An arc matching method and system for automatic labeling

Abstract

The application provides a curvature matching method and system for automatic labeling, and relates to the technical field of data processing.The method comprises the following steps: indexing each contour point in a forward order and a reverse order, identifying the change direction of the contour points and merging the contour points with the same change direction into a change interval, merging the change intervals with the same change direction into a section, identifying the offset degree of the closed section along the attachment direction, obtaining a section shape offset value, extracting the boundary contour points between adjacent closed sections, connecting the boundary connection relationship between each closed section in series, identifying the consistency degree of the connection between different closed sections, obtaining an overall contour coupling value, and correcting and adjusting the preset attachment path according to the section shape offset value and the overall contour coupling value to obtain target attachment path data.The application can realize the adaptation of the attachment path to the actual curved surface contour of the workpiece, so that the attachment path can be adjusted according to the contour change.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for automatic labeling with arc matching. Background Technology

[0002] In the field of instrument manufacturing, in order to achieve automated installation of nameplates, existing technologies usually adopt an installation method based on automatic labeling equipment. First, the connecting structures on the metal nameplate, such as metal pins, are pre-bent. Then, the nameplate and the instrument workpiece to be labeled are placed in the corresponding stations of the equipment, and the positioning structure completes the initial alignment. Subsequently, the equipment is started, and the multi-segment folding mold or pressing mechanism presses and covers the nameplate sequentially according to the preset trajectory, so that it is attached or fixed along the outer surface of the instrument, realizing the automatic installation process of the nameplate.

[0003] However, in practical applications, this automatic labeling method is usually based on a fixed mold trajectory or a preset pressing path, which may lack the ability to adaptively match the surface curvature of different instrument housings. This can lead to inconsistencies in localized adhesion during the application process on curved surfaces. For example, when installing labels on arc-shaped instrument housings with significant curvature, the label may not adjust its adhesion angle and force distribution synchronously with the surface curvature during the folding and pressing process, potentially resulting in warping or gaps at the ends or edges of the label. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for automatic labeling with arc matching, in order to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: In a first aspect, a method for matching the curvature of labels for automatic labeling, the method comprising: Obtain the contour point data of the workpiece to be mounted; Based on the contour point data, index sequences are established for each contour point in both forward and reverse order. The forward and reverse index sequences are then inserted alternately according to their positions to obtain a bidirectional fusion sequence. Based on the bidirectional fusion sequence, the changing direction of the contour points is identified and contour points with the same changing direction are merged into a change interval. The change intervals with lengths that meet the standard are used as the constraint source intervals to obtain the constraint interval sequence. Based on the constraint interval sequence, the intervals with the same direction of change are merged into segments, and the range of the segments is expanded to both sides with the constraint source interval as the center to obtain a closed segment set; Based on the set of closed segments, a unidirectional change path is established in each closed segment according to the contour point sequence. The degree of offset of the closed segment along the attachment direction is identified, and the segment shape offset value is obtained. Based on the set of closed segments, extract the boundary contour points between adjacent closed segments, and establish the boundary connection relationship between closed segments based on the boundary contour points to obtain the segment boundary mapping set; Based on the segment boundary mapping set, the boundary connection relationships between each closed segment are connected in series and combined to identify the degree of connection consistency between different closed segments and obtain the overall contour coupling value. By mapping the preset attachment path to each closed segment, the preset attachment path is corrected and adjusted according to the segment shape offset value and the overall contour coupling value to obtain the target attachment path data.

[0006] Furthermore, based on the contour point data, index sequences are established for each contour point in both forward and reverse order. These forward and reverse index sequences are then alternately inserted according to their positions to obtain a bidirectional fusion sequence, including: Based on the contour point data, each contour point is sequentially numbered according to the collection order of the contour points, and arranged in ascending order to form a forward point column, and in descending order to form a reverse point column. Based on the forward point column and the reverse point column, the contour points in the forward point column and the reverse point column are labeled with the number of each contour point as the index identifier, so as to obtain the forward index sequence and the reverse index sequence. By alternating the contour points at the same position in the forward index sequence and the reverse index sequence, ensuring that each position contains both forward index information and reverse index information, an alternating fusion sequence is obtained. Based on the alternating fusion sequence, the forward and reverse index information of each contour point is combined and identified, and the sequence position is recalibrated according to the alternating arrangement order to obtain the bidirectional fusion sequence.

[0007] Furthermore, based on the bidirectional fusion sequence, the direction of change of contour points is identified, and contour points with the same direction of change are merged into a change interval. Change intervals whose lengths meet the standard are used as constraint source intervals, resulting in a constraint interval sequence, including: Based on the bidirectional fusion sequence, the direction of change between contour points is determined by comparing the increase or decrease of the index values ​​of adjacent contour points, thus obtaining the direction identification sequence. Based on the direction identification sequence, the contour points with continuous and consistent change directions are merged according to their arrangement order in the bidirectional fusion sequence, and the start and end positions are recorded to obtain the change interval set; Based on the set of change intervals, count the number of contour points contained in each change interval, and determine the change intervals with a consistent number of contour points as the target intervals, thus obtaining the target interval set; The target interval is used as the source interval of the constraint, and the direction of change of the adjacent intervals is uniformly calibrated according to the direction of change of the source interval of the constraint, so as to obtain the sequence of constraint intervals.

[0008] Furthermore, based on the set of closed segments, a unidirectional change path is established within each closed segment according to the sequence of contour points. The degree of offset of the closed segment along the attachment direction is identified, and the segment shape offset value is obtained, including: Based on the set of closed segments, calculate the degree of difference between positive and negative index changes for adjacent contour points to obtain the index deviation item; Based on the set of closed segments, identify whether the cumulative changes of each contour point inside the closed segment are balanced when unfolded along the attachment direction, and obtain the path offset term. Based on the closed segment, calculate the difference between the forward and reverse indices at both ends of the closed segment on the bidirectional index, identify the boundary consistency of the closed segment at the beginning and end positions, and obtain the beginning and end coupling term; Based on the index deviation term and path offset term, the influence of local contour changes on the path distribution in the attachment direction is identified, and the segment interaction term is obtained. Based on the head-tail coupling term, the path offset terms of each contour point in the closed section are aggregated to identify the common influence between the overall internal offset state of the closed section and the head-tail boundary state, and the boundary compensation term is obtained. By fusing the segment interaction term and the boundary compensation term, the combined effect of changes in the internal contour of the closed segment and the difference between the beginning and end boundaries on the degree of attachment direction offset is identified, and the segment shape offset value is obtained.

[0009] Furthermore, based on the set of closed segments, boundary contour points between adjacent closed segments are extracted, and boundary connection relationships between closed segments are established based on these boundary contour points, resulting in a segment boundary mapping set, including: Based on the set of closed segments, the start and end positions of each closed segment in the bidirectional fusion sequence are extracted, and the contour points at the corresponding positions are determined as segment boundary contour points to obtain the set of segment boundary points. Based on the set of boundary points of each segment, the termination position of each closed segment is sequentially matched with the starting position of its adjacent closed segment to determine the pairs of segments with adjacent positional relationships, thus obtaining a set of adjacent segment pairs. Based on the adjacent segment pairs, the contour points are aligned according to their relative positions in the bidirectional fusion sequence, and the contour points at the termination position and the starting position are made to form a correspondence, thus obtaining the boundary correspondence set. Based on the boundary correspondence set, the boundary contour points with corresponding relationships are filtered, the correspondence with the smallest position difference is retained and the rest are deleted to obtain the effective boundary relationship set. Based on the effective boundary relationship set, the correspondence between each closed segment is arranged in the order of the bidirectional fusion sequence to obtain the segment boundary mapping set.

[0010] Furthermore, based on the segment boundary mapping set, the boundary connection relationships between each closed segment are concatenated and combined to identify the degree of consistency in the connection between different closed segments, thereby obtaining the overall contour coupling value, including: Based on the segment boundary mapping set, calculate the degree of difference between the boundary contour points of the front and back closed segments in the forward and reverse index information to obtain the boundary difference term. Based on the segment boundary mapping set, calculate the proportion of the connection span between adjacent closed segments relative to the boundary position range to obtain the connection span term; Based on the segment connection path formed by the concatenation of segment boundary mapping sets, the degree of deviation of the closed segment arrangement in the connection path relative to the continuous sequential advancement is calculated to obtain the sequence perturbation term; Based on the number of closed segments in the segment connection path, the degree of matching between the path coverage status and the segment distribution range is calculated to obtain the coverage discrete term; By fusing the boundary difference term and the connection span term, a local connection term is obtained; by fusing the sequence disturbance term and the coverage discrete term, an overall connection term is obtained; by fusing the local connection term and the overall connection term, the degree of connection consistency of the closed segment on the overall connection path is identified, and the overall contour coupling value is obtained.

[0011] Furthermore, by mapping the preset attachment path to each closed segment, and correcting and adjusting the preset attachment path according to the segment shape offset value and the overall contour coupling value, the target attachment path data is obtained, including: Based on the preset attachment path and the set of closed segments, each path node in the preset attachment path is divided into the corresponding closed segments according to its position range in the bidirectional fusion sequence, thus obtaining the segment mapping path. Based on the segment mapping path and segment shape offset value, the position of the path nodes in each closed segment is adjusted along the attachment direction to ensure that the arrangement of the path nodes changes accordingly with the degree of offset within the segment, thus obtaining the adjustment path within the segment. Based on the overall contour coupling value, the path nodes between adjacent closed segments are sequentially coordinated at the connection points to ensure that the path nodes at the connection points maintain a continuous transition between different closed segments, thus obtaining the inter-segment transition path. Based on the intra-segment adjustment path and inter-segment transition path, all path nodes are rearranged in the order of the bidirectional fusion sequence, and the order of overlapping and discontinuous path nodes is corrected to obtain the reconstructed attachment path. Based on the reconstructed attachment path, the positions of the path nodes in each closed segment are unified to obtain the target attachment path data.

[0012] Secondly, a curvature matching system for automatic labeling, the system comprising: The data module is used to acquire the contour point data of the workpiece to be mounted; The fusion module is used to create index sequences for each contour point according to the forward and reverse order based on the contour point data, and to alternately insert the forward and reverse index sequences according to their positions to obtain a bidirectional fusion sequence. The constraint module is used to identify the direction of change of contour points based on the bidirectional fusion sequence and merge contour points with the same direction of change into a change interval. The change intervals with lengths that meet the standard are used as constraint source intervals to obtain a constraint interval sequence. The closure module is used to merge intervals with the same direction of change into segments based on the constraint interval sequence, and to expand the range of the segments to both sides with the constraint source interval as the center to obtain a set of closed segments; The offset module is used to establish a unidirectional change path in each closed segment according to the contour point sequence based on the closed segment set, identify the degree of offset of the closed segment along the attachment direction, and obtain the segment shape offset value. The mapping module is used to extract the boundary contour points between adjacent closed segments based on the set of closed segments, and to establish the boundary connection relationship between closed segments based on the boundary contour points, thereby obtaining the segment boundary mapping set; The coupling module is used to connect and combine the boundary connection relationships between closed segments in series according to the segment boundary mapping set, identify the degree of connection consistency between different closed segments, and obtain the overall contour coupling value. The path module is used to map a preset attachment path to each closed segment, and then correct and adjust the preset attachment path according to the segment shape offset value and the overall contour coupling value to obtain the target attachment path data.

[0013] The above-described solution of the present invention has at least the following beneficial effects: This invention identifies the changing direction of contour point indices and merges contour points with consistent directions into change intervals, completing the transformation from point-level data to interval-level data. This transforms contour data from a discrete set into an interval set with continuity and defined boundaries. This interval division process is equivalent to segmenting and encoding contour changes, compressing the original point sequence into several interval units with consistent directions, forming a structured segmented expression. This enables a segmented description of curve shape changes and forms an interval data structure with start and end indices, direction labels, and length attributes. This provides a unified data unit for subsequent segment merging and path mapping, allowing subsequent processing to be based on interval sets that meet preset data conditions. A constraint control mechanism is introduced into the data stream.

[0014] This invention merges intervals with the same direction of change into segments and expands them outward from the constraint source interval as the center, thereby elevating the set of intervals to a set of segments. This enables the contour data to form a multi-level organizational relationship, i.e., from point to interval to segment. Closed segments, as higher-level data units, contain multiple continuous intervals and have boundary information, giving the data not only local continuity but also overall closure. This step is equivalent to constructing a grouped or block-structured data model, allowing subsequent calculations to be performed at the segment granularity. The segment can be regarded as a continuous morphological fragment in the contour curve. By expanding from the constraint interval as the center, a center-driven mechanism is introduced into the segment formation process, causing the segment range to expand around a specific condition interval, forming a constraint-centric regional aggregation model in data processing.

[0015] This invention establishes a unidirectional change path within a closed segment according to the sequence of contour points, and identifies the degree of offset along the attachment direction based on the path change, generating a segment shape offset value. The contour change within the segment is expressed in numerical form. The segment shape offset value, as the result extracted from the path change, represents the relationship between the contour change within the segment and the attachment direction. It is a parameterized result obtained from geometric relationships. This step generates data variables that can be used for subsequent calculations. By establishing a unidirectional path, the data within the segment has a clear sequential direction of advancement, providing a unified benchmark for offset calculation. The change information that originally only existed in the contour geometry is transformed into numerical variables, which can participate in subsequent path correction calculations.

[0016] This invention extracts boundary contour points between adjacent closed segments and establishes boundary connection relationships between segments to obtain a segment boundary mapping set. This transforms the originally independent set of segments into a network structure with connections. Each pair of adjacent segments establishes a mapping relationship through boundary points, forming adjacency relationship data between segments. This step constructs a data model similar to a graph structure or a chain connection structure, modeling the spatial adjacency relationship between corresponding contour segments. It not only records the connection relationship between segments but also implicitly contains the segment order and boundary position relationship, enabling subsequent path continuity calculations based on this mapping.

[0017] This invention identifies the degree of consistency between different segments by serially combining boundary connection relationships, thereby obtaining an overall contour coupling value. It integrates local connection relationships into global relationships. The overall contour coupling value is a global parameter formed by summarizing the connection relationships of multiple segments. Its essence is to numerically represent the overall consistency of the segment connection path, which is reflected by the aggregation calculation of multi-node connection relationships, reflecting the continuity of the overall contour. By serially combining segment connection relationships to form a continuous path structure, and performing calculations on this path, the scattered local connection information is transformed into a single global variable. Attached Figure Description

[0018] Figure 1 This is a flowchart of an arc matching method for automatic labeling provided by an embodiment of the present invention. Detailed Implementation

[0019] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0020] like Figure 1 As shown, an embodiment of the present invention proposes a curvature matching method for automatic labeling, the method comprising: Obtain the contour point data of the workpiece to be mounted; Based on the contour point data, index sequences are established for each contour point in both forward and reverse order. The forward and reverse index sequences are then inserted alternately according to their positions to obtain a bidirectional fusion sequence. Based on the bidirectional fusion sequence, the changing direction of the contour points is identified and contour points with the same changing direction are merged into a change interval. The change intervals with lengths that meet the standard are used as the constraint source intervals to obtain the constraint interval sequence. Based on the constraint interval sequence, the intervals with the same direction of change are merged into segments, and the range of the segments is expanded to both sides with the constraint source interval as the center to obtain a closed segment set; Based on the set of closed segments, a unidirectional change path is established in each closed segment according to the contour point sequence. The degree of offset of the closed segment along the attachment direction is identified, and the segment shape offset value is obtained. Based on the set of closed segments, extract the boundary contour points between adjacent closed segments, and establish the boundary connection relationship between closed segments based on the boundary contour points to obtain the segment boundary mapping set; Based on the segment boundary mapping set, the boundary connection relationships between each closed segment are connected in series and combined to identify the degree of connection consistency between different closed segments and obtain the overall contour coupling value. By mapping the preset attachment path to each closed segment, the preset attachment path is corrected and adjusted according to the segment shape offset value and the overall contour coupling value to obtain the target attachment path data.

[0021] In this embodiment of the invention, contour point data of the workpiece to be mounted is acquired to realize the mapping from the physical workpiece surface to digital contour data, providing a unified input format for subsequent processing. Based on the contour point data, each contour point is indexed in both forward and reverse order, and the forward and reverse index sequences are alternately inserted according to their positions to obtain a bidirectional fusion sequence. This ensures that each contour point has both forward and reverse order attributes in the data structure, so that subsequent change judgments do not depend on a single traversal order. Based on the bidirectional fusion sequence, the change direction of the contour points is identified, and contour points with the same change direction are merged into change intervals. Change intervals with lengths that meet the standard are used as constraint source intervals to obtain a constraint interval sequence, realizing the structural transformation of data from point level to interval level. Based on the constraint interval sequence, change intervals with the same change direction are merged into segments, and the segment range is expanded to both sides with the constraint source interval as the center to obtain a closed segment set, realizing the hierarchical organization of data and providing basic data blocks for subsequent calculations.

[0022] Based on the set of closed segments, a unidirectional change path is established within each closed segment according to the contour point sequence. The degree of offset of the closed segment along the attachment direction is identified, and the segment shape offset value is obtained. The geometric changes within the segment are converted into numerical features, realizing the mapping from structural data to feature data. Based on the set of closed segments, the boundary contour points between adjacent closed segments are extracted, and the boundary connection relationship between closed segments is established based on the boundary contour points, resulting in a segment boundary mapping set. This makes the segment set form a structure with adjacency relationships, providing a connection basis for subsequent overall analysis. Based on the segment boundary mapping set, the boundary connection relationships between each closed segment are connected in series and combined. The degree of connection consistency between different closed segments is identified, and the overall contour coupling value is obtained. The local connection relationship is integrated into global feature data, realizing the quantitative expression of the overall contour relationship. By mapping the preset attachment path to each closed segment, the preset attachment path is corrected and adjusted according to the segment shape offset value and the overall contour coupling value to obtain the target attachment path data. This realizes the fusion processing of path data and contour data, completing the data conversion from input path to output path.

[0023] Specifically, obtaining the contour point data of the workpiece to be mounted includes: The contour acquisition module, installed in the automatic labeling equipment, scans and acquires data from the outer surface of the workpiece to be labeled. This module can be a vision measurement unit comprised of a structured light scanning device, a laser contour sensor, or an industrial camera combined with image processing algorithms. During acquisition, the system continuously scans along the workpiece surface along a preset sampling path to obtain raw point cloud data reflecting the surface geometry. This raw point cloud data is then input into a data processing unit for preprocessing, including filtering out abnormal discrete points, downsampling dense areas, and interpolating sparse areas to form a contour point set. Further, the contour points are sorted according to the scanning path or time sequence, and each contour point is assigned a unique sequential number. Simultaneously, the corresponding spatial coordinates are associated with the number, generating a basic contour point data table. The contour points are then transformed to uniformly map to the equipment reference coordinate system or the labeling coordinate system, ensuring consistency in subsequent path calculations and obtaining the contour point data.

[0024] Specifically, based on the constraint interval sequence, intervals with the same direction of change are merged into segments, and the range of these segments is expanded outwards from the constraint source interval to obtain a set of closed segments, which includes: The sequence is traversed through each change interval. Based on the change direction identifier corresponding to each change interval, adjacent change intervals with the same direction are initially merged, integrating consecutive change intervals with the same change direction into an initial segment structure. During this process, the start and end indexes of each initial segment are recorded, and a mapping relationship is established between the segment and the change intervals it contains. Then, constraint source intervals that meet the length standard are selected sequentially from the constraint interval sequence. Using this constraint source interval as the center of segment expansion, adjacent change intervals before and after it in the sequence are expanded. Starting from the start position of the constraint source interval, adjacent change intervals are traversed backward, determining whether their change direction is consistent with the constraint source interval and whether they meet the continuity condition. If they do, the change interval is merged into the current segment, and the segment's start boundary is updated. Similarly, starting from the end position of the constraint source interval, the expansion is performed backward, incorporating change intervals that meet the conditions into the segment range, and the segment's end boundary is updated. When encountering change intervals with inconsistent change directions or that do not meet the continuity condition, expansion in that direction is stopped, and the final boundary range of the segment is determined. After constructing a segment centered on a single constraint source interval, the same expansion operation is performed on the next constraint source interval until all constraint source intervals have completed segment generation. For segments with overlapping or intersecting ranges, they are merged or deduplicated according to the segment boundary position and inclusion relationship to ensure that each segment in the final segment set has a clear boundary range and does not overlap with each other. All constructed segments are arranged according to their position order in the bidirectional fusion sequence to form a closed segment set.

[0025] In a preferred embodiment of the present invention, based on the contour point data, an index sequence is established for each contour point in a forward and reverse order, and the forward and reverse index sequences are alternately inserted according to their positions to obtain a bidirectional fusion sequence, including: Based on the contour point data, each contour point is sequentially numbered according to the collection order of the contour points, and arranged in ascending order to form a forward point column, and in descending order to form a reverse point column. Based on the forward point column and the reverse point column, the contour points in the forward point column and the reverse point column are labeled with the number of each contour point as the index identifier, so as to obtain the forward index sequence and the reverse index sequence. By alternating the contour points at the same position in the forward index sequence and the reverse index sequence, ensuring that each position contains both forward index information and reverse index information, an alternating fusion sequence is obtained. Based on the alternating fusion sequence, the forward and reverse index information of each contour point is combined and identified, and the sequence position is recalibrated according to the alternating arrangement order to obtain the bidirectional fusion sequence.

[0026] In this embodiment of the invention, based on the contour point data, each contour point is sequentially numbered according to the acquisition order, and arranged in ascending order to form a forward point column, and in descending order to form a reverse point column. This allows the same set of contour information to have both forward and reverse traversal sequence representations, avoiding mismatch problems caused by different data sources. Based on the forward and reverse point columns, the contour points in the forward and reverse point columns are labeled using the number of each contour point as an index identifier, resulting in a forward index sequence and a reverse index sequence. This unifies the identification system in the forward and reverse sequence representations, allowing the same contour point to share the same number in different directional sequences. Based on the foundation, by alternating the contour points at the same sequential positions in the forward and reverse index sequences, it is ensured that each position contains both forward and reverse index information, resulting in an alternating fusion sequence. This provides a data foundation for the joint identification of bidirectional information within the same position group, eliminating the need for frequent cross-table lookups during sequence processing. Based on the alternating fusion sequence, the forward and reverse index information of each contour point is combined and identified, and the sequence position is recalibrated according to the alternating arrangement order, resulting in a bidirectional fusion sequence. This allows subsequent operations to be carried out based on a unified fusion sequence position system, avoiding the data switching complexity caused by the coexistence of different sequence position systems.

[0027] Specifically, by alternating the contour points at the same sequential positions in the forward and reverse index sequences, ensuring that each position simultaneously contains both forward and reverse index information, an alternating fusion sequence is obtained, which includes: Read the data items of each contour point in the forward and reverse index sequences, check whether the two sequences originate from the same contour point data set, and verify whether the total number of contour points in the two sequences is consistent and whether the sets of contour point numbers correspond. If a local quantity difference is detected due to previous data cleaning, outlier deletion, or point supplementation, alignment processing is performed first. This involves inserting placeholder data at missing positions, truncating redundant data that exceeds the position, or rearranging locally misaligned data items to matching positions according to the correspondence of contour point numbers, ensuring that both the forward and reverse index sequences have data items that can participate in the arrangement at the same sequential position. Perform an alternating write operation on this set of data, with a pre-defined unified write rule, such as writing the forward index data item at the current sequential position first, and then writing the reverse index data item at the current sequential position; or writing the reverse index data item first, and then writing the forward index data item according to the opposite rule. Regardless of the rule used, the same write order is maintained throughout the sequence fusion process, so that the final alternating fused sequence has a definite arrangement logic. During the writing process, not only is the index value of the contour point written, but also the original number corresponding to that contour point, the contour point coordinates, the source sequence direction identifier, and the position number of the contour point in the original index sequence are written simultaneously, ensuring that each written record completely retains subsequent traceable information. Subsequently, the forward index data item and the reverse index data item at the next sequential position are read, and the aforementioned alternating writing process is repeated until all sequential positions in both sets of index sequences have been processed, forming an alternating fused sequence.

[0028] Specifically, based on the alternating fusion sequence, the forward and reverse index information of each contour point is combined and identified, and the sequence position is re-marked according to the alternating arrangement order to obtain the bidirectional fusion sequence, which includes: According to the aforementioned position group number or the correspondence between adjacent records in the alternating arrangement rule, the data items in the alternating fusion sequence are read group by group to recover the forward index data items and reverse index data items generated by the same sequential position in each group. For each group of data, its forward index value, reverse index value, original contour point number, contour point coordinates, and source direction identifier are extracted, and a combined identifier generation operation is performed on this basis. The combined identifier can be generated by using a composite field encapsulation method, that is, writing the forward index value and reverse index value into different fields of the same record to form a two-field structure of forward index field and reverse index field. The data organization method in the alternating fusion sequence is reconstructed, and the forward index information and reverse index information belonging to the same position group are recombined from the original two adjacent alternating records into a composite record. This composite record includes at least the following fields: the original number of the contour point, the coordinate information of the contour point, the forward index value, the reverse index value, the original forward position, the original reverse position, and the source relationship identifier within the group. If the forward and reverse index data items in a position group are associated with the same contour point, the system directly merges them into the same data object. If two data items in a position group are defined as a set of sequential position reference pairs in the implementation method, the system can write the entire position group as a fusion unit into a new sequence, instead of retaining two separate records. Subsequently, the reassembled composite records are renumbered. This re-marking of the sequence position means that the position numbers in the forward index sequence and the reverse index sequence are no longer used. Instead, the output order after fusion is used as the sole criterion, and new consecutive position numbers are assigned to each composite record starting from the beginning of the sequence. For example, all fusion units can be renumbered using consecutive integers from 1 to M, where M is the total number of composite records after fusion. This new position number is written into the fusion position field of each composite record to identify its actual sequential position in the bidirectional fusion sequence. Each composite record is checked to ensure that both the forward and reverse index fields have valid values. The fusion position numbers are checked to ensure they are consecutive and that there are no duplicate or skipped numbers. The combined identifier at any fusion position is verified to uniquely correspond to a set of forward and reverse index information. If a composite record is found to be missing one of the directional index values ​​or its content is inconsistent with the original alternating fusion sequence, the alternating fusion sequence is backtracked according to the original position group number to extract data again and repair the corresponding record, thus obtaining a bidirectional fusion sequence.

[0029] In a preferred embodiment of the present invention, based on the bidirectional fusion sequence, the changing direction of contour points is identified and contour points with the same changing direction are merged into a change interval. Change intervals with lengths conforming to a standard are designated as constraint source intervals, resulting in a constraint interval sequence, including: Based on the bidirectional fusion sequence, the direction of change between contour points is determined by comparing the increase or decrease of the index values ​​of adjacent contour points, thus obtaining the direction identification sequence. Based on the direction identification sequence, the contour points with continuous and consistent change directions are merged according to their arrangement order in the bidirectional fusion sequence, and the start and end positions are recorded to obtain the change interval set; Based on the set of change intervals, count the number of contour points contained in each change interval, and determine the change intervals with a consistent number of contour points as the target intervals, thus obtaining the target interval set; The target interval is used as the source interval of the constraint, and the direction of change of the adjacent intervals is uniformly calibrated according to the direction of change of the source interval of the constraint, so as to obtain the sequence of constraint intervals.

[0030] In this embodiment of the invention, based on the bidirectional fusion sequence, the change direction between contour points is determined by comparing the increase or decrease of the index values ​​of adjacent contour points, resulting in a direction identifier sequence. This ensures that the adjacent contour points in the sequence are no longer simply arranged sequentially, but have a clear semantic change direction, providing a direct basis for subsequent interval merging. Based on the direction identifier sequence, contour points with continuous and consistent change directions are merged according to their arrangement order in the bidirectional fusion sequence, and the start and end positions are recorded to obtain a set of change intervals. This provides a higher-level data unit for subsequent interval length statistics and constraint source interval selection. Based on the set of change intervals, the number of contour points contained in each change interval is counted, and the change intervals with consistent contour point numbers are determined as target intervals, resulting in a target interval set. This allows subsequent processing to be based on a set of intervals with relatively consistent length attributes, providing a direct basis for determining the target intervals as constraint source intervals in the next step. The target intervals are used as constraint source intervals, and the change directions of their adjacent change intervals are uniformly labeled according to the change direction of the constraint source intervals, resulting in a constraint interval sequence. This transforms the direction identifiers in the change interval set from a locally independent state into an ordered sequence structure with constraint propagation relationships, providing a basis for subsequently merging change intervals into segments.

[0031] Specifically, based on the bidirectional fusion sequence, the direction of change between contour points is determined by comparing the increase or decrease of the index values ​​of adjacent contour points, resulting in a direction identification sequence, which includes: The system extracts data from the current contour point and the next contour point sequentially, starting from the first record, according to the fusion position order in the bidirectional fusion sequence, forming adjacent contour point pairs. For each pair, the system compares the changes in both the forward and reverse index values. When the forward index value between the current and next contour points increases, decreases, or remains unchanged, the system records the forward index change state; similarly, the reverse index value undergoes the same type of change state identification. Subsequently, the system comprehensively interprets the forward and reverse index change states according to preset direction determination rules, generating a change direction identifier corresponding to the adjacent contour point pair. For example, when the forward index increases and the reverse index decreases, it can be identified as a continuous advancement state in a preset direction; when the change pattern of the forward and reverse indices is opposite to the aforementioned state, it can be identified as a continuous advancement state in another preset direction; when the change relationship between the two does not meet the continuous advancement condition, or when there are index jumps, local repetitions, or abnormal increases or decreases, it can be marked as a direction interruption state, an invalid state, or a state pending determination. The system repeats the above process for all adjacent contour point pairs in the bidirectional fusion sequence until the end of the sequence. For the last contour point, since there are no successor contour points, the direction identifier can be completed using boundary inheritance, null value filling, or termination marker methods. Subsequently, the system writes the direction judgment results corresponding to each pair of adjacent contour points into the direction identifier sequence in their order in the bidirectional fusion sequence, so that the direction identifier sequence and the bidirectional fusion sequence maintain a corresponding positional relationship. The order relationship that originally existed in the form of index combination in the bidirectional fusion sequence is transformed into a direction-type sequence result that can be used for subsequent segmentation processing.

[0032] Specifically, based on the set of variation intervals, the number of contour points contained in each variation interval is counted, and the variation intervals with a consistent number of contour points are determined as the target intervals, thus obtaining the target interval set, which includes: The system reads each interval record in the set of changing intervals one by one, and calculates the actual number of contour points contained in the changing interval based on the difference between its start and end positions, or based on the list of contour points already saved within the interval. The number of contour points can be written into the interval attribute field as the length parameter of the changing interval, forming a correspondence table between interval numbers and length parameters. The system summarizes and analyzes all interval length parameters to identify length patterns that recur in the set of changing intervals or length types that meet preset standards. For example, it can use frequency statistics to count the number of times each length value appears in all changing intervals; it can also filter changing intervals whose lengths fall within a preset standard length range; or it can group changing intervals with identical lengths or length differences within the allowable error range into the same candidate category according to the principle of length consistency. When determining the target interval, the system filters the changing intervals according to preset filtering rules. If maintaining a consistent number of points is used as the target criterion, the system selects a group of intervals with the same number of contour points from all changing intervals, or selects the group of intervals with the most concentrated number distribution, as the target interval. If there are slight deviations in the interval length, intervals with similar lengths can be uniformly regarded as the same length category by rounding, categorizing, or merging threshold tolerances. Intervals that do not meet the consistency requirements are temporarily retained in the changing interval set but not included in the current target interval set. The selected target intervals are reorganized according to their order in the bidirectional fusion sequence and saved as the target interval set.

[0033] In a preferred embodiment of the present invention, based on the set of closed segments, a unidirectional change path is established within each closed segment according to the contour point sequence, the degree of offset of the closed segment along the attachment direction is identified, and the segment shape offset value is obtained, including: Based on the set of closed segments, calculate the degree of difference between positive and negative index changes for adjacent contour points to obtain the index deviation item; Based on the set of closed segments, identify whether the cumulative changes of each contour point inside the closed segment are balanced when unfolded along the attachment direction, and obtain the path offset term. Based on the closed segment, calculate the difference between the forward and reverse indices at both ends of the closed segment on the bidirectional index, identify the boundary consistency of the closed segment at the beginning and end positions, and obtain the beginning and end coupling term; Based on the index deviation term and path offset term, the influence of local contour changes on the path distribution in the attachment direction is identified, and the segment interaction term is obtained. Based on the head-tail coupling term, the path offset terms of each contour point in the closed section are aggregated to identify the common influence between the overall internal offset state of the closed section and the head-tail boundary state, and the boundary compensation term is obtained. By fusing the segment interaction term and the boundary compensation term, the combined effect of changes in the internal contour of the closed segment and the difference between the beginning and end boundaries on the degree of attachment direction offset is identified, and the segment shape offset value is obtained.

[0034] In this embodiment of the invention, based on the set of closed segments, the degree of difference between the forward and reverse index changes of adjacent contour points is calculated to obtain an index deviation term, reflecting the consistency of the contours within the segment during bidirectional traversal; based on the set of closed segments, whether the cumulative changes of each contour point within the closed segment are balanced when unfolded along the attachment direction is identified to obtain a path offset term, describing the distribution state of path changes; based on the closed segments, the difference between the forward and reverse indexes at both ends of the closed segment on the bidirectional index is calculated to identify the boundary consistency of the closed segment at the beginning and end positions, obtaining a beginning-end coupling term, describing the consistency or deviation of the beginning and end of the segment in the sequence structure; based on the index deviation term and the path offset term, the degree of difference between the forward and reverse indexes at both ends of the closed segment is identified to obtain ... path offset term, describing the consistency or deviation of the beginning and end of the segment in the sequence structure; based on the index deviation term and the path offset term, the degree of difference between the forward and reverse indexes at both ends of the closed segment is identified to obtain a path offset term, describing the consistency or deviation of the beginning and end of the segment in the sequence structure; based on the index deviation term and the path offset term, the degree of difference between the forward and reverse indexes at both ends of the closed segment is identified to obtain a path offset term, describing the consistency or deviation of the beginning and end of the closed segment in the sequence structure; based on the index deviation term and the path offset term, the degree of difference between the forward and reverse indexes at both ends of the closed segment is identified to obtain a path offset term, describing the consistency or deviation of the beginning and end of the closed segment in the sequence structure; based on the index deviation term and the path offset term, the degree of difference between the forward and reverse indexes at both ends of the closed segment is identified to obtain a path offset term, describing the consistency or deviation of the beginning and The influence of local contour changes on the path distribution in the attachment direction is identified, resulting in segment interaction terms, which enable joint modeling of contour changes and path relationships. Based on the head-tail coupling terms, the path offset terms of each contour point within the closed segment are aggregated to identify the combined influence between the overall internal offset state of the closed segment and the head-tail boundary states, resulting in boundary compensation terms, which establish a correlation between the internal feature data of the segment and the boundary feature data. The segment interaction terms and boundary compensation terms are fused to identify the combined effect of contour changes within the closed segment and head-tail boundary differences on the degree of offset in the attachment direction, resulting in segment morphological offset values. This provides data input for subsequent path adjustments, realizing the conversion from multi-source features to unified parameters.

[0035] Specifically, based on the set of closed segments, the degree of difference between positive and negative index changes of adjacent contour points is calculated to obtain the index deviation term; based on the set of closed segments, the cumulative change of each contour point within a closed segment as it unfolds along the attachment direction is identified to obtain the path offset term; based on the closed segment, the difference between the positive and negative indices at both ends of the closed segment on the bidirectional index is calculated to identify the boundary consistency of the closed segment at the beginning and end positions, resulting in the beginning-end coupling term; based on the index deviation term and the path offset term, the influence of local contour changes on the path distribution in the attachment direction is identified to obtain the segment interaction term; based on the beginning-end coupling term, the path offset terms of each contour point within the closed segment are aggregated to identify the common influence between the overall internal offset state of the closed segment and the beginning-end boundary state, resulting in the boundary compensation term; the segment interaction term and the boundary compensation term are fused to identify the combined effect of internal contour changes and beginning-end boundary differences on the degree of offset in the attachment direction, resulting in the segment shape offset value, specifically including: For any closed segment, the system extracts all contour point data within that segment according to the fusion position order in the bidirectional fusion sequence, and constructs adjacent contour point pairs sequentially. For each pair of adjacent contour points, its forward and reverse index values ​​are read, and the changes in the forward and reverse indices are calculated. The difference between the two changes is then calculated, for example, by taking the difference, absolute difference, or normalized deviation, to obtain the change difference value of the adjacent contour points under the bidirectional indexing system. Subsequently, the difference values ​​of all adjacent contour points within the segment are accumulated or statistically processed according to their order in the segment. Summation, averaging, or weighted accumulation methods can be used to obtain the index deviation item, reflecting the degree of overall index change difference of the closed segment.

[0036] Based on a preset attachment direction, the spatial coordinates of each contour point within the segment are projected onto that direction, forming a path sequence arranged along the attachment direction. For this path sequence, the system calculates the displacement of adjacent contour points along the attachment direction point by point, and accumulates these displacements sequentially to obtain cumulative path change data within the segment. Subsequently, a balance analysis is performed on this cumulative change data, such as calculating the dispersion and gradient of displacements between segments, identifying whether there are localized concentrations or uneven distributions of change during path unfolding, and generating a path offset term to characterize the balance of path change within the closed segment.

[0037] The system extracts the first and last contour points of the closed segment, reads their forward and reverse index values ​​respectively, and calculates the difference between the first and last points under the forward indexing system and the difference under the reverse indexing system. Then, it compares and analyzes these two differences, such as calculating the degree of difference or determining whether their trends are consistent, identifying the consistency of the first and last boundaries of the closed segment under the bidirectional indexing system, and generating a first-and-last coupling term to reflect the index correspondence between the first and last points of the segment.

[0038] The system combines index deviation and path offset items according to preset data fusion rules. For example, it integrates the two data items into a single segment interaction item through weighted summation, proportional mapping, or function transformation. The weights of each item are set according to the data distribution within the segment or preset parameters, so that the differences in index changes and the distribution of path changes have corresponding effects in the calculation results. This segment interaction item is used to characterize the relationship between the changes in the contour within the closed segment and the changes in the attached path.

[0039] The system aggregates path offset data within a segment based on the beginning-end coupling term. It summarizes and calculates the path displacement or path offset data corresponding to each contour point within the segment, such as obtaining a sum, average, or weighted result, and then corrects the summarized result based on the beginning-end coupling term. For example, it adjusts the path offset summary result proportionally or compensates for the offset based on the value of the beginning-end coupling term, thus establishing a correlation between the path change data within the segment and the segment boundary state to obtain a boundary compensation term, which can simultaneously reflect the path distribution within the segment and the consistency of the beginning and end boundaries.

[0040] The system combines segment interaction items and boundary compensation items according to preset fusion rules, such as weighted superposition, normalization processing, or polynomial mapping, to integrate the internal variation features and boundary constraint features of the segment into a single output result. The weights are adjusted according to the feature parameters of different segments so that the fusion result contains both the internal contour change information and the boundary difference information of the segment. The fusion result is used as the segment shape offset value of the closed segment, and the offset value is associated with the corresponding closed segment.

[0041] In a preferred embodiment of the present invention, based on the set of closed segments, boundary contour points between adjacent closed segments are extracted, and boundary connection relationships between closed segments are established based on the boundary contour points to obtain a segment boundary mapping set, including: Based on the set of closed segments, the start and end positions of each closed segment in the bidirectional fusion sequence are extracted, and the contour points at the corresponding positions are determined as segment boundary contour points to obtain the set of segment boundary points. Based on the set of boundary points of each segment, the termination position of each closed segment is sequentially matched with the starting position of its adjacent closed segment to determine the pairs of segments with adjacent positional relationships, thus obtaining a set of adjacent segment pairs. Based on the adjacent segment pairs, the contour points are aligned according to their relative positions in the bidirectional fusion sequence, and the contour points at the termination position and the starting position are made to form a correspondence, thus obtaining the boundary correspondence set. Based on the boundary correspondence set, the boundary contour points with corresponding relationships are filtered, the correspondence with the smallest position difference is retained and the rest are deleted to obtain the effective boundary relationship set. Based on the effective boundary relationship set, the correspondence between each closed segment is arranged in the order of the bidirectional fusion sequence to obtain the segment boundary mapping set.

[0042] In this embodiment of the invention, based on the set of closed segments, the start and end positions of each closed segment in the bidirectional fusion sequence are extracted, and the contour points at the corresponding positions are determined as segment boundary contour points, resulting in a set of segment boundary points. This ensures that each segment has a clear boundary point data representation, providing basic data for subsequent segment-to-segment relationship matching. Based on the set of segment boundary points, the end position of each closed segment is sequentially matched with the start position of its adjacent closed segments to determine segments with adjacent positional relationships, resulting in a set of adjacent segment pairs. This establishes a sequential connection relationship between the originally independent sets of segments, providing a foundation for subsequent segment-level connections. Based on the set of adjacent segment pairs, the contour points in the bidirectional fusion sequence are... Aligning the positions establishes a correspondence between the contour points of the ending and starting positions, resulting in a boundary correspondence set. This creates a more refined data association structure, providing a foundation for subsequent filtering and mapping. Based on the boundary correspondence set, the boundary contour points with corresponding relationships are filtered, retaining the correspondence with the smallest position difference and deleting the rest, resulting in a valid boundary correspondence set. This eliminates redundancy and conflicts in the boundary correspondences, ensuring the uniqueness and stability of the boundary mapping relationships between segments. Based on the valid boundary correspondence set, the correspondences between each closed segment are arranged according to the bidirectional fusion sequence, resulting in a segment boundary mapping set, thus forming an ordered mapping structure for the connection relationships between segments.

[0043] Specifically, based on the adjacent segment pairs, the contour points are aligned according to their relative positions in the bidirectional fusion sequence, establishing a correspondence between the contour points at the termination position and the starting position, thus obtaining a boundary correspondence set, which specifically includes: For each pair of adjacent segments, the system extracts the termination boundary contour points of the preceding segment and the starting boundary contour points of the following segment. It then retrieves the fusion position numbers, corresponding forward and reverse index values, and segment numbers of these two types of boundary contour points in the bidirectional fusion sequence. The termination boundary contour point of the preceding segment is used as the forward boundary reference point, and the starting boundary contour point of the following segment is used as the backward boundary candidate point. Using the positional order in the bidirectional fusion sequence as a unified reference, the system first determines whether the sequential arrangement of the forward boundary reference point and the backward boundary candidate point in the sequence satisfies the connection order requirements of adjacent segments. When the termination boundary contour point is in front and the starting boundary contour point is in back, and there is no abnormal jump across unrelated segments, this group of boundary points is included in the alignment range, forming a boundary candidate matching unit. The system then uses the position of the terminating boundary contour point in the bidirectional fusion sequence as a benchmark to calculate the difference between the position numbers of the starting and ending boundary contour points, and determines the relative distance between them based on this difference. Simultaneously, it considers the changing directions of the two points on the forward and reverse indices to determine if the position difference aligns with the current segment connection direction. If the position difference in the fusion sequence is within a preset allowable range, and the changes in the forward and reverse indices satisfy the adjacent advancement relationship, then the two points are identified as a set of boundary contour points satisfying the alignment condition, and this set of boundary points is confirmed as a valid candidate corresponding point pair. The system writes the matching result between the terminating and starting boundary contour points into the boundary correspondence record. If multiple candidate boundary points satisfying the alignment condition exist for the same adjacent segment pair, the system continues to retain the correspondence between these candidate points and writes them all into the candidate correspondence result. The system repeats the above process for each pair of adjacent segment pairs in the adjacent segment pair set to form a boundary correspondence set.

[0044] Specifically, based on the boundary correspondence set, boundary contour points with corresponding relationships are filtered, retaining the correspondence with the smallest positional difference and deleting the remaining correspondences to obtain the effective boundary relationship set, which includes: The system can group multiple candidate starting boundary contour point correspondences belonging to the same previous segment's termination boundary contour point into the same filtering group, or group all candidate correspondences belonging to the same adjacent segment pair into the same filtering group. It reads the termination boundary contour point position, the starting boundary contour point position, and the position difference between them recorded in each candidate correspondence group, using the position difference as a priority filtering parameter. For multiple candidate correspondences within the same filtering group, it compares the position differences of each correspondence, determining the candidate relationship with the smallest position difference as the priority retention object. When two or more candidate relationships have the same or similar position differences, the system further calls auxiliary judgment parameters such as the forward index difference and reverse index difference recorded in each candidate relationship to perform a secondary comparison on these candidate relationships with the same difference. The system writes this candidate relationship into the valid boundary relationship result set and marks the remaining unselected candidate correspondences in the same filtering group as redundant relationships. These redundant relationships are then deleted, either removed from the boundary correspondence set or their status field is modified to invalid. The system checks whether a certain terminating boundary contour point has already established a valid correspondence with a certain starting boundary contour point in the previous screening process. If it has been established, then even if the position difference of other candidate relationships involving the terminating boundary contour point is small, it is necessary to further determine whether it will cause a one-to-many mapping conflict. If a certain starting boundary contour point has already formed a unique correspondence with the previous terminating boundary contour point, then other candidate correspondences related to the starting boundary contour point will no longer be retained, so that each boundary contour point retains only a unique corresponding object in the set of valid boundary relationships.

[0045] The system performs cross-conflict checks on the retained valid candidate relationships. Following the order in the bidirectional fusion sequence, it sequentially traverses each valid candidate relationship, checking for the following situations: the terminating boundary contour point of the previous valid relationship is located after the terminating boundary contour point of the subsequent relationship, or the starting boundary contour point of the previous valid relationship is located after the starting boundary contour point of the subsequent relationship; i.e., a mapping order intersection occurs. If such a cross-mapping relationship is found, it indicates that although the positional difference of a single candidate relationship is small, its overall order does not meet the continuity requirement of segment boundary mapping. In response, the system re-compares the total positional difference, index difference degree, and segment order consistency among two or more valid candidate relationships with intersections, retaining the correspondence that better conforms to the overall order continuity and deleting the remaining relationships causing the intersection. The system also checks each valid boundary relationship to see if the terminating boundary contour point and the starting boundary contour point both belong to their recorded adjacent segment pairs, whether their positional difference is still within the allowable range, and whether, after deleting redundant relationships, some adjacent segment pairs have no valid boundary relationships available. If a pair of adjacent segments fails to retain any valid relationship after screening, the system can backtrack to the corresponding screening group and reselect the relationship with the second smallest position difference that does not cause conflict from the deleted candidate relationships as a supplementary relationship. Through the above processing, the multiple candidate relationships in the boundary relationship set are compressed and regularized into a valid boundary relationship set.

[0046] In a preferred embodiment of the present invention, based on the segment boundary mapping set, the boundary connection relationships between each closed segment are concatenated and combined to identify the degree of consistency in the connection between different closed segments, thereby obtaining the overall contour coupling value, including: Based on the segment boundary mapping set, calculate the degree of difference between the boundary contour points of the front and back closed segments in the forward and reverse index information to obtain the boundary difference term. Based on the segment boundary mapping set, calculate the proportion of the connection span between adjacent closed segments relative to the boundary position range to obtain the connection span term; Based on the segment connection path formed by the concatenation of segment boundary mapping sets, the degree of deviation of the closed segment arrangement in the connection path relative to the continuous sequential advancement is calculated to obtain the sequence perturbation term; Based on the number of closed segments in the segment connection path, the degree of matching between the path coverage status and the segment distribution range is calculated to obtain the coverage discrete term; By fusing the boundary difference term and the connection span term, a local connection term is obtained; by fusing the sequence disturbance term and the coverage discrete term, an overall connection term is obtained; by fusing the local connection term and the overall connection term, the degree of connection consistency of the closed segment on the overall connection path is identified, and the overall contour coupling value is obtained.

[0047] In this embodiment of the invention, based on the segment boundary mapping set, the degree of difference between the boundary contour points of the preceding and following closed segments in terms of forward and reverse index information is calculated to obtain a boundary difference term. This maps the connection relationship originally based on sequence position into quantifiable difference data, providing basic data for subsequent analysis. Based on the segment boundary mapping set, the proportion of the connection span between adjacent closed segments relative to the boundary position range is calculated to obtain a connection span term. This provides a unified quantitative standard for the span relationship of different segment connections in the overall sequence, providing comparable data for subsequent overall connection structure analysis. Based on the segment connection path formed by the concatenation of the segment boundary mapping set, the degree of deviation of the arrangement of closed segments in the connection path relative to the continuous sequential advancement is calculated to obtain a sequence disturbance term. This allows the sequence offset or jump phenomenon in the segment connection process to be quantitatively described, ensuring overall connection consistency. The system provides data for performance evaluation; based on the number of closed segments in the segment connection path, it calculates the matching degree between the path coverage status and the segment distribution range, obtaining a coverage discrete term, enabling the path coverage status to be expressed in numerical form, and providing distribution dimension data for subsequent overall connection relationship analysis; it fuses the boundary difference term and the connection span term to obtain a local connection term, representing the comprehensive state of the local connection relationship between adjacent segments; it fuses the sequence disturbance term and the coverage discrete term to obtain an overall connection term, describing the sequence consistency and distribution status of the segment connection path in the global scope; it fuses the local connection term and the overall connection term to identify the degree of connection consistency of closed segments on the overall connection path, obtaining the overall contour coupling value, transforming the connection relationship between segments from multi-dimensional data into a unified numerical expression, and providing global input data for subsequent attachment path correction.

[0048] Specifically, based on the segment boundary mapping set, the degree of difference between the boundary contour points of the preceding and following closed segments in terms of forward and reverse index information is calculated to obtain a boundary difference term; based on the segment boundary mapping set, the proportion of the connection span between adjacent closed segments relative to the boundary position range is calculated to obtain a connection span term; based on the segment connection path formed by the concatenation of segment boundary mapping sets, the degree of deviation of the arrangement of closed segments in the connection path relative to the continuous sequential advancement is calculated to obtain a sequence perturbation term; based on the number of closed segments in the segment connection path, the degree of matching between the path coverage state and the segment distribution range is calculated to obtain a coverage discrete term; the boundary difference term and the connection span term are fused to obtain a local connection term; the sequence perturbation term and the coverage discrete term are fused to obtain an overall connection term; the local connection term and the overall connection term are fused to identify the degree of connection consistency of closed segments on the overall connection path to obtain an overall contour coupling value, specifically including: Read the records of each valid boundary connection relationship in the segment boundary mapping set, and extract the corresponding termination boundary contour point of the previous segment and the starting boundary contour point of the next segment for each connection relationship. Read the forward index value and reverse index value of these two boundary contour points in the bidirectional fusion sequence, and calculate the difference between the forward and reverse index values ​​of the previous and next boundary points. Perform combined operations on the forward and reverse differences, such as obtaining the difference amount, ratio, or weighted deviation, to obtain a single difference result that reflects the degree of inconsistency in the change of the boundary connection relationship under the bidirectional index system. Perform the above calculation on all connections in the segment boundary mapping set one by one, and accumulate or statistically analyze each difference result according to the connection order, such as summing, averaging, or normalizing to form a unified value, thus forming a boundary difference item.

[0049] The system reads the position numbers of the termination boundary contour point and the start boundary contour point in the bidirectional fusion sequence of the connection relationship, and calculates the distance between them as the connection span. The system obtains the segment boundary position range corresponding to the connection relationship, such as the sequence interval length between the termination position of the previous segment and the start position of the next segment, or the boundary span of the segment relative to the overall range. Then, the system calculates the ratio of the connection span to the corresponding boundary position range to obtain the span proportion of the connection relationship. The system repeats the above operation for all boundary connections and summarizes the span proportions, for example, by taking the average or weighting by segment weights, to form the connection span item.

[0050] The system sequentially connects closed segments according to the segment connection order recorded in the mapping set, forming a complete segment connection path sequence. This connection path is compared with the natural order in the bidirectional fusion sequence. The order in which segments in the bidirectional fusion sequence are arranged in ascending or descending order of position is used as a reference order, and the segment arrangement in the actual connection path is compared one by one. By calculating the offset of each segment's position in the actual path relative to the reference order—such as segment jump distance, number of order reversals, or deviation position difference—deviation data describing the connection path's deviation from continuous advancement is obtained. The deviation data of all segments are summarized to generate a sequence perturbation term, thus numerically representing the deviation of the segment connection path at the sequence level.

[0051] The system counts the number of closed segments contained in the path and determines the overall distribution range of these segments in the bidirectional fusion sequence, i.e., obtaining the span between the minimum and maximum position numbers in the path. Then, it performs a matching analysis between the number of segments and this distribution range, such as calculating the segment density per unit area, the distribution of segment spacing, or the uniformity of segment arrangement. Further, it statistically analyzes the intervals between segments, such as calculating the variance, range, or average spacing of each interval, to determine whether there is a concentration or dispersion in the segment distribution. The above statistical results are then summarized to obtain the coverage discrete term.

[0052] The system combines boundary difference terms and connection span terms according to preset fusion rules, such as through weighted summation or normalization mapping, to integrate the two data points into a single local connectivity term, which represents the comprehensive state of adjacent segments at the local connectivity level. Subsequently, the system performs similar fusion processing on the order disturbance term and coverage discrete term, integrating the two data points into an overall connectivity term, which describes the order consistency and distribution state of segment connection paths at the overall structural level. Based on preset weights or mapping functions, the system combines the local connectivity term and the overall connectivity term, so that local connectivity features and overall path features are reflected in the same data result, obtaining an overall contour coupling value. This value is then associated with and stored in relation to the corresponding segment connection path and segment set, realizing a step-by-step calculation from segment boundary connection relationships to the overall contour coupling value, thus integrating local connectivity features and overall structural features within a unified data system.

[0053] In a preferred embodiment of the present invention, by mapping a preset attachment path to each closed segment, and correcting and adjusting the preset attachment path according to the segment shape offset value and the overall contour coupling value, target attachment path data is obtained, including: Based on the preset attachment path and the set of closed segments, each path node in the preset attachment path is divided into the corresponding closed segments according to its position range in the bidirectional fusion sequence, thus obtaining the segment mapping path. Based on the segment mapping path and segment shape offset value, the position of the path nodes in each closed segment is adjusted along the attachment direction to ensure that the arrangement of the path nodes changes accordingly with the degree of offset within the segment, thus obtaining the adjustment path within the segment. Based on the overall contour coupling value, the path nodes between adjacent closed segments are sequentially coordinated at the connection points to ensure that the path nodes at the connection points maintain a continuous transition between different closed segments, thus obtaining the inter-segment transition path. Based on the intra-segment adjustment path and inter-segment transition path, all path nodes are rearranged in the order of the bidirectional fusion sequence, and the order of overlapping and discontinuous path nodes is corrected to obtain the reconstructed attachment path. Based on the reconstructed attachment path, the positions of the path nodes in each closed segment are unified to obtain the target attachment path data.

[0054] In this embodiment of the invention, based on a preset attachment path and a set of closed segments, each path node in the preset attachment path is divided into corresponding closed segments according to its position range in the bidirectional fusion sequence, resulting in a segment mapping path. The preset attachment path is decomposed into the data range corresponding to each closed segment, establishing a direct correspondence between the path data and the contour segment data. Based on the segment mapping path and the segment shape offset value, the position of the path nodes in each closed segment is adjusted along the attachment direction to ensure that the arrangement position of the path nodes changes accordingly with the degree of internal offset of the segment, resulting in an adjustment path within the segment. This ensures that the distribution state of the path nodes in each closed segment changes accordingly with the shape offset characteristics of the segment, realizing the conversion of segment feature data into path node coordinate data. Based on the overall contour coupling value, adjacent closed segments are... The path nodes between merging segments are sequentially coordinated at the connection points to ensure a continuous transition between different closed segments, resulting in inter-segment transition paths. This establishes a continuous transition relationship at the boundaries of the paths that have been corrected within different closed segments. Based on the intra-segment adjustment paths and inter-segment transition paths, all path nodes are rearranged according to the bidirectional fusion sequence, and overlapping and discontinuous path nodes are sequentially corrected to obtain reconstructed attachment paths. The intra-segment adjustment results and inter-segment transition results are integrated into a continuous path. Based on the reconstructed attachment paths, the positions of the path nodes in each closed segment are unified to obtain target attachment path data. This standardizes the organizational relationship of path nodes within and between segments, achieving a complete transformation from preset path input to target path output.

[0055] Specifically, based on the segment mapping path and segment shape offset value, the positions of the path nodes within each closed segment are adjusted along the attachment direction to ensure that the arrangement of the path nodes changes accordingly with the degree of offset within the segment, thus obtaining the adjustment path within the segment, which specifically includes: The system reads the set of path nodes corresponding to each closed segment in the segment mapping path and obtains the segment shape offset value associated with each closed segment. For each closed segment, the system establishes an internal direction reference coordinate system based on a preset attachment direction or the local attachment direction corresponding to the segment on the workpiece surface. This direction reference coordinate system is determined based on the spatial distribution of contour points in the bidirectional fusion sequence, such as using the direction of the line connecting the first and last contour points within the segment, the direction of the tangent of the local surface, or the device execution direction as the attachment direction reference. For each path node, the system calculates the corresponding adjustment weight based on its relative position within the segment, such as the node number, the normalized position of the node in the segment length, or the distance between the node and the segment boundary, and distributes the segment shape offset value to each node according to this weight. Subsequently, the system applies an offset to the spatial coordinates of the path nodes along the attachment direction to obtain the adjusted node position. For path nodes located in the middle of the segment, a larger offset can be assigned to reflect the influence of the internal contour changes of the segment on the path; for path nodes near the segment boundary, the offset can be appropriately reduced to avoid excessive impact on the connection of subsequent segments. The system re-associates the adjusted node coordinates with the original node numbers and segment identifiers to form the adjusted path data within the closed segment.

[0056] Specifically, based on the overall contour coupling value, the path nodes between adjacent closed segments are sequentially coordinated at the connection points to ensure a continuous transition between different closed segments, resulting in inter-segment transition paths, which include: The system reads the overall contour coupling value and the segment boundary mapping relationship, and determines the connection positions between adjacent closed segments based on this mapping relationship. Then, it extracts the set of boundary nodes of adjacent segments from the adjusted paths within each segment, i.e., several path nodes at the end of the previous segment and several path nodes at the beginning of the next segment, forming a candidate node set for the connection region. Using the overall contour coupling value as an adjustment parameter, the system performs sequential coordination processing on these candidate nodes, analyzing the differences in spatial position, node spacing, and direction changes between the end nodes of the previous segment and the beginning nodes of the next segment, and determining the coordination strength based on the overall contour coupling value. When the overall contour coupling value reflects a high degree of connection consistency between segments, the system enhances the alignment between nodes, for example, by reducing the spatial gap between preceding and following nodes and adjusting the node arrangement order, making the paths of the two segments tend to be continuous at the boundary. When the overall contour coupling value reflects a low degree of connection consistency, a certain degree of node difference is allowed, but abrupt connections are still avoided through smoothing processing. Interpolation calculations are performed between boundary nodes to generate several transition nodes between the end nodes of the previous segment and the beginning nodes of the next segment, or the original node positions are interpolated and adjusted. For example, when a misalignment is detected between the end node of the previous segment and the beginning node of the next segment, the system renumbers or adjusts the node arrangement to maintain a progressive relationship between the nodes at the connection point. The system summarizes the coordination results at the connection points of all adjacent segments to form the transition path data between segments.

[0057] Specifically, based on the intra-segment adjustment path and inter-segment transition path, all path nodes are rearranged according to the bidirectional fusion sequence, and the order of overlapping and discontinuous path nodes is corrected to obtain the reconstructed attachment path, which specifically includes: First, the system summarizes the intra-segment adjustment paths of all closed segments and the inter-segment transition paths between adjacent segments to form a unified set of path nodes. Then, based on the arrangement order of the closed segments in the bidirectional fusion sequence, the system sequentially splices the path nodes and corresponding transition nodes within each segment according to the segment order. Using the node's position in the bidirectional fusion sequence or its segment position as the sorting criterion, all nodes are reordered to form a preliminary reconstructed path sequence. The system performs consistency checks on the spatial and sequential relationships between path nodes. For overlapping nodes (multiple nodes being spatially close or describing the same position repeatedly in sequence), the system eliminates them by deleting redundant nodes, merging adjacent nodes, or averaging node coordinates. For discontinuous nodes (where the distance between adjacent nodes is too large or the path continuity is interrupted), the system generates supplementary nodes through interpolation or adjusts the positions of neighboring nodes. The system always uses the order of the bidirectional fusion sequence as a constraint to ensure that the arrangement order of all nodes is consistent with the overall order of the contour data. The system redistributes node spacing based on the total path length, making the nodes more evenly distributed in space. For areas with abrupt changes in direction in the path, a local smoothing algorithm is used to correct the node directions, ensuring the path changes remain continuous. After completing the above-mentioned overlap elimination, discontinuity repair, and sequence correction, a path node sequence is obtained, which is the reconstructed attachment path. This path has structurally integrated the adjustment results within the segments and the transition results between segments, and the sequence correction eliminates local anomalies, giving it complete path continuity and consistency, which can serve as the basis for the final attachment path data.

[0058] Embodiments of the present invention also provide a curvature matching system for automatic labeling, the system comprising: The data module is used to acquire the contour point data of the workpiece to be mounted; The fusion module is used to create index sequences for each contour point according to the forward and reverse order based on the contour point data, and to alternately insert the forward and reverse index sequences according to their positions to obtain a bidirectional fusion sequence. The constraint module is used to identify the direction of change of contour points based on the bidirectional fusion sequence and merge contour points with the same direction of change into a change interval. The change intervals with lengths that meet the standard are used as constraint source intervals to obtain a constraint interval sequence. The closure module is used to merge intervals with the same direction of change into segments based on the constraint interval sequence, and to expand the range of the segments to both sides with the constraint source interval as the center to obtain a set of closed segments; The offset module is used to establish a unidirectional change path in each closed segment according to the contour point sequence based on the closed segment set, identify the degree of offset of the closed segment along the attachment direction, and obtain the segment shape offset value. The mapping module is used to extract the boundary contour points between adjacent closed segments based on the set of closed segments, and to establish the boundary connection relationship between closed segments based on the boundary contour points, thereby obtaining the segment boundary mapping set; The coupling module is used to connect and combine the boundary connection relationships between closed segments in series according to the segment boundary mapping set, identify the degree of connection consistency between different closed segments, and obtain the overall contour coupling value. The path module is used to map a preset attachment path to each closed segment, and then correct and adjust the preset attachment path according to the segment shape offset value and the overall contour coupling value to obtain the target attachment path data.

[0059] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0060] Embodiments of the present invention also provide a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0061] Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0062] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for matching the curvature of an automatic label, characterized in that, The method includes: Obtain the contour point data of the workpiece to be mounted; Based on the contour point data, index sequences are established for each contour point in both forward and reverse order. The forward and reverse index sequences are then inserted alternately according to their positions to obtain a bidirectional fusion sequence. Based on the bidirectional fusion sequence, the changing direction of the contour points is identified and contour points with the same changing direction are merged into a change interval. The change intervals with lengths that meet the standard are used as the constraint source intervals to obtain the constraint interval sequence. Based on the constraint interval sequence, the intervals with the same direction of change are merged into segments, and the range of the segments is expanded to both sides with the constraint source interval as the center to obtain a closed segment set; Based on the set of closed segments, a unidirectional change path is established in each closed segment according to the contour point sequence. The degree of offset of the closed segment along the attachment direction is identified, and the segment shape offset value is obtained. Based on the set of closed segments, extract the boundary contour points between adjacent closed segments, and establish the boundary connection relationship between closed segments based on the boundary contour points to obtain the segment boundary mapping set; Based on the segment boundary mapping set, the boundary connection relationships between each closed segment are connected in series and combined to identify the degree of connection consistency between different closed segments and obtain the overall contour coupling value. By mapping the preset attachment path to each closed segment, the preset attachment path is corrected and adjusted according to the segment shape offset value and the overall contour coupling value to obtain the target attachment path data.

2. The arc matching method for automatic labeling according to claim 1, characterized in that, Based on the contour point data, index sequences are created for each contour point in both forward and reverse order. These forward and reverse index sequences are then alternately inserted according to their positions to obtain a bidirectional fused sequence, including: Based on the contour point data, each contour point is sequentially numbered according to the collection order of the contour points, and arranged in ascending order to form a forward point column, and in descending order to form a reverse point column. Based on the forward point column and the reverse point column, the contour points in the forward point column and the reverse point column are labeled with the number of each contour point as the index identifier, so as to obtain the forward index sequence and the reverse index sequence. By alternating the contour points at the same position in the forward index sequence and the reverse index sequence, ensuring that each position contains both forward index information and reverse index information, an alternating fusion sequence is obtained. Based on the alternating fusion sequence, the forward and reverse index information of each contour point is combined and identified, and the sequence position is recalibrated according to the alternating arrangement order to obtain the bidirectional fusion sequence.

3. The arc matching method for automatic labeling according to claim 2, characterized in that, Based on the bidirectional fusion sequence, the direction of change of contour points is identified, and contour points with the same direction of change are merged into a change interval. Change intervals with lengths conforming to a standard are used as constraint source intervals, resulting in a constraint interval sequence, including: Based on the bidirectional fusion sequence, the direction of change between contour points is determined by comparing the increase or decrease of the index values ​​of adjacent contour points, thus obtaining the direction identification sequence. Based on the direction identification sequence, the contour points with continuous and consistent change directions are merged according to their arrangement order in the bidirectional fusion sequence, and the start and end positions are recorded to obtain the change interval set; Based on the set of change intervals, count the number of contour points contained in each change interval, and determine the change intervals with a consistent number of contour points as the target intervals, thus obtaining the target interval set; The target interval is used as the source interval of the constraint, and the direction of change of the adjacent intervals is uniformly calibrated according to the direction of change of the source interval of the constraint, so as to obtain the sequence of constraint intervals.

4. The arc matching method for automatic labeling according to claim 3, characterized in that, Based on the set of closed segments, a unidirectional change path is established within each closed segment according to the sequence of contour points. The degree of offset of the closed segment along the attachment direction is identified, and the segment shape offset value is obtained, including: Based on the set of closed segments, calculate the degree of difference between positive and negative index changes for adjacent contour points to obtain the index deviation item; Based on the set of closed segments, identify whether the cumulative changes of each contour point inside the closed segment are balanced when unfolded along the attachment direction, and obtain the path offset term. Based on the closed segment, calculate the difference between the forward and reverse indices at both ends of the closed segment on the bidirectional index, identify the boundary consistency of the closed segment at the beginning and end positions, and obtain the beginning and end coupling term; Based on the index deviation term and path offset term, the influence of local contour changes on the path distribution in the attachment direction is identified, and the segment interaction term is obtained. Based on the head-tail coupling term, the path offset terms of each contour point in the closed section are aggregated to identify the common influence between the overall internal offset state of the closed section and the head-tail boundary state, and the boundary compensation term is obtained. By fusing the segment interaction term and the boundary compensation term, the combined effect of changes in the internal contour of the closed segment and the difference between the beginning and end boundaries on the degree of attachment direction offset is identified, and the segment shape offset value is obtained.

5. The arc matching method for automatic labeling according to claim 4, characterized in that, Based on the set of closed segments, extract the boundary contour points between adjacent closed segments, and establish the boundary connection relationships between closed segments based on the boundary contour points, thus obtaining the segment boundary mapping set, including: Based on the set of closed segments, the start and end positions of each closed segment in the bidirectional fusion sequence are extracted, and the contour points at the corresponding positions are determined as segment boundary contour points to obtain the set of segment boundary points. Based on the set of boundary points of each segment, the termination position of each closed segment is sequentially matched with the starting position of its adjacent closed segment to determine the pairs of segments with adjacent positional relationships, thus obtaining a set of adjacent segment pairs. Based on the adjacent segment pairs, the contour points are aligned according to their relative positions in the bidirectional fusion sequence, and the contour points at the termination position and the starting position are made to form a correspondence, thus obtaining the boundary correspondence set. Based on the boundary correspondence set, the boundary contour points with corresponding relationships are filtered, the correspondence with the smallest position difference is retained and the rest are deleted to obtain the effective boundary relationship set. Based on the effective boundary relationship set, the correspondence between each closed segment is arranged in the order of the bidirectional fusion sequence to obtain the segment boundary mapping set.

6. The arc matching method for automatic labeling according to claim 5, characterized in that, Based on the segment boundary mapping set, the boundary connection relationships between each closed segment are concatenated and combined to identify the degree of consistency between different closed segments, thereby obtaining the overall contour coupling value, including: Based on the segment boundary mapping set, calculate the degree of difference between the boundary contour points of the front and back closed segments in the forward and reverse index information to obtain the boundary difference term. Based on the segment boundary mapping set, calculate the proportion of the connection span between adjacent closed segments relative to the boundary position range to obtain the connection span term; Based on the segment connection path formed by the concatenation of segment boundary mapping sets, the degree of deviation of the closed segment arrangement in the connection path relative to the continuous sequential advancement is calculated to obtain the sequence perturbation term; Based on the number of closed segments in the segment connection path, the degree of matching between the path coverage status and the segment distribution range is calculated to obtain the coverage discrete term; By fusing the boundary difference term and the connection span term, a local connection term is obtained; by fusing the sequence disturbance term and the coverage discrete term, an overall connection term is obtained; by fusing the local connection term and the overall connection term, the degree of connection consistency of the closed segment on the overall connection path is identified, and the overall contour coupling value is obtained.

7. The arc matching method for automatic labeling according to claim 6, characterized in that, By mapping a preset attachment path to each closed segment, and correcting and adjusting the preset attachment path based on the segment shape offset value and the overall contour coupling value, the target attachment path data is obtained, including: Based on the preset attachment path and the set of closed segments, each path node in the preset attachment path is divided into the corresponding closed segments according to its position range in the bidirectional fusion sequence, thus obtaining the segment mapping path. Based on the segment mapping path and segment shape offset value, the position of the path nodes in each closed segment is adjusted along the attachment direction to ensure that the arrangement of the path nodes changes accordingly with the degree of offset within the segment, thus obtaining the adjustment path within the segment. Based on the overall contour coupling value, the path nodes between adjacent closed segments are sequentially coordinated at the connection points to ensure that the path nodes at the connection points maintain a continuous transition between different closed segments, thus obtaining the inter-segment transition path. Based on the intra-segment adjustment path and inter-segment transition path, all path nodes are rearranged in the order of the bidirectional fusion sequence, and the order of overlapping and discontinuous path nodes is corrected to obtain the reconstructed attachment path. Based on the reconstructed attachment path, the positions of the path nodes in each closed segment are unified to obtain the target attachment path data.

8. A curvature matching system for automatic labeling, characterized in that, The system is used to perform the method as described in any one of claims 1 to 7, the system comprising: The data module is used to acquire the contour point data of the workpiece to be mounted; The fusion module is used to create index sequences for each contour point according to the forward and reverse order based on the contour point data, and to alternately insert the forward and reverse index sequences according to their positions to obtain a bidirectional fusion sequence. The constraint module is used to identify the direction of change of contour points based on the bidirectional fusion sequence and merge contour points with the same direction of change into a change interval. The change intervals with lengths that meet the standard are used as constraint source intervals to obtain a constraint interval sequence. The closure module is used to merge intervals with the same direction of change into segments based on the constraint interval sequence, and to expand the range of the segments to both sides with the constraint source interval as the center to obtain a set of closed segments; The offset module is used to establish a unidirectional change path in each closed segment according to the contour point sequence based on the closed segment set, identify the degree of offset of the closed segment along the attachment direction, and obtain the segment shape offset value. The mapping module is used to extract the boundary contour points between adjacent closed segments based on the set of closed segments, and to establish the boundary connection relationship between closed segments based on the boundary contour points, thereby obtaining the segment boundary mapping set; The coupling module is used to connect and combine the boundary connection relationships between closed segments in series according to the segment boundary mapping set, identify the degree of connection consistency between different closed segments, and obtain the overall contour coupling value. The path module is used to map a preset attachment path to each closed segment, and then correct and adjust the preset attachment path according to the segment shape offset value and the overall contour coupling value to obtain the target attachment path data.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images