A QR calibration board calibration method and system for multi-scene cooperation

Abstract

The application belongs to the technical field of computer vision, and discloses a QR calibration board calibration method and system for multi-scene cooperation, which comprises the following steps: constructing a fusion calibration board, adopting a unified two-dimensional matrix structure, integrating multiple functional areas in the same calibration board, and cooperating with layout according to preset space rules; generating a positioning mark area and an adjacent formatted information area in the two-dimensional matrix structure; configuring the physical size of the positioning mark area with a fixed internal structure, and storing format constraint information through the formatted information area; configuring a visual corner detection pattern in the calibration board according to different application scenarios, and performing structural completion on the corner area affected by the two-dimensional matrix layout; writing preset task information into the data content area in the multi-functional area, and generating a code word based on a finite field coding rule before writing; and the QR calibration board has improved adaptability and calibration precision in multi-scene cooperative application.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer vision, more particularly, the present application relates to a QR calibration board calibration method and system for multi-scene cooperation. BACKGROUND

[0002] The existing QR calibration board calibration method and system for multi-scene cooperation mainly have the following problems:

[0003] With the wide application of robots, augmented reality and multi-camera systems in complex scenes, camera calibration and spatial positioning tasks have higher requirements for the functions of calibration boards. Existing calibration boards mainly include checkerboards, circular positioning patterns, and Aruco, ChAruco, and other coded calibration boards, which are mainly used for camera parameter solving and corner positioning. However, these traditional calibration boards have many limitations in multi-scene cooperation and information carrying:

[0004] Limited information carrying capacity: Existing checkerboards, circular positioning or Aruco calibration boards usually only provide corner positioning information and cannot carry a large amount of task or planning information, which cannot be directly used for robot path planning or cooperative tasks in complex scenes. The data or coding information of the calibration board usually depends on prior knowledge or additional databases for analysis, and the information is separated from the visual corner, resulting in a complex analysis process.

[0005] Insufficient analysis and error correction capability: Traditional calibration boards cannot complete information analysis and error correction by themselves, and the calibration and positioning process is highly sensitive to the environment and imaging conditions, with low recognition rate. When ordinary QR codes or Aruco codes are directly used for calibration, due to the limitations of coding rules and patterns, the calibration pattern is too complex, the corner information is blocked or missing. The error correction mechanism of traditional two-dimensional codes is not fully integrated with corner detection, and the pattern complexity is high and the recognition requires strict imaging conditions.

[0006] Lack of unified coding and error correction scheme in multi-functional area: When the corner detection area, positioning mark area and data storage area coexist in a multi-functional calibration board, there is a lack of unified coding and error correction scheme, which can easily lead to partial code errors or unanalyzable data, thereby affecting the calibration accuracy and task execution reliability.

[0007] Decoding difficulty in complex scenes: In the data storage and decoding method of existing calibration boards, QR codes or similar codes are prone to code errors in complex scenes, which can cause the calibration board information to be unable to be correctly analyzed. Traditional decoding schemes usually rely on complete image reading or global error correction, and lack precise positioning and correction mechanisms for local code errors, which cannot guarantee data analysis success when there are partial code errors.

[0008] In view of this, the present application proposes a QR calibration board calibration method for multi-scene cooperation to solve the above problems. SUMMARY

[0009] In order to overcome the above-mentioned defects of the prior art, in order to achieve the above-mentioned purpose, the present application provides the following technical scheme: a QR calibration board calibration method for multi-scene cooperation, comprising:

[0010] S1, a fusion calibration board is constructed, a unified two-dimensional matrix structure is adopted, multi-functional areas are integrated in the same calibration board, and collaborative layout is performed according to a preset space rule;

[0011] S2, a positioning mark area and an adjacent formatted information area are generated in the two-dimensional matrix structure; the physical size of the positioning mark area is configured by a fixed internal structure, and format constraint information is stored through the formatted information area;

[0012] S3, the visual corner point detection pattern in the calibration board is configured according to a preset different application scenario, and the corner point area affected by the two-dimensional matrix layout is structurally completed;

[0013] S4, preset task information is written into the data content area in the multi-functional area, and a code word is generated based on a finite field coding rule before writing; whether there is an error in the code stream is determined during decoding; when it is determined that there is no error, valid information is output; when it is determined that there is an error, error position determination and error amplitude correction are performed;

[0014] S5, the position, direction and unit scale of the calibration board in the image are determined based on the positioning mark area; a standard matrix mapping relationship is generated according to the format constraint information and code stream analysis is completed; target positioning, path planning and action control are realized by combining the calibration board identity information obtained by analysis and the corrected code stream.

[0015] Preferably, the integration method of the multi-functional area comprises:

[0016] The fusion calibration board is constructed, the whole calibration board is abstracted as a unified two-dimensional matrix structure, the minimum visual unit is taken as a matrix basic unit, and multi-functional areas are integrated in the same two-dimensional matrix according to a preset space layout rule;

[0017] The multi-functional area includes a positioning mark area, a formatted information area, a time alignment pattern area, a version information area, a calibration corner point detection pattern area, a data storage area and an edge auxiliary corner point pattern area;

[0018] In the two-dimensional matrix, different functional areas are cooperatively laid out according to preset spatial rules, so that the calibration board has the abilities of corner point detection, unique identity recognition, information storage and self-analysis on the same physical carrier, and can still complete calibration and positioning analysis under the condition of occlusion or limited imaging conditions.

[0019] Preferably, the method for generating a positioning identification area and an adjacent formatted information area in a two-dimensional matrix structure comprises:

[0020] After completing the two-dimensional matrix structure and the division of the multi-functional areas of the calibration board, at least one matrix reference position in the two-dimensional matrix is selected as an analysis entry, a positioning identification area is generated according to a preset positioning structure rule, the preset positioning structure rule includes a reference position selection rule, a structure form constraint rule, an adjacent area mapping rule and an analysis entry constraint rule; and a formatted information area is generated in a matrix area adjacent to the positioning identification area and composed of different matrix basic units, with the positioning identification area as a reference.

[0021] Preferably, the method for configuring the physical size of the positioning identification area by using a fixed internal structure comprises:

[0022] After generating the positioning identification area and the adjacent formatted information area, the physical size of the positioning identification area is configured by using a fixed internal structure, the positioning identification area is composed of a preset number of matrix basic units, and the row and column sizes and the internal unit arrangement relationship in the two-dimensional matrix remain unchanged;

[0023] Format constraint information is stored in the formatted information area adjacent to the positioning identification area, the analysis rule of the calibration board is described through the format constraint information, the analysis rule includes one or more of error correction level, mask mode, version number or function configuration identification, the format constraint information is written into the formatted information area according to a preset bit width and arrangement order, and the positioning identification area and the formatted information area maintain a fixed relative spatial position relationship;

[0024] When the calibration board is analyzed by using an image, the formatted information area is directly positioned according to the fixed relative spatial position relationship after the positioning identification area is recognized and the two-dimensional matrix is established, and the code stream reading order and the analysis mode of the data storage area are determined based on the format constraint information stored therein.

[0025] Preferably, the method for structurally completing the corner point area affected by the two-dimensional matrix layout comprises:

[0026] The visual corner point detection pattern in the calibration board is configured according to different preset application scenarios, the different preset application scenarios include a calibration application scenario and a positioning planning application scenario; in the calibration application scenario, a regularly arranged matrix basic unit in a two-dimensional matrix is selected to construct a corner point detection pattern; in the positioning planning application scenario, the number or coverage of the visual corner point detection pattern is simplified and configured;

[0027] An edge auxiliary corner point pattern region is generated at a corner point position affected by a two-dimensional matrix layout, a missing or incomplete corner point structure is structurally completed, so that the completed corner point is consistent with a complete corner point in terms of geometric position relationship, gradient direction and gray scale change; in a corner point detection process, the structurally completed corner point and the complete corner point are uniformly detected and utilized as effective visual corner points.

[0028] Preferably, the method of generating a code word based on a finite field coding rule before writing comprises:

[0029] According to the preset task information required by the preset different application scenarios, the preset task information includes calibration board cell number, cell size, unique code identification range and positioning planning related information;

[0030] The preset task information is converted into a finite field element sequence according to a preset data format, each element in the finite field element sequence belongs to a finite field; the finite field element sequence is constructed into an original code stream polynomial, in the finite field, a Reed-Solomon code is selected to construct a RS code polynomial, and the RS code polynomial is encoded to generate a check code polynomial, to obtain a final code word sequence;

[0031] The generated code word sequence is mapped into the matrix unit of the data content area of the calibration board according to the two-dimensional matrix layout rule, the mapping follows a fixed row and column order and keeps the relative spatial positions of the calibration corner point detection pattern area, the positioning identification area and the formatted information area unchanged.

[0032] Preferably, the method of determining the error position and correcting the error amplitude when it is determined that there is an error comprises:

[0033] In the decoding process, it is determined whether there is an error in the code stream, a Reed-Solomon decoding algorithm is applied to the code word sequence to obtain an error code word sequence, when the error code word sequence is equal to zero, it is determined that there is no error in the code stream, and the decoded code word sequence is directly output as effective information;

[0034] When the error code word sequence is not equal to zero, the error position polynomial is constructed according to the error code word by using the Berlekamp-Massey algorithm, and the error position is determined; the roots of the error position polynomial are solved to obtain the position of each error code word in the code word sequence, the amplitude of each error code word is calculated based on the Forney algorithm, and the corresponding position of the code word sequence is corrected to obtain the corrected effective code word sequence.

[0035] Preferably, the method for generating a standard matrix mapping relationship according to format constraint information and completing code stream analysis comprises:

[0036] The position, direction and unit scale of the calibration plate in the image are determined based on the positioning identification area, the position of the positioning identification area in the image is detected, the image translation position and rotation direction of the calibration plate are determined through at least three corner positioning identifications, and the image scale of each matrix unit is calculated in combination with the corresponding relationship between the known physical size of the positioning identification and the pixel scale in the image;

[0037] According to the calibration plate type and version information data read from the formatting information area, the matrix size, symbol arrangement order and data analysis rule are determined, the mapping relationship from the image pixel coordinates to the matrix logical unit coordinates is established, and the standard matrix mapping table is generated; the symbol information in the data content area is read according to the standard matrix mapping relationship, the arrangement order, blocking rule and masking mode of the code word sequence are determined according to the format constraint information, so as to complete the code stream analysis.

[0038] Preferably, the method for realizing target positioning, path planning and action control comprises:

[0039] After completing the code stream analysis in the calibration plate image and correcting the code word sequence, the calibration plate identity information obtained by analysis is extracted, the calibration plate identity information includes unique coding identification, calibration plate type, size information and version number, and is matched with preset task information to confirm the function area corresponding to the calibration plate;

[0040] In combination with the corrected code word sequence, the number of cells, the size of the cells and the positioning planning parameters are obtained; the matrix unit logical coordinates are mapped to the actual space position based on the position, direction and unit scale of the calibration plate in the image;

[0041] The global coordinates of the target position node are calculated by using the spatial position information of the calibration plate and the preset task information, and the path planning data structure required for robot navigation or operation task is constructed; the path planning data is combined with the robot motion control model, the corresponding control instruction sequence is generated according to the action control algorithm; in the process of robot execution control, the position error is dynamically corrected by continuously collecting the visual feedback of the calibration plate, so as to realize target positioning, path planning and action control.

[0042] A QR calibration board calibration system for multi-scene cooperation, comprising:

[0043] A calibration board construction module is configured to construct a fusion calibration board, adopt a unified two-dimensional matrix structure, integrate multiple functional areas in the same calibration board, and perform cooperative layout according to preset spatial rules.

[0044] A formatted information generation module is configured to generate a positioning identification area and an adjacent formatted information area in the two-dimensional matrix structure, configure the physical size of the positioning identification area by using a fixed internal structure, and store format constraint information through the formatted information area.

[0045] A visual feature construction module is configured to configure visual corner point detection patterns in the calibration board according to preset different application scenarios, and perform structural completion on the corner point area affected by the two-dimensional matrix layout.

[0046] An information encoding error correction module is configured to write preset task information into a data content area in the multiple functional areas, generate a code word based on a finite field encoding rule before writing, determine whether there is an error in the code stream during the decoding process, output valid information when it is determined that there is no error, and perform error position determination and error amplitude correction when it is determined that there is an error.

[0047] A detection and analysis cooperation module is configured to determine the position, direction and unit scale of the calibration board in an image based on the positioning identification area, generate a standard matrix mapping relationship according to the format constraint information and complete code stream analysis, and realize target positioning, path planning and action control in combination with the corrected code stream and the calibration board identity information obtained through analysis.

[0048] Compared with the prior art, the present application has the following beneficial effects:

[0049] The application realizes the unification of calibration and information storage by converting task information into a finite field element sequence according to a preset data format and generating an RS code check code, so that the calibration plate contains high-precision corner point information in a two-dimensional matrix and can store task information and planning data; the calibration plate itself supports task information analysis and error correction without relying on external databases or prior knowledge. The check code generated by finite field coding realizes multi-symbol error detection and correction, so that the task information can be correctly analyzed even in the case of partial occlusion or limited imaging conditions. The fusion of the check code and the original information polynomial ensures the integrity of the data content area and improves the robustness of the calibration plate in complex scenes. The codeword sequence is mapped to the data content area according to a fixed row-column order, and the relative positions of the corner detection area, the positioning mark area and the formatted information area remain unchanged, so that the corner detection and data analysis can still be stably completed under the coordinated layout of the multi-functional area. By defining the data format and mapping rules, the consistency of analysis between different calibration plates is ensured, and the operability of multi-calibration plate cooperation is improved. The task information and error correction capability required to be stored can be selected according to different application scenarios, and the system is flexibly adapted to camera calibration, robot positioning, path planning and other scenes.

[0050] By directly generating error code word sequences in the decoding process, the presence or absence of errors in the code stream is accurately determined, the valid information is quickly output when there is no error, and the analysis efficiency is improved. When there is an error in the code stream, the Berlekamp-Massey algorithm is used to construct an error position polynomial and the Forney algorithm is used to correct the position and amplitude of each error code word, so that the local error is accurately repaired and the overall decoding failure is avoided. The corrected code word sequence can recover the original calibration plate task information, thereby ensuring high reliability and high precision calibration and positioning analysis capability in complex scenes. The multi-functional calibration plate can still be stably analyzed in complex environments, supports partial occlusion, low contrast and multi-calibration plate cooperation scenarios, and improves the application range and overall robustness of the system. BRIEF DESCRIPTION OF DRAWINGS

[0051] Figure 1 A multi-scene cooperative QR calibration plate calibration method flowchart according to the application.

[0052] Figure 2 A multi-scene cooperative QR calibration plate calibration system structure diagram according to the application.

[0053] Figure 3 A multi-functional area diagram according to the application. DETAILED DESCRIPTION

[0054] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described, obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0055] Embodiment 1:

[0056] Please refer to Figure 1 and Figure 3 The embodiment provides a QR calibration plate calibration method for multi-scene cooperation, and specifically comprises the following steps:

[0057] S1, a fusion calibration plate is constructed, a unified two-dimensional matrix structure is adopted, multi-functional areas are integrated in the same calibration plate, and cooperative layout is performed according to a preset space rule;

[0058] S2, a positioning mark area and an adjacent formatted information area are generated in the two-dimensional matrix structure; the physical size of the positioning mark area is configured by using a fixed internal structure, and format constraint information is stored through the formatted information area;

[0059] S3, a visual corner point detection pattern in the calibration plate is configured according to a preset different application scene, and a structural completion is performed on a corner point area affected by the two-dimensional matrix layout;

[0060] S4, preset task information is written into a data content area in the multi-functional area, and a code word is generated based on a finite field coding rule before writing; whether there is an error in the code stream is determined in the decoding process, when it is determined that there is no error, valid information is output; when it is determined that there is an error, error position determination and error amplitude correction are performed;

[0061] S5, the position, direction and unit scale of the calibration plate in the image are determined based on the positioning mark area; a standard matrix mapping relationship is generated according to the format constraint information and code stream analysis is completed; target positioning, path planning and action control are realized in combination with the calibration plate identity information obtained by analysis and the corrected code stream.

[0062] The integration method of the multi-functional area comprises:

[0063] The fusion calibration plate is constructed, the whole calibration plate is abstracted as a unified two-dimensional matrix structure, a minimum visual unit (code element / black and white grid / minimum pixelization square) is taken as a matrix basic unit, and multi-functional areas are integrated in the same two-dimensional matrix according to a preset space layout rule; wherein, the fusion calibration plate is a calibration plate structure which integrates calibration corner points, coding marks and data storage areas into a whole by function division and reuse of the matrix basic unit in the same two-dimensional matrix coordinate system.

[0064] The multi-functional area includes a positioning mark area, a format information area, a time alignment pattern area, a version information area, a calibration corner detection pattern area, a data storage area, and an edge auxiliary corner pattern area.

[0065] It should be noted that the positioning mark area is arranged at at least three corner positions of the two-dimensional matrix, and is used to determine the overall size, direction and reference coordinates of the matrix, and the unit size is consistent with the calibration corner detection unit; the size of the overall matrix is marked, the positioning mark is optimized, and only one black and white grid size is occupied relative to the traditional calibration board; in the use of unique code identification, the size of the positioning mark does not change with the code identification, and a blank area of 4 symbols does not need to be left, so that after inversion, it can be detected and recognized in a manner similar to Aruco.

[0066] The format information area is arranged adjacent to the positioning mark area, and is used to store calibration board type, error correction level, mask type and size information, and to perform redundancy check through error correction coding; the format information area saves information data correction level and mask type, the present application extends 15bits format information of QR code to 18bits to store the format information of the current calibration board, and the format information includes calibration board type, correction level, mask and correction code; in order to ensure the accuracy of corner detection, the data position is adjusted correspondingly.

[0067] The time alignment pattern area is arranged along the row direction and column direction of the two-dimensional matrix, and is used to constrain the periodic arrangement relationship of the matrix unit and assist in symbol size estimation; the time alignment pattern of ordinary QR code is used, mainly for unique identification and large capacity information storage.

[0068] The version information area is used for calibration as the main purpose, and the size of the calibration board is determined by the black and white squares of the calibration board, so that the 18bits version information can be reserved or stored as calibration board corner detection pattern information. For positioning and planning guidance scene, the version information stores the QR code version information.

[0069] The calibration corner detection pattern area is located in the main internal area of the two-dimensional matrix, and is arranged in the manner of chessboard or fusion of chessboard and code unit, and is used to provide high-precision visual corner features; in the calibration scene, the calibration pattern still uses the chessboard or ChAruco calibration pattern, and in the positioning and planning guidance scene, the number of positioning detection corners is simplified compared with the calibration scene, so as to leave more data space for planning information guidance, and the data space is the right lower area surrounded by the three positioning mark areas.

[0070] The data storage area is arranged in a predetermined matrix sub-area enclosed by the positioning identification area, and is used to store calibration plate parameters, unique coded identification, and positioning or planning information; the data content includes the number of calibration plate cells, cell size, and the start and end range of the unique coded identification contained.

[0071] The edge auxiliary corner point pattern area is arranged at the edge of the calibration plate or near the boundary of the information area, and is used to solve the problem of missing or incomplete detection of part of the corner points caused by the integration of multiple functional areas. Compared with the traditional pattern of all black and white cells, the area between the positioning identification areas is used to store the format calibration plate information, which causes the corner points in the second row and the second column to be arranged from the third to be unable to be normally detected, and therefore the edge auxiliary corner point pattern is added to complete the corner points.

[0072] In the two-dimensional matrix, different functional areas are cooperatively laid out according to a preset spatial rule, so that the calibration plate has the capabilities of corner point detection, unique identity recognition, information storage, and self-analysis on the same physical carrier, and can still complete calibration and positioning analysis under the condition of occlusion or limited imaging conditions.

[0073] It should be noted that the preset spatial rule includes taking the matrix cell where the positioning identification area is located as the reference position of the matrix direction and scale, fixedly arranging the formatted information area and the time alignment pattern according to the relative row-column relationship with the positioning identification area, dividing and arranging the calibration corner point detection area, the unique coded area, and the data content area in different sub-areas of the two-dimensional matrix without destroying the continuity of the calibration corner point detection, and introducing the edge auxiliary corner point to structurally complete the missing corner point position caused by the division of the functional area, so that the functional areas cooperatively coexist on the same physical carrier.

[0074] The method for generating the positioning identification area and the adjacent formatted information area in the two-dimensional matrix structure includes:

[0075] After the two-dimensional matrix structure of the calibration plate and the division of the multiple functional areas are completed, at least one matrix reference position is selected as an analysis entrance in the two-dimensional matrix, the positioning identification area is generated according to the preset positioning structure rule, the preset positioning structure rule includes the reference position selection rule, the structure form constraint rule, the adjacent area mapping rule, and the analysis entrance constraint rule; and the formatted information area (for example, a 4×6 formatted information block) is generated in the matrix area adjacent to the positioning identification area and composed of different matrix basic units, with the positioning identification area as a reference.

[0076] It should be noted that the reference position selection rule is used to determine the generation starting point of the positioning mark area in the two-dimensional matrix structure. The reference position is a predefined matrix coordinate in the two-dimensional matrix, which has a determined offset relationship with respect to the overall boundary of the calibration plate, so as to ensure that the positioning mark area can be preferentially detected and used as a spatial reference for subsequent analysis under different imaging scales and cropping conditions.

[0077] The structure form constraint rule is used to limit the internal matrix arrangement mode of the positioning mark area. The positioning mark area is composed of a plurality of matrix basic units according to a fixed black and white distribution relationship, and the structure still has unique geometric and topological characteristics under the conditions of rotation, scaling or local occlusion, so that the analysis end can determine the direction, scale and matrix coordinate system of the calibration plate through local pattern recognition.

[0078] The adjacent area mapping rule is used to define the relative position relationship between the formatted information area and the positioning mark area. According to the rule, the formatted information area is only allowed to be generated in the matrix area directly adjacent to the positioning mark area, and the relative direction, interval unit number and arrangement order between them are fixed, so that the analysis end can determine the specific position and reading direction of the formatted information area without global search after detecting the positioning mark area.

[0079] The analysis entry constraint rule is used to define the positioning mark area as the starting mark of the two-dimensional matrix analysis process. In the analysis process, the analysis end first establishes the matrix coordinate system and the symbol scale according to the positioning mark area, and then analyzes the formatted information area adjacent to the positioning mark area according to the adjacent area mapping rule, thereby forming a unified and reproducible analysis entry mechanism.

[0080] The method for configuring the physical size of the positioning mark area by using a fixed internal structure includes:

[0081] After generating the positioning mark area and the adjacent formatted information area, the physical size of the positioning mark area is configured by using a fixed internal structure. The positioning mark area is composed of a preset number of matrix basic units, and the row and column size and internal unit arrangement relationship in the two-dimensional matrix remain unchanged, so that the positioning mark area occupies a fixed matrix range under different calibration plate configurations and different data content conditions, thereby serving as a scale reference structure for determining the symbol scale and overall scaling ratio in the two-dimensional matrix.

[0082] The format constraint information is stored in the formatted information area adjacent to the positioning mark area. The analysis rule includes one or more of the error correction level, the mask mode, the version number or the function configuration identification, and the format constraint information is written into the formatted information area according to a preset bit width and arrangement order, and maintains a fixed relative spatial position relationship with the positioning mark area.

[0083] When performing image analysis on the calibration board, after identifying the positioning marker area and establishing a two-dimensional matrix, the formatted information area is directly located based on the fixed relative spatial position relationship, and the code stream reading order and parsing method of the data storage area are determined based on the format constraint information stored therein.

[0084] Methods for structural completion of corner regions affected by two-dimensional matrix layout include:

[0085] The visual corner detection patterns in the calibration board are configured according to different preset application scenarios, including calibration application scenarios and positioning planning application scenarios. In the calibration application scenario, a regular corner detection pattern is constructed by selecting continuously arranged matrix basic units in a two-dimensional matrix to form a visual corner with accuracy that meets sub-pixel level positioning requirements. In the positioning planning application scenario, the number or coverage of the visual corner detection patterns is simplified to reserve more matrix units for the data storage area.

[0086] Furthermore, when the introduction of positioning marker areas, formatted information areas, or data storage areas causes some corner points to no longer have complete unit structures in the two-dimensional matrix, edge auxiliary corner point pattern areas are generated at the corner point positions affected by the two-dimensional matrix layout to structurally complete the missing or incomplete corner point structures. This ensures that the completed corner points are consistent with the complete corner points in terms of geometric positional relationships, gradient directions, and grayscale changes. During the corner point detection process, the structurally completed corner points and the complete corner points are uniformly used as effective visual corner points for detection and utilization. Thus, even under the circumstances of multi-functional area collaborative layout and the presence of occlusion or limited imaging conditions, stable camera calibration, target localization, or path planning analysis can still be completed.

[0087] Methods for generating codewords based on finite field encoding rules before writing include:

[0088] Preset task information according to the storage requirements of different application scenarios. The preset task information includes the number of calibration board cells, cell size, unique code identification range and positioning planning related information.

[0089] The preset task information is converted into a finite field element sequence according to a preset data format. The finite field element sequence is as follows: ;in, It represents a sequence of elements in a finite field, that is, a set of encoded symbols for the preset task information in a finite field; This indicates the 0th element in the sequence of elements in the finite field, which is the first information field in the sequence of task information, such as the calibration board type; This represents the first element in the sequence of elements of a finite field; Represents the th element in the sequence of elements of a finite field. One element; denotes a sequence length symbol, and is a total length of the finite field element sequence;

[0090] Each element in the finite field element sequence belongs to a finite field GF(256); wherein, GF(256) denotes a finite field, which is a Galois field of 256 elements; it should be noted that the preset data format is a task information coding specification determined before writing, and the task information coding specification includes field order, bit number occupied by each field, how to pack into a finite field element, and how to correspond to a final two-dimensional matrix code element, which ensures the uniformity of coding, writing, parsing and error correction; the finite field coding requires that each symbol belongs to a finite field, therefore, the preset data format usually divides all task information into a plurality of symbols, each symbol has a fixed length (such as 1 byte = 8 bits), and forms a finite field element sequence. The data format needs to agree on the arrangement order of each task information in the code word sequence in advance, otherwise the decoding end cannot correctly parse. For example, the first 3 bits: calibration plate type, the last 2 bits: error correction level, the next 8 bits: unique code identification start, the next 8 bits: unique code identification end, and the remaining bits: path planning or other task data.

[0091] The finite field element sequence is constructed into an original code stream polynomial, and the original code stream polynomial is ; wherein, denotes an original code stream polynomial of the preset task information in the finite field; denotes a formal variable of the original code stream polynomial, which is used to represent the position of each element in the original code stream polynomial; in the finite field, a Reed-Solomon code is selected to construct a RS code polynomial, and the RS code polynomial is: ; wherein, denotes the RS code polynomial; denotes a correctable code word number; denotes an index of a generator of the RS code polynomial, and the value is from 0 to ; denotes a generator of the finite field GF(256); denotes a non-zero element in GF(256);

[0092] The RS code polynomial is encoded to generate a check code polynomial, and a final code word sequence is obtained; the final code word sequence is: ; wherein, denotes the final code word sequence; denotes that the original information polynomial is left shifted by bits to give a position to the check code; denotes a remainder of the RS code polynomial generated by taking a modulus of ;

[0093] The generated code word sequence is mapped into the matrix units of the data content area of the calibration board according to a two-dimensional matrix layout rule, and the mapping follows a fixed row-column order and keeps the relative spatial positions of the calibration corner detection pattern area, the positioning mark area and the formatted information area unchanged.

[0094] The technical problems existing in the prior art are solved. The existing chessboard, circular positioning or Aruco calibration board can usually only provide corner positioning information, cannot carry a large amount of task or planning information, and cannot be directly used for robot path planning or complex scene cooperation. In the traditional method, the data or coding information of the calibration board often depends on prior knowledge or an additional database for analysis, the information is separated from the visual corner, and the analysis process is complex. The calibration board itself cannot complete information analysis and error correction, resulting in sensitivity to the environment and imaging conditions during calibration and positioning, and low recognition rate. The ordinary QR code or Aruco code is directly used for calibration, which is limited by the coding rules and patterns, and is easy to cause the calibration pattern to be too complex, the corner information to be blocked or missing. The error correction mechanism of the traditional two-dimensional code is not fully integrated with the corner detection of the calibration board, the pattern complexity is high, and the recognition requires strict imaging conditions. When the corner detection area, the positioning mark area and the data storage area coexist in the multifunctional calibration board, there is a lack of unified coding and error correction scheme, which is easy to cause part of the code elements to be wrong or the data to be unanalyzable.

[0095] When it is determined that there is an error, the method for error position determination and error amplitude correction comprises:

[0096] In the decoding process, it is determined whether there is an error in the code stream, a decoding algorithm of Reed-Solomon coding is applied to the code word sequence to obtain an error code word sequence, when the error code word sequence is equal to zero, it is determined that there is no error in the code stream, and the decoded code word sequence is directly output as valid information;

[0097] When the error code word sequence is not equal to zero, a Berlekamp-Massey algorithm is used to construct an error position polynomial according to the error code word to determine the error position; the error position polynomial is: ; wherein, represents the error position polynomial; represents the number of error code words; represents the position element of the th error code word in the finite field GF(256); represents the index of the error code word;

[0098] The root of the error position polynomial is solved to obtain the position of each error code word in the code word sequence, the amplitude of each error code word is calculated based on the Forney algorithm, and the corresponding position of the code word sequence is corrected to obtain a corrected valid code word sequence.

[0099] The corrected valid codeword sequence is: ; wherein, represents the corrected valid codeword sequence, that is, the final correct codeword sequence; represents the error codeword sequence;

[0100] The following technical problems existing in the prior art are solved: In the data storage and decoding method of the existing calibration board, the QR code or similar code is prone to codeword errors in a complex scene (such as partial occlusion, blurred imaging, reflection or low contrast), which leads to the failure to correctly parse the calibration board information. The traditional decoding scheme usually relies on complete reading of the entire image or global error correction, lacks accurate positioning and correction mechanism for local codeword errors, and cannot ensure successful data parsing when there are partial errors. In the cooperative use of multiple scenes, the calibration board not only carries the corner positioning information, but also stores the task planning, unique code and other data. If the codeword parsing fails, it will cause the calibration, positioning and robot planning tasks to be interrupted, and the system robustness and reliability will be reduced.

[0101] The method for generating a standard matrix mapping relationship according to the format constraint information and completing the bitstream parsing includes:

[0102] The position, direction and unit scale of the calibration board in the image are determined based on the positioning mark area, the position of the positioning mark area in the image is detected, the image translation position and rotation direction of the calibration board are determined through at least three corner positioning marks, and the image scale of each matrix unit is calculated by combining the known physical size of the positioning mark and the corresponding relationship between the pixel scale in the image.

[0103] It should be noted that calculating the image scale of each matrix unit includes first accurately identifying the position of the positioning mark area through an image processing algorithm (such as binarization, edge detection or gradient detection). Generally, the calibration board is provided with at least three corner positioning marks, and these marks have known real sizes (such as side length) on the physical calibration board. According to the pixel coordinates of the three corner marks in the image, the affine transformation of the calibration board in the image can be determined, including the translation vector, the rotation angle and the scaling factor. By calculating the pixel distances Px and Py between the corner marks in the image, and corresponding them with the actual physical distance, the proportional relationship between the pixels and the physical units (pixels) in the image can be obtained. Using this proportional relationship, the physical size of the logical unit (that is, each code element or black and white grid) of the entire two-dimensional matrix is mapped to the image pixel scale, so as to accurately determine the pixel size and position of each matrix unit in the image.

[0104] According to the calibration plate type and version information data read by the formatting information area, the matrix size, symbol arrangement order and data parsing rule are determined, and a mapping relationship from image pixel coordinates to matrix logical unit coordinates is established to generate a standard matrix mapping table; symbol information in the data content area is read according to the standard matrix mapping relationship, and the arrangement order of code word sequence, blocking rule and mask mode are determined according to the format constraint information, so as to complete code stream parsing.

[0105] It should be noted that the process of reading the calibration plate type and version information according to the formatting information, determining the matrix size, symbol arrangement order and establishing the pixel to logical unit mapping relationship includes: after detecting the positioning mark area in the image and calculating the matrix unit image size, reading the encoded data in the formatting information area adjacent to the positioning mark area. The formatting information usually contains fields such as calibration plate type, version number, error correction level, mask type, each field occupies a fixed number of bits, and is arranged in a predetermined order. According to the calibration plate type and version information, the size (number of rows and columns) of the logical matrix, the arrangement order of each symbol or unit in the matrix, and the blocking mode of the data content area can be determined.

[0106] Using the known image size and matrix logical size, the image pixel coordinates are mapped to the matrix logical unit coordinates, that is, the corresponding mapping relationship from image coordinates to matrix row and column indexes is established. This mapping relationship can be represented as a standard matrix mapping table, in which each matrix unit corresponds to a pixel area in the image. According to the mapping table, the symbol information of the data content area is read in order, and combined with the code word arrangement order, blocking rule and mask mode specified in the formatting information, the code stream parsing can be completed, and the preset task information in the calibration plate can be accurately decoded.

[0107] The method for realizing target positioning, path planning and action control includes:

[0108] After completing the code stream parsing in the calibration plate image and correcting the code word sequence, the calibration plate identity information obtained by parsing is extracted, including unique code identification, calibration plate type, size information and version number, and is matched with the preset task information to confirm the function area corresponding to the calibration plate;

[0109] Combined with the corrected code word sequence, the number of cells, cell size and positioning planning parameters are obtained; based on the position, direction and cell size of the calibration plate in the image, the matrix unit logical coordinates are mapped to the actual space position to realize accurate positioning of the calibration plate in the three-dimensional space;

[0110] The spatial position information of the calibration board and the preset task information are used to calculate the global coordinates of the target position node, and a path planning data structure required for robot navigation or operation task is constructed. Specifically, the logical coordinates of each matrix unit on the calibration board are mapped to geometric nodes in a unified spatial coordinate system to form a spatial node set containing node coordinates, relative distances between nodes, and reachable relationships. Meanwhile, the node set is functionally labeled in combination with the obtained preset task information to distinguish target nodes, passing nodes, and forbidden area nodes, thereby constructing a path planning data structure suitable for robot navigation or operation tasks.

[0111] The path planning data is combined with a robot motion control model to generate a corresponding control instruction sequence according to a motion control algorithm. It should be noted that after obtaining the path planning data structure, the system converts the spatial path nodes into a set of motion parameters matching the robot's execution capabilities based on the robot's kinematics or dynamics model. Specifically, the path node sequence is decomposed into continuous pose changes or trajectory segments, and the path is smoothed and timed according to the robot joint constraints, speed constraints, and acceleration constraints to generate path planning data that meets the requirements of the robot control interface. Subsequently, the path planning data is associated with the robot motion control model, and the spatial path information is mapped to a specific control instruction sequence, including displacement instructions, steering instructions, or joint driving instructions, through the motion control algorithm, so that the robot can execute the corresponding navigation or operation actions according to the planned path.

[0112] During robot execution control, the system continuously collects visual feedback of the calibration board to dynamically correct position errors and achieve target positioning, path planning, and motion control.

[0113] It should be noted that during the execution of the control instructions by the robot, the system continuously collects real-time images containing the calibration board as visual feedback information, and repeatedly detects and updates the current spatial position of the calibration board based on the positioning mark area. By comparing the real-time detected spatial position of the calibration board with the spatial position information of the calibration board used in the planning stage, the current position deviation and attitude deviation are calculated, and the deviation amount is fed back to the path planning data structure or the motion control model for dynamic correction of subsequent control instructions. Thus, a closed-loop control mechanism is formed with the calibration board as the spatial reference, so that the robot can continuously correct the motion trajectory in the presence of environmental disturbances or execution errors, and ultimately achieve stable and accurate target positioning, path planning, and motion control.

[0114] In this embodiment, the task information is converted into a finite field element sequence according to a preset data format and an RS code check code is generated. The calibration plate contains high-precision corner point information in a two-dimensional matrix and can store task information and planning data, realizing the unification of calibration and information storage. The calibration plate itself supports task information analysis and error correction without relying on external databases or prior knowledge. The check code is generated by finite field coding to realize multi-symbol error detection and correction. Even in the case of partial occlusion or limited imaging conditions, the task information can be correctly parsed. The fusion of the check code and the original information polynomial ensures the integrity of the data content area and improves the robustness of the calibration plate in complex scenes. The code word sequence is mapped to the data content area according to a fixed row-column order, and the relative positions of the corner detection area, positioning mark area and formatted information area remain unchanged, ensuring stable corner detection and data parsing in the collaborative layout of the multi-functional area. By defining the data format and mapping rules, the consistency of parsing between different calibration plates is ensured, and the operability of multi-calibration plate collaboration is improved. The task information and error correction capability required to be stored can be selected according to different application scenarios, and the system is flexibly adapted to camera calibration, robot positioning, path planning and other scenarios.

[0115] By directly generating error code word sequences in the decoding process, the presence or absence of errors in the code stream is accurately determined, ensuring fast output of valid information when there are no errors and improving parsing efficiency. When there are errors in the code stream, the Berlekamp-Massey algorithm is used to construct an error position polynomial and the Forney algorithm is used to correct the position and amplitude of each error code word, realizing accurate repair of local errors and avoiding overall decoding failure. The corrected code word sequence can recover the original calibration plate task information, thereby ensuring high reliability and high-precision calibration and positioning analysis capability in complex scenes. The multi-functional calibration plate can still be parsed stably in complex environments, supporting partial occlusion, low contrast and multi-calibration plate collaborative use scenarios, and improving the application range and overall robustness of the system.

[0116] Embodiment 2:

[0117] Please refer to Figure 2 The embodiment not described in detail is described in embodiment 1. A multi-scene collaborative QR calibration plate calibration system is provided, which includes:

[0118] A calibration plate construction module is used to construct a fusion calibration plate. A unified two-dimensional matrix structure is adopted to integrate multi-functional areas in the same calibration plate and to collaboratively layout according to a preset spatial rule.

[0119] The formatting information generation module generates a positioning identification area and an adjacent formatting information area in a two-dimensional matrix structure; a physical size of the positioning identification area is configured by using a fixed internal structure, and format constraint information is stored through the formatting information area;

[0120] The visual feature construction module configures a visual corner point detection pattern in the calibration board according to a preset different application scenario, and performs structural completion on a corner point area affected by a two-dimensional matrix layout;

[0121] The information encoding and error correction module writes preset task information into a data content area in the multifunctional area, and generates a code word based on a finite field coding rule before writing; in a decoding process, it is determined whether there is an error in the code stream, when it is determined that there is no error, valid information is output; when it is determined that there is an error, error position determination and error amplitude correction are performed;

[0122] The detection and analysis coordination module determines the position, direction and unit scale of the calibration board in the image based on the positioning identification area; generates a standard matrix mapping relationship based on the format constraint information and completes code stream analysis; and combines the corrected code stream and the calibration board identity information obtained through analysis to realize target positioning, path planning and action control.

[0123] Since the electronic device introduced in the embodiment is the electronic device used to implement the QR calibration board calibration method and system based on a multi-scene cooperation in the embodiment, the specific implementation of the electronic device and its various forms can be understood by those skilled in the art based on the QR calibration board calibration method and system based on a multi-scene cooperation in the embodiment, so the implementation of the method in the embodiment will not be described in detail. As long as the electronic device used in the QR calibration board calibration method and system based on a multi-scene cooperation in the embodiment is implemented by those skilled in the art, it belongs to the scope of protection of the present application.

[0124] The above formulas are dimensionless numerical calculations, the formulas are obtained by software simulation of a large amount of data to obtain a formula closest to the real situation, and the preset parameters and threshold values in the formula are set by those skilled in the art according to the actual situation.

[0125] The above is only the preferred embodiment of the present application, the protection scope of the present application is not limited to the above-mentioned embodiments, any technical solution belonging to the idea of the present application is also within the protection scope of the present application. It should be noted that for ordinary technical operators in the technical field, some improvements and decorations without departing from the principle of the present application are also considered as the protection scope of the present application.

Claims

1. A QR calibration board calibration method for multi-scene cooperation, characterized in that, The application relates to a method for constructing a fusion calibration board. S1, a fusion calibration board is constructed, a unified two-dimensional matrix structure is adopted, multifunctional areas are integrated in the same calibration board, and collaborative layout is performed according to preset space rules; S2, a positioning mark area and an adjacent formatted information area are generated in the two-dimensional matrix structure; the physical size of the positioning mark area is configured by using a fixed internal structure, and format constraint information is stored in the formatted information area; S3, visual corner point detection patterns in the calibration board are configured according to preset different application scenarios, and the structure of the corner point area affected by the two-dimensional matrix layout is complemented; S4, preset task information is written into a data content area in the multifunctional area, and a code word is generated based on a finite field coding rule before writing; whether an error exists in the code stream is judged during the decoding process; when it is judged that no error exists, valid information is output; when it is judged that an error exists, error position determination and error amplitude correction are performed; S5, the position, direction and unit scale of the calibration board in an image are determined based on the positioning mark area; a standard matrix mapping relationship is generated according to the format constraint information, and code stream analysis is completed; target positioning, path planning and action control are realized by combining the calibration board identity information obtained through analysis and the corrected code stream. The integration method of the multifunctional area comprises the following steps: The fusion calibration board is constructed, the whole calibration board is abstracted as a unified two-dimensional matrix structure, the smallest visual unit is taken as a matrix basic unit, and multifunctional areas are integrated in the same two-dimensional matrix according to preset space layout rules; 2. The QR calibration board calibration method for multi-scene cooperation according to claim 1, wherein, The multifunctional area comprises a positioning mark area, a formatted information area, a time alignment pattern area, a version information area, a calibration corner point detection pattern area, a data storage area and an edge auxiliary corner point pattern area; In the two-dimensional matrix, different functional areas are collaboratively laid out according to preset space rules, so that the calibration board simultaneously has the abilities of corner point detection, unique identity recognition, information storage and self-analysis on the same physical carrier, and the calibration and positioning analysis can still be completed under the condition that the calibration board is shielded or the imaging condition is limited. The method for generating the positioning mark area and the adjacent formatted information area in the two-dimensional matrix structure comprises the following steps: After the two-dimensional matrix structure and the multifunctional area division of the calibration board are completed, at least one matrix reference position is selected as an analysis entrance in the two-dimensional matrix, the positioning mark area is generated according to preset positioning structure rules, the preset positioning structure rules comprise reference position selection rules, structure form constraint rules, adjacent area mapping rules and analysis entrance constraint rules; and the formatted information area is generated in the matrix area composed of different matrix basic units and directly adjacent to the positioning mark area by taking the positioning mark area as a reference.

3. The QR calibration board calibration method for multi-scene cooperation according to claim 2, characterized in that, The method for configuring the physical size of the positioning mark area by using a fixed internal structure comprises the following steps: After the positioning mark area and the adjacent formatted information area are generated, the physical size of the positioning mark area is configured by using a fixed internal structure, the positioning mark area is composed of a preset number of matrix basic units, and the row and column sizes and the internal unit arrangement relationship in the two-dimensional matrix remain unchanged; 4. The QR calibration board calibration method for multi-scene cooperation according to claim 3, characterized in that, ​ ​ The format constraint information is stored in a formatted information area adjacent to the positioning identification area, the parsing rule of the calibration board is described by the format constraint information, the parsing rule includes one or more of error correction level, mask mode, version number or function configuration identification, the format constraint information is written in the formatted information area according to a preset bit width and arrangement order, and a fixed relative spatial position relationship is maintained with the positioning identification area; When the calibration board is parsed, the formatted information area is directly positioned according to the fixed relative spatial position relationship after the positioning identification area is recognized and the two-dimensional matrix is established, and the code stream reading sequence and the parsing mode of the data storage area are determined based on the format constraint information stored therein.

5. The QR calibration board calibration method for multi-scene cooperation according to claim 4, characterized in that, The method for structurally completing the corner point area affected by the two-dimensional matrix layout includes: The visual corner point detection pattern in the calibration board is configured according to different application scenarios, the different application scenarios include a calibration application scenario and a positioning planning application scenario; in the calibration application scenario, the regularly arranged matrix basic units in the two-dimensional matrix are selected to construct the corner point detection pattern; in the positioning planning application scenario, the number or coverage of the visual corner point detection pattern is simplified and configured; An edge auxiliary corner point pattern area is generated at the corner point position affected by the two-dimensional matrix layout, the missing or incomplete corner point structure is structurally completed, and the completed corner point is consistent with the complete corner point in terms of geometric position relationship, gradient direction and gray change; in the corner point detection process, the structurally completed corner point and the complete corner point are uniformly used as effective visual corner points for detection and utilization.

6. The QR calibration board calibration method for multi-scene cooperation according to claim 5, characterized in that, The method for generating a code word based on a finite field coding rule before writing includes: The preset task information is required according to the task requirement in the preset different application scenarios, the preset task information includes the number of calibration board cells, the cell size, the unique code identification range and the positioning planning related information; The preset task information is converted into a finite field element sequence according to a preset data format, each element in the finite field element sequence belongs to a finite field, the finite field element sequence is constructed into an original code stream polynomial, in the finite field, the Reed-Solomon code is selected to construct an RS code polynomial, and the RS code polynomial is encoded to generate a check code polynomial, to obtain a final code word sequence; The generated code word sequence is mapped into the matrix unit of the data content area of the calibration board according to the two-dimensional matrix layout rule, the mapping follows a fixed row and column order and maintains a fixed relative spatial position with the calibration corner point detection pattern area, the positioning identification area and the formatted information area.

7. The QR calibration board calibration method for multi-scene cooperation according to claim 6, characterized in that, The method for determining the error position and correcting the error amplitude when it is determined that there is an error includes: In the decoding process, it is determined whether there is an error in the code stream, the Reed-Solomon decoding algorithm is applied to the code word sequence to obtain an error code word sequence, when the error code word sequence is equal to zero, it is determined that there is no error in the code stream, and the decoded code word sequence is directly output as valid information. When the error code word sequence is not equal to zero, the error position polynomial is constructed according to the error code word by using the Berlekamp-Massey algorithm, and the error position is determined; the roots of the error position polynomial are solved, the position of each error code word in the code word sequence is obtained, the amplitude of each error code word is calculated based on the Forney algorithm, and the corresponding position of the code word sequence is corrected to obtain the corrected effective code word sequence.

8. The QR calibration board calibration method for multi-scene cooperation according to claim 7, characterized in that, The method for generating a standard matrix mapping relationship according to the format constraint information and completing code stream analysis comprises: The position, direction and unit scale of the calibration plate in the image are determined based on the positioning identification area, the position of the positioning identification area in the image is detected, the image translation position and rotation direction of the calibration plate are determined through at least three corner positioning identifications, and the image scale of each matrix unit is calculated in combination with the correspondence between the known physical size of the positioning identification and the pixel scale in the image; The matrix size, symbol arrangement order and data analysis rule are determined according to the calibration plate type and version information data read from the format information area, and the mapping relationship from the image pixel coordinates to the matrix logical unit coordinates is established to generate a standard matrix mapping table; the symbol information in the data content area is read according to the standard matrix mapping relationship, the arrangement order, blocking rule and masking mode of the code word sequence are determined according to the format constraint information, and thus the code stream analysis is completed.

9. The QR calibration board calibration method for multi-scene cooperation according to claim 8, wherein, The method for realizing target positioning, path planning and action control comprises: After the code stream analysis in the calibration plate image is completed and the code word sequence is corrected, the calibration plate identity information obtained through analysis is extracted, the calibration plate identity information comprises a unique code identification, a calibration plate type, size information and a version number, and is matched with preset task information to confirm the function area corresponding to the calibration plate; In combination with the corrected code word sequence, the number of unit cells, the size of unit cells and positioning planning parameters are obtained; based on the position, direction and unit scale of the calibration plate in the image, the matrix unit logical coordinates are mapped to actual space positions; The global coordinates of the target position node are calculated by using the space position information of the calibration plate and the preset task information, and the path planning data structure required for robot navigation or operation tasks is constructed; the path planning data are combined with a robot motion control model to generate corresponding control instruction sequences according to an action control algorithm; in the process of robot execution control, the position error is dynamically corrected by continuously collecting the visual feedback of the calibration plate to realize target positioning, path planning and action control.

10. A multi-scenario cooperative oriented QR calibration board calibration system for implementing the method of any one of claims 1 to 9, characterized in that, The method comprises: The calibration plate construction module is configured to construct a fusion calibration plate, adopt a unified two-dimensional matrix structure, integrate multiple functional areas in the same calibration plate, and perform collaborative layout according to a preset space rule; The format information generation module is configured to generate a positioning identification area and an adjacent format information area in the two-dimensional matrix structure, configure the physical size of the positioning identification area by using a fixed internal structure, and store format constraint information through the format information area; The visual feature construction module is configured to configure visual corner point detection patterns in the calibration plate according to preset different application scenarios, and perform structural completion on the corner point areas affected by the two-dimensional matrix layout. The information coding error correction module writes preset task information into a data content area in the multifunctional area, and generates a code word based on a finite field coding rule before writing; in a decoding process, it is determined whether there is an error in the code stream; when it is determined that there is no error, valid information is output; When it is determined that there is an error, error position determination and error amplitude correction are performed; The detection and analysis coordination module determines the position, direction and unit scale of the calibration plate in the image based on the positioning identification area; According to the format constraint information, a standard matrix mapping relationship is generated and code stream analysis is completed; combined with the calibration plate identity information obtained by analysis and the corrected code stream, target positioning, path planning and action control are realized.

Images