] (Relationship 2).

[0049] The smoothing algorithm is used to suppress high-frequency noise and optimize surface continuity while retaining key edge features as much as possible. Gaussian filtering is a linear smoothing algorithm that suppresses high-frequency noise by weighted averaging of neighborhood point coordinates, and is suitable for retaining low-frequency overall structure. The calculation model of the smoothing algorithm satisfies the following relationship 3:

[0050] (Relationship 3).

[0051] Bilateral filtering is a nonlinear smoothing algorithm that combines spatial domain and feature domain (such as normal vector and curvature) weights to suppress noise while retaining key edge features, and is suitable for edge-sensitive point cloud processing tasks. Its calculation model satisfies the following relationship 4:

[0052] (Relationship 4).

[0053] The interpolation algorithm (such as surface reconstruction based on moving least squares or Poisson reconstruction) is used to intelligently fill in the point cloud gap area and restore the missing surface geometric information. The moving least squares method is a local weighted interpolation method that fits a low-order polynomial surface in the neighborhood of each point to achieve progressive filling of the gap area. Let the neighborhood point set be , and the coordinates be . The design matrix is constructed, where the jth row element is ϕk(Q j ) (k = 0, 1, …, m). The weighted residual sum of squares satisfies the following relationship 5:

[0054] (Relation 5).

[0055] The application preserves the details of the curved surface by local polynomial fitting, and is suitable for small-scale cavity filling.

[0056] In step S300, the reference data is discretized into a plurality of two-dimensional sections along the axial direction of the cigarette, and a multi-thread technology is used to synchronously extract the upper and lower semicircle feature points in different two-dimensional sections, to perform offset vector calculation and mirror translation transformation, to reconstruct a plurality of continuous two-dimensional reference sections in a virtual space, and to perform curved surface reconstruction on the continuous two-dimensional reference sections to obtain a three-dimensional topographic model of the cigarette.

[0057] Specifically, the point cloud data of the surface of the cigarette is huge, and directly performing curved surface reconstruction, filtering, interpolation and other operations on the reference data will result in a very large amount of calculation, which will cause slow processing speed and cannot meet the efficiency requirements of actual industrial detection. Therefore, the application adopts a layered processing and parallel computing mode to optimize the calculation efficiency. More specifically, the premise of the layered processing of the application is to use a double semicircle space splicing technology to solve the problem of curved cigarette measurement difficulty, and to convert the complex global three-dimensional modeling problem into a large number of simple and independently processable two-dimensional section fitting problems, and then effectively extract and splice the upper and lower semicircles.

[0058] Specifically, the application establishes a space coordinate system with the axial direction of the cigarette (Y axis) as the reference, and discretizes the reference data into a plurality of equidistant thin layers along the axial direction of the cigarette. In actual operation, in order to ensure the integrity of the geometric structure of the thin layers during the discretization (also referred to as cutting) process, the application adopts an adaptive layering strategy to obtain a plurality of two-dimensional sections with complete geometric structure and noise removal. The adaptive layering strategy includes intelligently expanding the axial search interval when the effective point cloud density of any thin layer is insufficient, until continuous point clusters are captured, and discarding discrete noise clusters with less than a threshold number of points. That is, first, calculate the three-dimensional Euclidean distance of adjacent points, set a distance threshold of 70 microns to construct a connected domain, and discard discrete noise clusters with less than 50 points. This processing fully utilizes the spatial continuity feature of the surface of the cigarette, effectively eliminates floating noise points while preserving the true geometric structure. The point cloud graph after noise processing is shown in FIG. 2. Figure 4 Then the application adopts a parallel computing architecture, that is, the upper and lower semicircle layered processing tasks are executed synchronously by multiple threads, fully utilizing the computing resources of the multi-core processor to reduce the overall processing time. Experimental data shows that compared with traditional serial processing, this parallel architecture improves the simultaneous measurement efficiency of double cigarettes.

[0059] Specifically, the application extracts the upper and lower semicircle feature points in different two-dimensional sections, calculates the offset vector, and performs mirror translation transformation to reconstruct a series of continuous two-dimensional reference sections in the virtual space.

[0060] Specifically, the application reconstructs a complete and accurate three-dimensional topographic model of the cigarette based on the smooth surface reconstruction of the series of continuous two-dimensional reference sections.

[0061] Step S400: Based on the three-dimensional topographic model, a hybrid measurement model is used to extract the circumference and length of the cigarette, wherein in the process of extracting the length of the cigarette, an adaptive Simpson integral algorithm oriented to the geometric features of the cigarette is used for length measurement, and the adaptive Simpson integral algorithm includes automatically densifying sampling points at the end turning area of the three-dimensional space curve of the three-dimensional topographic model and using sparse sampling in the flat area.

[0062] Specifically, the hybrid measurement model of the application includes the least square method and the template matching method. Based on the basic geometric parameter extraction of the least square method and combined with the complex feature recognition of the template matching method, a high-precision and high-robustness three-dimensional size measurement algorithm of the cigarette can be established.

[0063] Specifically, this application uses a hybrid measurement model to extract the perimeter of a cigarette, including: dividing the three-dimensional topography model into several single-slice layers; projecting the point cloud data in the single-slice layers onto a two-dimensional fitting plane to establish a virtual coordinate system; and customizing the template; converting the template into a local circle model with a center and radius based on the least squares method; constructing several local circle models from all single-slice layers using the hybrid measurement model; and aggregating the radius parameters of all local circle models along the cigarette axis to form a global circumferential distribution feature to extract the perimeter.

[0064] More specifically, in its implementation, this application projects point cloud data in three-dimensional space onto a two-dimensional fitting plane to establish a virtual local coordinate system. This step simplifies the complexity of subsequent calculations and is a crucial preprocessing step for accurate geometric calculations. Then, the template defined in this application is a mathematical model of an ideal circle, whose equation is expressed as: The center (h, k) and radius r are parameters to be determined. The matching criterion is the least squares criterion, and the goal of matching is to find an ideal circle that minimizes the sum of the squares of the distances from all data points to the circumference of this circle.

[0065] More specifically, this application constructs a design matrix. With observation vector Using the least squares method Solve for parameters Solve for the center and radius of the circle using the least squares method to obtain the circle parameters: The optimal fitted circle, i.e., the local circle model, is determined. Then, this application repeats the above operation for each single slice layer on the tobacco branch axis to obtain a series of local circles with radii r1, r2, ... r... n The radius of each local area can be converted into its local perimeter. The distribution of all these local perimeters along the cigarette's axis constitutes the global circumferential distribution characteristic. By analyzing this circumferential distribution characteristic, such as taking the average, maximum, and minimum values, the perimeter can be extracted, enabling accurate measurement and quality assessment of the cigarette's circumference.

[0066] More specifically, this application strictly adheres to the principles of rigid body kinematics, achieving precise alignment of the three-dimensional point set through a homogeneous transformation matrix. Particularly noteworthy is the introduction of a local coordinate system calibration mechanism: a cylindrical coordinate reference system is established with the cigarette's main axis as the reference axis, ensuring that the reconstructed circumference strictly satisfies spatial closure characteristics. This approach completely solves the circumferential breakage problem encountered by traditional methods when measuring bent cigarettes, achieving virtual repair of incomplete contours.

[0067] Specifically, the length of the cigarette is extracted by using a hybrid measurement model, including: dividing the three-dimensional topographic model into a plurality of single slice layers, projecting the point cloud data in the single slice layer into a two-dimensional fitting plane to establish a virtual coordinate system, and performing template self-definition; converting the template into a cross-sectional layer with a center and a radius based on the least square method, and constructing all single slice layers into a plurality of cross-sectional layers by using the hybrid measurement model; connecting the center parameters of all cross-sectional layers in the axial sequence to form a discrete space point sequence, and fitting the discrete space point sequence into a three-dimensional space curve, and measuring the length by using an adaptive Simpson integral algorithm.

[0068] Specifically, the measurement of the length of the cigarette is based on the depth analysis of the three-dimensional projection characteristics, and the essence is to reconstruct the spatial topological form of the cigarette entity from the discrete point cloud data. The measurement process starts from the original point cloud collected by the laser scanner. The system first establishes a reference coordinate system along the axial direction (Y axis) of the cigarette, calculates the transverse cross-sectional characteristics at each discrete axial position, and forms a continuous profile curve with the height value changing with the axial position. From the theoretical level, the essence of the cigarette length calculation is to solve the effective straight length of the three-dimensional space curve. Its mathematical model can be described as a parametric curve The arc length integral on the interval [y1, y2] is shown in equation 6.

[0069] (Equation 6).

[0070] Specifically, in the actual system, the numerical solution is obtained by axial discrete sampling, and the approximation error is directly related to the sampling density. Therefore, the adaptive Simpson integral algorithm for the geometric characteristics of the cigarette is proposed. Simpson's rule is an important method in the field of numerical integration. Its core idea is to approximate the integral function by using a quadratic polynomial, so as to obtain a higher precision integral approximation value than the trapezoidal method. For a function f(x) defined on the interval [a, b], the basic integral formula satisfies equation 7:

[0071] (Equation 7).

[0072] Specifically, the application automatically increases the sampling points in the key feature area (end turning area) to capture the curve form with higher accuracy; in the flat area, sparse sampling is used to optimize the calculation efficiency while maintaining the accuracy. In actual engineering application, the application can effectively solve three key technical problems: the value conversion of the actual length and the projection length of the curved cigarette, the length compensation of the local deformation area, and the boundary recognition conflict when multiple cigarettes are in contact.

[0073] Please refer to Figure 5 The framework diagram of the three-dimensional topographic reconstruction system 500 of the cigarette based on laser three-dimensional scanning of the application is shown in FIG. 1.

[0074] The data acquisition module 510 is configured to acquire three-dimensional point cloud data of the whole surface of the cigarette based on a laser triangulation reflection principle.

[0075] The data processing module 520 is configured to pre-process the three-dimensional point cloud data by using a multi-level filtering algorithm strategy to obtain reference data.

[0076] The three-dimensional topographic model reconstruction module 530 is configured to discretize the reference data into a plurality of two-dimensional sections along the axial direction of the cigarette, synchronously extract feature points of the upper and lower semicircles of the cigarette in different two-dimensional sections by using a multi-thread technology, perform offset vector calculation and mirror translation transformation, reconstruct a plurality of continuous two-dimensional reference sections in a virtual space, and perform surface reconstruction to obtain a three-dimensional topographic model of the cigarette.

[0077] The cigarette parameter extraction module 540 is configured to extract the circumference and length of the cigarette based on the three-dimensional topographic model by using a hybrid measurement model. In the process of extracting the length of the cigarette, an adaptive Simpson integral algorithm oriented to the geometric features of the cigarette is used for length measurement. The adaptive Simpson integral algorithm includes automatically encrypting sampling points at the end transition zone of the three-dimensional space curve of the three-dimensional topographic model and using sparse sampling in the flat area.

[0078] It should be noted that the laser three-dimensional scanning based cigarette three-dimensional topographic reconstruction system provided in the above embodiment and the laser three-dimensional scanning based cigarette three-dimensional topographic reconstruction method provided in the above embodiment belong to the same concept, and the specific manner in which each module and unit performs operations has been described in detail in the method embodiments, which will not be described here. In actual application, the above functions can be completed by different functional modules according to needs, that is, the internal structure of the system is divided into different functional modules to complete all or part of the functions described above, and this is not limited herein.

[0079] Embodiments of the present application also provide a computer device, comprising: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the computer device implements the laser three-dimensional scanning based cigarette three-dimensional topographic reconstruction method provided in each of the above embodiments.

[0080] Figure 6 The structure schematic diagram of the computer system of the computer device suitable for the embodiments of the present application is shown. It should be noted that, Figure 6 The computer system 600 of the electronic device shown is only an example and should not limit the functions and use range of the embodiments of the present application.

[0081] As Figure 6As shown, the computer system 600 includes a central processing unit (CPU) 601 which can perform various appropriate actions and processes in accordance with programs stored in a read only memory (ROM) 602 or loaded into a random access memory (RAM) 603 from a storage section 608, such as the methods described in the above embodiments. Various programs and data required for the operation of the system are also stored in the RAM 603. The CPU 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604. The following are connected to the I / O interface 605: an input section 606 including a keyboard, a mouse, etc.; an output section 607 including a display such as a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to the I / O interface 605 as necessary. A removable recording medium 611 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is attached to the drive 610 as necessary, so that a computer program read therefrom is installed into the storage section 608 as necessary.

[0082] In particular, according to embodiments of the present application, the processes described above with reference to the flowcharts can be implemented as a computer tool program. For example, embodiments of the present application include a computer program product including a computer program carried on a computer-readable medium, the computer program containing a computer program for executing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network by the communication section 609, and / or installed from the removable recording medium 611. When the computer program is executed by the central processing unit (CPU) 601, various functions defined in the system of the present application are performed.

[0083] It should be noted that the computer readable medium in the embodiments of the present application can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may, for example, be an electronic, magnetic, optical, electromagnetic, infrared or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, a flash memory, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination of the above. In this application, the computer readable signal medium can include a data signal propagated in a baseband or as a carrier wave in a propagated data signal, in which the computer readable computer program is carried. Such a propagated data signal can take on various forms, including but not limited to an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium can also be any computer readable medium other than the computer readable storage medium, which can send, propagate or transmit the program for use by or in connection with an instruction execution system, apparatus or device. The computer program contained on the computer readable medium can be transmitted by any suitable medium, including but not limited to wireless, wired, or the like, or any suitable combination of the above.

[0084] The flowcharts and block diagrams in the drawings illustrate the possible implementation architectures, functions and operations of the systems, methods and computer program products according to various embodiments of the present application. In the flowcharts or block diagrams, each block can represent a module, a program segment or a part of code containing one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in different order from that shown in the drawings. For example, two blocks represented in succession can actually be executed substantially in parallel, and sometimes in reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and the combination of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0085] The units described in the embodiments of the present application can be implemented in the form of tools, or can be implemented in the form of hardware, and the described units can also be arranged in a processor. In some cases, the names of these units do not constitute a limitation on the units themselves.

[0086] Another aspect of the present application also provides a computer readable storage medium, which stores a computer program. The computer program is executed by a processor of a computer, so that the computer executes the method for reconstructing a three-dimensional appearance of a cigarette based on a laser three-dimensional scanning as described above. The computer readable storage medium can be included in the computer device described in the above embodiments, or can exist separately without being assembled into the computer device.

[0087] Another aspect of the present application also provides a computer program product or a computer program, which includes computer instructions stored in a computer readable storage medium. A processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method for reconstructing a three-dimensional appearance of a cigarette based on a laser three-dimensional scanning provided in the above embodiments.

[0088] The above embodiments only exemplarily illustrate the principles and effects of the present application, and are not used to limit the present application. Any person skilled in the art can modify or change the above embodiments without departing from the spirit and scope of the present application. Therefore, all equivalent modifications or changes completed by those skilled in the art without departing from the spirit and technical thought disclosed by the present application should be covered by the claims of the present application.

Claims

1. A method for reconstructing the three-dimensional appearance of a cigarette based on laser three-dimensional scanning, characterized in that, The method comprises the following steps: Step S100, based on the laser triangulation reflection principle, collecting three-dimensional point cloud data of the entire surface of the cigarette; Step S200, using a multi-level filtering algorithm strategy to preprocess the three-dimensional point cloud data to obtain reference data; Step S300, discretizing the reference data into a plurality of two-dimensional cross sections along the axial direction of the cigarette, using multi-thread technology to synchronously extract the upper and lower semicircle feature points of the cigarette in different two-dimensional cross sections, performing offset vector calculation and mirror translation transformation, reconstructing a plurality of continuous two-dimensional reference cross sections in virtual space, and performing surface reconstruction to obtain a three-dimensional topographic model of the cigarette; Step S400, based on the three-dimensional topographic model, using a hybrid measurement model to extract the circumference and length of the cigarette, wherein during the extraction of the length of the cigarette, an adaptive Simpson integral algorithm oriented to the geometric features of the cigarette is used for length measurement, and the adaptive Simpson integral algorithm comprises automatically encrypting sampling points at the end turning area of the three-dimensional space curve of the three-dimensional topographic model and using sparse sampling in flat areas.

2. The cigarette three-dimensional topography reconstruction method according to claim 1, characterized in that, In the step S200, the multi-level filtering algorithm strategy comprises a filtering algorithm, a smoothing algorithm, and an interpolation algorithm.

3. The cigarette three-dimensional topography reconstruction method according to claim 2, characterized in that, In the step S300, discretizing the reference data into a plurality of two-dimensional cross sections along the axial direction of the cigarette comprises discretizing the reference data into a plurality of equidistant thin layers along the axial direction of the cigarette, using an adaptive layering strategy to obtain a plurality of two-dimensional cross sections with complete geometric structure and noise removal, and the adaptive layering strategy comprises intelligently expanding the axial search interval when the effective point cloud density of any thin layer is insufficient, until continuous point clusters are captured, and discarding discrete noise clusters with a number of points less than a threshold value.

4. The cigarette three-dimensional topography reconstruction method according to claim 3, characterized in that, In the step S300, extracting the upper and lower semicircle feature points of the cigarette in different two-dimensional cross sections, performing offset vector calculation and mirror translation transformation, and reconstructing a plurality of continuous two-dimensional reference cross sections in virtual space comprises extracting the upper semicircle peak points and the lower semicircle valley points of the cigarette in different two-dimensional cross sections, calculating the three-dimensional offset vector between the two, and through a composite space transformation comprising mirroring and translation, reconstructing the upper semicircle data and the lower semicircle data into a complete virtual circumference in virtual space, and reconstructing a plurality of continuous two-dimensional reference cross sections.

5. The cigarette three-dimensional topography reconstruction method according to claim 4, characterized in that, In the step S400, the hybrid measurement model comprises a least squares method and a template matching method.

6. The cigarette three-dimensional topography reconstruction method according to claim 5, characterized in that, In the step S400, using a hybrid measurement model to extract the circumference of the cigarette comprises dividing the three-dimensional topographic model into a plurality of single slice layers, projecting the point cloud data in the single slice layers into a two-dimensional fitting plane to establish a virtual coordinate system, and defining a template; converting the template into a local circular model with a center and a radius based on the least squares method, constructing all the single slice layers into a plurality of local circular models through the hybrid measurement model, and aggregating the radius parameters of all the local circular models along the axial direction of the cigarette to form a global circumference distribution feature, so as to extract the circumference.

7. The cigarette three-dimensional topography reconstruction method according to claim 5, characterized in that, The step S400 includes: dividing the three-dimensional topography model into a plurality of single slice layers, projecting point cloud data in the single slice layers into a two-dimensional fitting plane to establish a virtual coordinate system, and performing template self-definition; converting the template into a cross-section layer with a center and a radius based on the least square method; and constructing all the single slice layers into a plurality of cross-section layers through the hybrid measurement model. The center parameters of all the cross-section layers are connected in axial sequence to form a discrete space point sequence, which is fitted into a three-dimensional space curve, and the length is measured through an adaptive Simpson integral algorithm.

8. A system for reconstructing the three-dimensional appearance of a cigarette based on a laser three-dimensional scan, characterized in that it comprises: The method comprises the following steps: The data acquisition module is configured to collect three-dimensional point cloud data of the entire surface of a cigarette based on a laser triangulation reflection principle. The data processing module is configured to pre-process the three-dimensional point cloud data through a multi-level filtering algorithm strategy to obtain reference data. The three-dimensional topography model reconstruction module is configured to discretize the reference data into a plurality of two-dimensional cross-sections along the axial direction of the cigarette, extract feature points of upper and lower semicircles of the cigarette in different two-dimensional cross-sections synchronously through a multi-thread technology, perform offset vector calculation and mirror translation transformation, reconstruct a plurality of continuous two-dimensional reference cross-sections in a virtual space, and perform surface reconstruction to obtain a three-dimensional topography model of the cigarette. The cigarette parameter extraction module is configured to extract the circumference and length of the cigarette based on the three-dimensional topography model through a hybrid measurement model.

9. A computer-readable storage medium, characterized in that, The computer readable instructions are stored on the computer readable medium and are executed by the processor of the computer to cause the computer to perform the method of any one of claims 1 to 7.

10. A computer device comprising: The memory, the processor, and the computer program stored on the memory and executable on the processor are characterized in that the processor executes the computer program to implement the steps of the method of any one of claims 1 to 7.

Images