Intelligent detection method and system for drawing differences

Abstract

The present application provides a method and system for intelligent detection of drawing differences, which relates to the technical field of drawing data processing, including: obtaining a drawing transmission module, wherein the drawing transmission module includes a plurality of different drawing transmission modes; reading information items such as drawing geometric features, annotation information, and file format information of the drawings to be transmitted; combining the drawing transmission module, using a drawing difference prediction model to predict drawing differences for information items of the drawings to be transmitted, and outputting a plurality of drawing difference indicators corresponding to the plurality of drawing transmission modes; combining and matching drawing transmission modes according to the plurality of drawing difference indicators, outputting a matching drawing transmission mode combination, and combining and transmitting the drawings to be transmitted with the matching drawing transmission mode combination. The present application solves the technical problem that the traditional drawing transmission process cannot flexibly adjust the transmission strategy according to the drawing characteristics, resulting in insufficient accuracy in the transmission of drawing information, and achieves the technical effect of improving the accuracy and consistency of drawing transmission.

Description

Technical Field

[0001] The present application relates to the technical field of drawing data processing, and in particular to a method and system for intelligent detection of drawing differences. Background Art

[0002] In the field of engineering design and manufacturing, drawings, as the core carrier of project planning, design and construction, play an important role in transmission. With the widespread application of digital design, the form of drawings has gradually shifted from traditional paper versions to electronic versions, and they are transmitted between different systems and platforms. However, due to the complexity of drawing files, especially the diversity of geometric features, annotation information and file formats, the drawing transmission process faces many challenges.

[0003] Most existing drawing transmission methods rely on a single transmission method, usually selecting a fixed transmission path based on file size or format. This method is more effective when transmitting simple drawings, but it is difficult to guarantee transmission efficiency and accuracy when facing complex drawings containing multiple geometric features and annotation information. During the transmission process, drawing information is frequently lost, distorted, or inconsistent, resulting in the inability to fully match versions of design information on different platforms, affecting the smooth progress of design and construction. Summary of the invention

[0004] The present application provides an intelligent drawing difference detection method and system, which solves the technical problem that the traditional drawing transmission process has a single transmission method and cannot flexibly adjust the transmission strategy according to the different characteristics of the drawings, resulting in loss or distortion of drawing information, and achieves the technical effect of improving the accuracy and consistency of drawing transmission.

[0005] In view of the above problems, on the one hand, the present application provides an intelligent drawing difference detection method, which includes: obtaining a drawing transmission module, wherein the drawing transmission module includes multiple drawing transmission modes, and each drawing transmission mode is different; reading information items of the drawings to be transmitted, wherein the information items of the drawings to be transmitted include drawing geometric features, annotation information, and file format information; combining the drawing transmission module, using a drawing difference prediction model to predict drawing differences for the information items of the drawings to be transmitted, and outputting multiple drawing difference indicators corresponding to the multiple drawing transmission modes respectively; performing drawing transmission mode combination matching according to the multiple drawing difference indicators, outputting a matching drawing transmission mode combination, and combining and transmitting the drawings to be transmitted with the matching drawing transmission mode combination.

[0006] On the other hand, the present application also provides an intelligent drawing difference detection system, the system comprising: a transmission mode reading unit, the transmission mode reading unit is used to obtain a drawing transmission module, wherein the drawing transmission module includes multiple drawing transmission modes, and each drawing transmission mode is different; a drawing information reading unit, the drawing information reading unit is used to read information items of drawings to be transmitted, and the information items of drawings to be transmitted include drawing geometric features, annotation information and file format information; a drawing difference prediction unit, the drawing difference prediction unit is used to combine the drawing transmission module, use a drawing difference prediction model to predict drawing differences for the information items of the drawings to be transmitted, and output multiple drawing difference indicators corresponding to the multiple drawing transmission modes respectively; a combined transmission matching unit, the combined transmission matching unit is used to perform drawing transmission mode combination matching according to the multiple drawing difference indicators, output a matching drawing transmission mode combination, and use the matching drawing transmission mode combination to perform combined transmission of the drawings to be transmitted.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] A drawing transmission module is obtained, wherein the drawing transmission module includes a plurality of different drawing transmission modes, which can adapt to the geometric features, annotation complexity and file format requirements of different drawings, and provide a rich basic option for drawing transmission. The drawing geometric features, annotation information and file format information of the drawings to be transmitted are read, the characteristics and transmission requirements of each drawing are accurately identified, and a data basis is provided for difference detection and transmission mode selection. In conjunction with the drawing transmission module, the drawing difference prediction model is used to predict the information items of the drawings to be transmitted, and a plurality of drawing difference indicators corresponding to the plurality of drawing transmission modes are output, and the influence of each transmission mode on the geometric features and annotations of the drawings is clarified, so as to identify potential problems in advance before transmission, and effectively reduce the problems of inconsistent data and distortion after transmission. The drawing transmission mode combination is matched according to the plurality of drawing difference indicators, and the optimal matching drawing transmission mode combination is output, and the drawings to be transmitted are combined and transmitted with the matching drawing transmission mode combination, so as to realize efficient and accurate drawing transmission. Compared with a single transmission mode, the combined transmission can be dynamically optimized according to the different requirements of the drawings, greatly improving the efficiency and accuracy of drawing transmission, and reducing the risk of information loss.

[0009] To summarize, this application dynamically adapts to the needs of different types of drawings through intelligent prediction and optimized combination, automatically matches the optimal combined transmission method, effectively reduces errors and information loss in transmission, makes the entire transmission process more flexible and efficient, and improves the accuracy and consistency of drawing transmission.

[0010] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A schematic diagram of a process flow of a method for intelligently detecting drawing differences provided in an embodiment of the present application;

[0012] Figure 2 A schematic diagram of a process of predicting drawing differences for information items of drawings to be transmitted in the drawing difference intelligent detection method provided in an embodiment of the present application;

[0013] Figure 3 A schematic diagram of the structure of the intelligent drawing difference detection system provided in an embodiment of the present application.

[0014] Explanation of the reference numerals: transmission mode reading unit 10 , drawing information reading unit 20 , drawing difference prediction unit 30 , combined transmission matching unit 40 . DETAILED DESCRIPTION

[0015] The embodiments of the present application provide an intelligent drawing difference detection method and system to solve the technical problem that the traditional drawing transmission process has a single transmission mode and cannot flexibly adjust the transmission strategy according to the different characteristics of the drawings, resulting in loss or distortion of drawing information, thereby achieving the technical effect of improving the accuracy and consistency of drawing transmission.

[0016] Embodiment 1, as Figure 1 As shown, the embodiment of the present application provides a method for intelligently detecting drawing differences, the method comprising:

[0017] Step S1: Acquire a drawing transmission module, wherein the drawing transmission module includes a plurality of drawing transmission modes, and each drawing transmission mode is different from another.

[0018] Specifically, the drawing transmission module is a functional component used to transmit drawing data. It is responsible for managing and executing the drawing transmission process, and contains multiple pre-defined ways to transmit drawings and related operation logic. Each way includes different network protocols, compression algorithms, encryption methods, etc.

[0019] Get the configuration information of the drawing transfer module from the system resource library or server, and read various transmission methods, such as FTP file transfer, HTTP upload and download, SFTP encrypted transmission protocol, etc. By obtaining these different transmission methods, you can flexibly select the most appropriate method for transmission according to the specific drawing type, complexity, file format and transmission requirements, which provides a basis for subsequent transmission method selection.

[0020] Step S2: reading information items of the drawing to be transmitted, wherein the information items of the drawing to be transmitted include drawing geometric features, annotation information, and file format information.

[0021] Specifically, the drawings to be transmitted are usually digital files, such as CAD drawings, PDF drawings or 3D models. Extract and analyze the key data of the drawings to be transmitted, that is, the information items of the drawings to be transmitted, clarify the content and characteristics of the drawings, and then judge the complexity, importance and applicability of the drawings, providing a basis for the subsequent selection of transmission methods. These information items include geometric features, annotation information and file format information.

[0022] Drawing geometric features include the size, shape, lines, angles and other geometric properties of the drawing, which determine the complexity of the drawing. For example, for a relatively small amount of data in a flat architectural design (two-dimensional geometric features), simple network transmission can meet the transmission requirements. For complex three-dimensional mechanical parts model drawings (such as .STL files), the geometric features are complex and the file size is large, so it is necessary to consider compression or segmented transmission when transmitting. To read geometric features, you can use image processing libraries such as OpenCV to pre-process the drawings, including grayscale, binarization, edge detection, etc., to highlight the graphic features; and use computer vision algorithms, such as Hough transform or contour detection, to identify and measure the size and position of geometric figures.

[0023] Annotation information includes text descriptions, notes, labels, etc. on drawings, which are used to explain or supplement the visual content of drawings. To read annotation information, OCR (Optical Character Recognition) technology, such as Tesseract OCR engine, can be used to recognize text information on drawings, or natural language processing (NLP) technology can be used to understand the semantics of annotation information, such as extracting dimension values ​​and units.

[0024] File format information refers to the format type of the drawing file, such as PDF, DWG, JPEG, etc., which determines how the drawings are stored, viewed, and edited. For example, CAD drawings in DWG format usually require specialized CAD software support, and may need to be processed through cloud collaboration tools (such as AutoCAD 360) when transmitted. Drawings in PDF format have wider compatibility and are suitable for sharing via email or cloud storage. Reading file format information can determine the file format by checking the file's metadata or file extension.

[0025] Step S3: in conjunction with the drawing transmission module, a drawing difference prediction model is used to predict drawing differences on the information items of the drawings to be transmitted, and a plurality of drawing difference indicators corresponding to the plurality of drawing transmission modes are output.

[0026] Specifically, the drawing difference prediction model is a machine learning model trained based on historical data, which is used to predict the possible differences in drawings under different transmission methods. The drawing difference index is a quantitative index output by the drawing difference prediction model, which is used to measure the degree of information difference that may be caused by different transmission methods.

[0027] Through the drawing difference prediction model, each transmission method in the drawing transmission module is analyzed, and the geometric characteristics, annotation information, file format information and other information items of the drawing are comprehensively analyzed to predict the differences that may occur under different transmission methods, so as to evaluate the degree of drawing information difference that may be caused by each method and determine the drawing difference index corresponding to each transmission method. The drawing difference index can quantify and compare the potential risks under each transmission method, provide data support for the subsequent selection of the most appropriate transmission method, thereby improving the accuracy of drawing transmission and effectively avoiding information loss or errors caused by inappropriate transmission methods.

[0028] Step S4: performing drawing transmission mode combination matching according to the multiple drawing difference indicators, outputting a matching drawing transmission mode combination, and combining and transmitting the drawings to be transmitted with the matching drawing transmission mode combination.

[0029] Specifically, according to the multiple drawing difference indicators generated in step S3, different transmission modes are combined and matched to formulate multiple transmission strategies. The optimal transmission strategy is selected from the multiple transmission strategies through the optimization algorithm, that is, the matching drawing transmission mode combination is used to meet the complex and diversified drawing transmission requirements. After the matching drawing transmission mode combination is determined, the drawings are combined and transmitted according to the matching drawing transmission mode combination.

[0030] Combined transmission first generates multiple drawing files through different transmission methods. Each drawing file corresponds to a drawing transmission method and retains different characteristics. For example, a PDF file may lose some layer information, but the dimensioning is still accurate; while a DWG file retains all layers, but lacks some visual details. Then, according to the difference indicators of each transmission method, these files are compared and integrated, and a transmission method (or file) is selected as the standard file. Supplementary information is extracted from other transmission methods to supplement the standard file, and the final transmission result file is generated. The transmission result file maximizes the retention of information items in the original drawing.

[0031] Exemplarily, corresponding PDF drawing files, DWG drawing files and SVG drawing files are generated through three transmission methods: PDF, DWG and SVG. Since DWG files can retain complete geometry and editing features, DWG files are selected as the basis. The PDF file retains some annotation information that is not clearly displayed in the DWG file. These annotations are extracted and integrated into the DWG file. Due to the characteristics of the conversion method, some lines of the vector graphics in the SVG file are retained more accurately. These vector data are replaced in the DWG file to ensure the accuracy of the lines. Through supplementary integration, an accurate DWG file is generated, which contains complete geometry data, clear annotation information, lossless layer structure and high-precision graphic elements.

[0032] By combining and matching drawing transmission methods and finding the combination that best suits the current drawing characteristics among multiple transmission methods, it can be ensured that the drawings are transmitted in the shortest time, in the safest and most accurate way. This not only speeds up the transmission efficiency, but also effectively maintains the integrity and security of the drawings, achieving the best transmission effect.

[0033] Furthermore, before combining and matching the drawing transmission modes according to the plurality of drawing difference indicators, the method further includes:

[0034] Configure a preset drawing difference index; compare the multiple drawing difference indexes according to the preset drawing difference index, and if there is a drawing transmission method that is smaller than the preset drawing difference index, output a first drawing transmission method, and perform a single transmission on the drawing to be transmitted based on the first drawing transmission method; wherein the first drawing transmission method is a drawing transmission method with the minimum drawing difference index.

[0035] Specifically, before combining and matching the drawing transmission methods, one or more preset drawing difference indicators are first set according to the importance of the drawing, the purpose of the transmission and the requirements of the recipient. The preset drawing difference indicator is used to measure the maximum difference or deviation allowed during the drawing transmission process. It can be a specific value or range, indicating the acceptable geometric feature deformation, annotation information loss or file format compatibility issues during the transmission process.

[0036] Compare the difference index of each transmission method with the preset drawing difference index one by one to determine whether there is a drawing transmission method with a difference index smaller than the preset drawing difference index. If so, sort these single transmission methods that meet the drawing difference index requirements according to the size of the drawing difference index, select the transmission method with the smallest difference index as the first drawing transmission method, and directly use the first drawing transmission method for single transmission without combining multiple transmission methods.

[0037] By presetting standards and comparison, if there is a transmission method with different indicators that meets the requirements, a single method will be directly selected for transmission without combining complex transmission methods. While ensuring the quality of drawing transmission, the transmission process is simplified to avoid unnecessary waste of resources.

[0038] Further, if there is no drawing transmission method with a difference less than the preset drawing difference index, the drawing transmission method combination matching is performed according to the multiple drawing difference indexes, specifically including:

[0039] According to the multiple drawing transmission modes included in the drawing transmission module, the drawing feature retention capability is identified to generate a retention feature label; the multiple drawing transmission modes are randomly combined for dual transmission to generate multiple dual transmission combination schemes, and multiple feature retention label groups corresponding to the multiple dual transmission combination schemes, wherein the dual transmission combination scheme sample is obtained by combining two different drawing transmission modes; the dual transmission combination scheme sample is optimized by a simulated annealing algorithm to output a matching drawing transmission mode combination.

[0040] Specifically, if the difference index comparison shows that there is no drawing transmission method with a difference index less than the preset drawing difference index, it is necessary to combine and match multiple drawing transmission methods to find a drawing transmission method combination that meets the transmission requirements. In order to save resources, first consider combining two drawing transmission methods.

[0041] Analyze the historical data and transmission performance of each drawing transmission method in the drawing transmission module, determine the ability of each transmission method to retain geometric features, annotation information, file format and other features during the transmission process, and generate a retained feature label for each drawing transmission method based on the transmission results, indicating the feature types and retention capabilities of the corresponding drawing transmission method. For example, the compression transmission method may perform better in retaining geometric features, while the encryption transmission method is more stable in retaining annotation information.

[0042] All drawing transmission modes in the drawing transmission module are randomly combined in pairs to generate multiple dual transmission combination schemes, each of which contains two different drawing transmission modes. At the same time, a feature retention label group consisting of the retained feature labels of the two drawing transmission modes is generated for each dual transmission combination scheme. This label group reflects the overall capability of the scheme in feature retention.

[0043] Exemplarily, there are four transmission modes A, B, C, and D, among which A is good at preserving details, B is good at color restoration, C has an advantage in file size control, and D can ensure a higher transmission speed. Label the retained feature labels for these four transmission modes respectively. A is labeled with a "details" label, B is labeled with a "color" label, C is labeled with a "volume" label, and D is labeled with a "speed" label. Through random matching of two by two, six combinations of AB, AC, AD, BC, BD, and CD are generated. Label each combination with the corresponding feature retention label group. For example, the AB combination obtains the label of "details + color", while the CD combination is labeled with "volume + speed".

[0044] Since there are multiple dual transmission combination schemes, it is necessary to select an optimal transmission mode combination from them. The embodiment of the present application adopts a simulated annealing algorithm to evaluate all dual transmission combination schemes and find the optimal drawing transmission mode combination, that is, the matching drawing transmission mode combination. The simulated annealing algorithm is a heuristic search algorithm that finds the global optimal solution by random search and accepting suboptimal solutions. During the optimization process, the algorithm evaluates the feature retention capability of each dual transmission combination scheme, and gradually adjusts the combination to find the best combination scheme that can retain key drawing features and meet the transmission requirements.

[0045] The specific process of optimizing using the simulated annealing algorithm includes: defining an initial temperature T, that is, the amount of randomness allowed during the exploration process. The higher the temperature, the wider the range of random searches allowed. Setting an upper limit on the number of iterations, that is, the maximum number of iterations, is used to control the execution cycle of the algorithm.

[0046] Then, a sample of solutions is randomly selected from all dual transmission combination solutions as the initial solution. The initial solution is evaluated and its feature retention capability is calculated. The initial solution is set as the current solution. Based on the current solution, a neighborhood solution is generated, that is, one of the transmission modes in the transmission mode combination is changed to generate a new solution. For example, if the original combination is "compression + encrypted transmission", the neighborhood solution may be "compression + network transmission". The feature retention label group of the neighborhood solution is calculated. If the feature retention capability of the neighborhood solution is worse than the current solution, it is used as the new current solution. If the objective function of the neighborhood solution is not as good as the current solution, the inferior solution is accepted with a certain probability. This probability is obtained by the formula P=exp((-ΔE) / T), where ΔE is the difference in feature retention capability between the current solution and the new solution; T is the current temperature. After each iteration, the temperature T is reduced according to the set cooling strategy. Common cooling strategies include linear cooling (T=T-α, α is a constant decay rate) and exponential cooling (T=T•β, β is a decay factor less than 1). As the temperature decreases, the randomness of the algorithm decreases, the search pace gradually converges, and finally converges to a more optimized state. When the temperature T drops to a certain set value (that is, close to 0), or the number of iterations reaches the set upper limit, the algorithm terminates. At this time, the current solution is the optimized optimal solution. Output the drawing transmission mode combination corresponding to the solution, that is, the matching drawing transmission mode combination optimized by the simulated annealing algorithm.

[0047] The above dual transmission combination matching process makes full use of the characteristics of different transmission methods and finds the optimal or near-optimal combination solution through the simulated annealing algorithm, thereby solving the problem that a single transmission method cannot meet the needs and ensuring that all key features of the drawing can be transmitted as intactly as possible even in the absence of a single perfect method.

[0048] Further, the dual transmission combination scheme samples are optimized by using a simulated annealing algorithm to output a matching drawing transmission mode combination, and further comprising:

[0049] Decompose the multiple drawing difference indicators and output feature difference labels; perform matching calculations on the feature retention label groups corresponding to each dual transmission combination scheme according to the feature difference labels to obtain a matching score for each dual transmission combination scheme; and perform perturbations and updates on the dual transmission combination scheme according to the matching scores using a simulated annealing algorithm until a global optimization result is obtained, which is output as a matching drawing transmission mode combination.

[0050] Specifically, when using the simulated annealing algorithm for optimization, it is necessary to compare the feature retention capabilities of different dual transmission combination schemes. The comparison process first decomposes the difference indicators of the drawings and extracts the feature difference labels. These labels identify the processing performance of different transmission methods on the geometric features, annotation information, file format, etc. of the drawings. For example, a certain transmission method may have small differences in geometric feature processing, but large differences in the processing of annotation information, which will generate labels of "low geometric difference" and "high annotation difference" respectively.

[0051] For each generated dual transmission combination scheme, the corresponding feature retention label group is compared with the generated feature difference label, and the matching degree between the two is calculated to obtain a matching score. The higher the score, the better the dual transmission combination scheme retains the specific drawing information. During the optimization process, the simulated annealing algorithm is used to make slight adjustments (i.e., perturbations) to the current dual transmission combination scheme (current solution), such as replacing one of the transmission methods or changing the transmission order, so as to generate new dual transmission combination scheme samples. Each time a new scheme is generated, its matching score is calculated again. The matching scores of the new and old solutions are compared, and it is determined whether to update the current solution based on the comparison results. With the continuous iteration of the algorithm, the scheme gradually converges to a global optimal solution, and the dual transmission combination scheme represented by the output optimal solution is the matching drawing transmission method combination.

[0052] Furthermore, the optimization result of the global optimization is obtained, and it is identified whether the matching score of the optimization result is less than the preset matching score. If it is less than the preset matching score, three transmission combination scheme samples are generated; wherein, the three transmission combination scheme samples are obtained by combining three different drawing transmission methods; the three transmission combination scheme samples are optimized through a simulated annealing algorithm, and an updated matching drawing transmission method combination is output.

[0053] Specifically, the preset matching score is a matching threshold set according to the transmission requirements, which is used to measure the quality of the transmission scheme. Only schemes higher than this preset value are considered acceptable solutions.

[0054] First, evaluate the final optimization result of the dual transmission combination scheme and calculate its matching score. Compare the matching score of the optimization result with the preset matching score. If the score is less than the preset matching score, it means that the combination of the two drawing transmission methods alone cannot meet the drawing transmission requirements and needs further optimization.

[0055] At this time, three different drawing transmission methods are randomly combined from all drawing transmission methods to generate multiple three-transmission combination schemes. Then, a method similar to the two-transmission combination scheme is adopted, and the simulated annealing algorithm is used again to search for the best among the three-transmission combination scheme samples. Through iteration and perturbation, the combination method is gradually adjusted, and the matching score of each group of schemes is calculated to evaluate its performance in terms of feature retention and drawing differences. Finally, a new optimal solution is found, and the three-transmission combination scheme represented by the optimal solution is output as a new matching drawing transmission method combination. The three-transmission combination scheme will have a higher matching score and meet the requirements of drawing feature retention and transmission efficiency.

[0056] When the matching degree of the dual transmission scheme is insufficient, the above steps further explore a wider and deeper combination space of transmission modes by constructing a three-transmission combination scheme and optimizing it using the simulated annealing algorithm to find a drawing transmission solution that can better meet the transmission needs.

[0057] Further, such as Figure 2 As shown, the drawing difference prediction model is used to predict the drawing difference of the information item of the drawing to be transmitted, and further includes:

[0058] A drawing difference prediction model is constructed, wherein the difference items of the drawing difference prediction model include format conversion difference, dimensioning difference and position offset difference; the drawing difference prediction model predicts the size of the format conversion difference, the size of the dimensioning difference and the size of the position offset difference under the multiple drawing transmission modes according to the information items of the drawings to be transmitted; linear fitting is performed according to the size of the format conversion difference, the size of the dimensioning difference and the size of the position offset difference, and a plurality of drawing difference indicators corresponding to the multiple drawing transmission modes are output.

[0059] Specifically, a drawing difference prediction model is first constructed using a machine learning algorithm (such as linear regression, support vector machine, or neural network). This model is used to simulate and predict the difference items that may be caused by different drawing transmission methods.

[0060] The construction process of the drawing difference prediction model includes: collecting a large number of transmission data sets of different drawings under various transmission methods. These data sets should cover multiple drawing formats (such as CAD, PDF, JPG, etc.), as well as the difference results of dimensioning and position information before and after transmission under different transmission methods. The collected data is cleaned and standardized, and divided into training sets and test sets for model learning and verification.

[0061] Select a machine learning algorithm (such as linear regression, support vector machine, or neural network) and use the training set to train the model. During the training process, the model will use the difference data as the target variable and learn the patterns of various differences under different transmission methods based on the data in the training set. Use common loss functions such as mean square error (MSE) or mean absolute error (MAE) to evaluate whether the prediction results of the model on the test set are accurate. If the prediction accuracy is not ideal, use optimization algorithms such as gradient descent to further adjust the model's hyperparameters, such as learning rate, number of layers, number of nodes, etc., to optimize the model's prediction performance so that the model can better fit the relationship between input data and output differences. Through continuous iterative training, the model gradually converges, and finally a well-trained drawing difference prediction model is obtained.

[0062] The geometric features, annotation information, file format and other information items of the drawings to be transmitted are input into the drawing difference prediction model as input information. The model can predict the size of format conversion differences, dimension annotation differences and position offset differences that may occur under different transmission methods.

[0063] Each predicted difference size is linearly fitted to form a comprehensive drawing difference index, which comprehensively reflects the overall difference in format, size and position of the drawings. Linear fitting can be obtained by setting weights for different difference items and then performing weighted summation. By linearly fitting the predicted results of each drawing transmission method, the drawing difference index corresponding to each drawing transmission method is finally determined. These indicators reflect the degree of influence of different transmission methods on the integrity of drawing information.

[0064] The above steps construct a drawing difference prediction model and linear fitting to quantitatively predict and evaluate the loss or deviation of drawing information in multiple aspects such as format conversion, dimensioning, and position offset under different transmission methods, providing a basis for the combination optimization of transmission methods, so as to find the best combination of transmission methods for retaining drawing information.

[0065] Furthermore, the format conversion differences include the loss of features of special line types, layer structures, and drawing symbols; the dimension annotation differences include the offset, font changes, and scale distortion of annotation text; and the position offset differences include the plane displacement differences and space displacement differences that may be caused by different transmission methods.

[0066] Specifically, the difference items of the drawing difference prediction model include format conversion difference, dimensioning difference and position offset difference.

[0067] When converting drawings between different formats, special line types, layer structures, and drawing symbol features may be lost. This includes: specific line types (such as dashed lines and dotted lines) may not be fully recognized and supported by the target format or the receiving software after the format conversion, resulting in the loss or deformation of line type features. For example, some complex dashed lines or dotted lines are simplified to ordinary straight lines when converted to PDF. The layer information in the original drawing may be ignored or merged during the format conversion, causing layer confusion and affecting the readability and editability of the drawing. For example, different layers in CAD may be merged into a single layer in PDF. Professional symbols (such as electrical symbols and mechanical symbols) may be simplified or replaced during the conversion process, resulting in information transmission errors. For example, arrows, text boxes, etc. may be lost or deformed due to incomplete conversion.

[0068] During the drawing transmission process, there may be differences in dimensioning, such as offset of annotation text, font changes, and scale distortion. For example, the position of the dimensioning after transmission does not match the position on the original drawing, resulting in incorrect interpretation of the drawing information; the font of the annotation text becomes unclear in certain formats, or the size and value change, affecting the uniformity and professionalism of the drawing; the scale of the annotation dimension changes, which directly affects the accuracy and practicality of the drawing.

[0069] In addition, there will be plane displacement differences and spatial displacement differences in the drawing transmission process. For example, the objects in the drawing will be offset in the plane coordinates, affecting the geometric relationship and spatial layout of the drawing. After the three-dimensional drawing is transmitted, the position of the object on the X, Y, and Z axes will deviate, affecting the integrity of the model.

[0070] The above differences in the drawing transmission process will affect the accuracy of drawing information transmission to varying degrees. In order to evaluate these losses, the drawing difference prediction model needs to set the following difference items: format conversion differences including special line types, layer structures, and feature loss of drawing symbols; dimension annotation differences including offset, font changes, and proportional distortion of annotation text; position offset differences including plane displacement differences and spatial displacement differences that may be caused by different transmission methods. Through the drawing difference prediction model, the size of each difference item under different transmission methods can be carefully analyzed and quantified, and the transmission risk of each drawing transmission method can be more accurately evaluated, thereby providing a reference for the subsequent selection of transmission methods.

[0071] In summary, the intelligent drawing difference detection method provided in the embodiment of the present application has the following technical effects:

[0072] A drawing transmission module is obtained, wherein the drawing transmission module includes a plurality of different drawing transmission modes, which can adapt to the geometric features, annotation complexity and file format requirements of different drawings, and provide a rich basic option for drawing transmission. The drawing geometric features, annotation information and file format information of the drawings to be transmitted are read, and the characteristics and transmission requirements of each drawing are accurately identified, so as to provide a data basis for difference detection and transmission mode selection. A drawing difference prediction model is constructed by machine learning algorithm and historical data, and the drawing transmission module is combined to predict the drawing difference of the information items of the drawings to be transmitted by the drawing difference prediction model, and a plurality of drawing difference indicators corresponding to the plurality of drawing transmission modes are output, and the influence of each transmission mode on the geometric features and annotation of the drawings is clarified, so as to identify potential problems in advance before transmission, and effectively reduce the problems of inconsistent data and distortion after transmission. According to the plurality of drawing difference indicators, the drawing transmission mode combination is matched, the combination complexity of the transmission mode is gradually increased, and the simulated annealing algorithm is combined to optimize, and the matching drawing transmission mode combination is output, so as to ensure that the drawing transmission requirements are met as much as possible with the least resources in the drawing transmission process, and the efficiency of drawing transmission and the accuracy of drawing information transmission are guaranteed. The drawings to be transmitted are combined and transmitted by combining the matching drawing transmission modes to achieve efficient and accurate drawing transmission. Compared with a single transmission mode, the combined transmission can be dynamically optimized according to different requirements of drawings, greatly improving the efficiency and accuracy of drawing transmission and reducing the risk of information loss.

[0073] Overall, the embodiments of the present application dynamically adapt to the needs of different types of drawings through intelligent prediction and optimized combination, automatically match the optimal combined transmission method, effectively reduce errors and information loss in transmission, make the entire transmission process more flexible and efficient, and improve the accuracy and consistency of drawing transmission.

[0074] Embodiment 2, as Figure 3 As shown, the embodiment of the present application provides a drawing difference intelligent detection system, the system comprising:

[0075] The transmission mode reading unit 10 is used to obtain a drawing transmission module, wherein the drawing transmission module includes a plurality of drawing transmission modes, and each drawing transmission mode is different from another.

[0076] The drawing information reading unit 20 is used to read information items of the drawing to be transmitted, and the information items of the drawing to be transmitted include drawing geometric features, annotation information and file format information.

[0077] The drawing difference prediction unit 30 is used to cooperate with the drawing transmission module to predict the drawing difference of the information item of the drawing to be transmitted by using a drawing difference prediction model, and output a plurality of drawing difference indicators corresponding to the plurality of drawing transmission modes.

[0078] The combined transmission matching unit 40 is used to perform combined matching of drawing transmission modes according to the multiple drawing difference indicators, output a matching drawing transmission mode combination, and perform combined transmission of the drawings to be transmitted using the matching drawing transmission mode combination.

[0079] Furthermore, the combined transmission matching unit 40 in the embodiment of the present application is further configured to perform the following steps:

[0080] Configure a preset drawing difference index; compare the multiple drawing difference indexes according to the preset drawing difference index, and if there is a drawing transmission method that is smaller than the preset drawing difference index, output a first drawing transmission method, and perform a single transmission on the drawing to be transmitted based on the first drawing transmission method; wherein the first drawing transmission method is a drawing transmission method with the minimum drawing difference index.

[0081] Further, if there is no drawing transmission mode with a difference less than the preset drawing difference index, the combined transmission matching unit 40 performs drawing transmission mode combination matching according to the multiple drawing difference indexes, and the execution steps include:

[0082] According to the multiple drawing transmission modes included in the drawing transmission module, the drawing feature retention capability is identified to generate a retention feature label; the multiple drawing transmission modes are randomly combined for dual transmission to generate multiple dual transmission combination schemes, and multiple feature retention label groups corresponding to the multiple dual transmission combination schemes, wherein the dual transmission combination scheme sample is obtained by combining two different drawing transmission modes; the dual transmission combination scheme sample is optimized by a simulated annealing algorithm to output a matching drawing transmission mode combination.

[0083] Furthermore, the combined transmission matching unit 40 in the embodiment of the present application is further configured to perform the following steps:

[0084] Decompose the multiple drawing difference indicators and output feature difference labels; perform matching calculations on the feature retention label groups corresponding to each dual transmission combination scheme according to the feature difference labels to obtain a matching score for each dual transmission combination scheme; and perform perturbations and updates on the dual transmission combination scheme according to the matching scores using a simulated annealing algorithm until a global optimization result is obtained, which is output as a matching drawing transmission mode combination.

[0085] Furthermore, the combined transmission matching unit 40 in the embodiment of the present application is further configured to perform the following steps:

[0086] Obtain the optimization result of the global optimization, identify whether the matching score of the optimization result is less than the preset matching score, and if it is less than the preset matching score, generate three transmission combination scheme samples; wherein the three transmission combination scheme samples are obtained by combining three different drawing transmission methods; optimize the three transmission combination scheme samples through a simulated annealing algorithm, and output an updated matching drawing transmission method combination.

[0087] Furthermore, the drawing difference prediction unit 30 in the embodiment of the present application is also used to perform the following steps:

[0088] A drawing difference prediction model is constructed, wherein the difference items of the drawing difference prediction model include format conversion difference, dimensioning difference and position offset difference; the drawing difference prediction model predicts the size of the format conversion difference, the size of the dimensioning difference and the size of the position offset difference under the multiple drawing transmission modes according to the information items of the drawings to be transmitted; linear fitting is performed according to the size of the format conversion difference, the size of the dimensioning difference and the size of the position offset difference, and a plurality of drawing difference indicators corresponding to the multiple drawing transmission modes are output.

[0089] Furthermore, the format conversion differences described in the embodiments of the present application include the loss of features of special line types, layer structures, and drawing symbols; the dimension annotation differences include the offset, font changes, and proportional distortion of the annotation text; and the position offset differences include the plane displacement differences and spatial displacement differences that may be caused by different transmission methods.

[0090] Through the above detailed description of the intelligent drawing difference detection method in this specification, those skilled in the art can clearly understand the intelligent drawing difference detection system in this embodiment. For the system disclosed in Example 2, since it corresponds to the method disclosed in Example 1 and has corresponding functional units and beneficial effects, the relevant parts can be referred to the method part description.

[0091] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. The intelligent detection method of drawing differences is characterized by: The method comprises: Acquire a drawing transmission module, wherein the drawing transmission module includes multiple drawing transmission modes, and each drawing transmission mode is different from another; Reading information items of the drawing to be transmitted, wherein the information items of the drawing to be transmitted include drawing geometric features, annotation information, and file format information; In conjunction with the drawing transmission module, a drawing difference prediction model is used to predict drawing differences for the information items of the drawings to be transmitted, and a plurality of drawing difference indicators corresponding to the plurality of drawing transmission modes are output; Perform drawing transmission mode combination matching according to the multiple drawing difference indicators, output a matching drawing transmission mode combination, and perform combined transmission of the drawings to be transmitted using the matching drawing transmission mode combination; The drawing difference prediction model is used to predict the information item of the drawing to be transmitted, and the method includes: Constructing a drawing difference prediction model, wherein the difference items of the drawing difference prediction model include format conversion difference, dimensioning difference and position offset difference; The drawing difference prediction model predicts the format conversion difference, dimensioning difference and position offset difference predicted under the multiple drawing transmission modes according to the information items of the drawings to be transmitted; Linear fitting is performed according to the format conversion difference, dimensioning difference and position offset difference, and a plurality of drawing difference indicators corresponding to the plurality of drawing transmission modes are output.

2. The intelligent drawing difference detection method according to claim 1, characterized in that: Before combining and matching the drawing transmission modes according to the plurality of drawing difference indicators, the method includes: Configure preset drawing difference indicators; Comparing the plurality of drawing difference indicators according to the preset drawing difference indicator, outputting a first drawing transmission method if there is a drawing transmission method that is smaller than the preset drawing difference indicator, and performing a single transmission on the drawing to be transmitted based on the first drawing transmission method; Among them, the first drawing transmission mode is a drawing transmission mode with a minimum drawing difference index.

3. The intelligent drawing difference detection method according to claim 2, characterized in that: If there is no drawing transmission method with a difference index less than the preset drawing difference index, combining and matching drawing transmission methods according to the multiple drawing difference indexes, the method includes: Identify the drawing feature retention capability according to the multiple drawing transmission modes included in the drawing transmission module, and generate a retention feature tag; Performing dual transmission random combinations on the multiple drawing transmission modes to generate multiple dual transmission combination schemes and multiple feature retention label groups corresponding to the multiple dual transmission combination schemes, wherein the dual transmission combination scheme samples are obtained by combining two different drawing transmission modes; The simulated annealing algorithm is used to optimize the dual transmission combination scheme samples and output a matching drawing transmission mode combination.

4. The intelligent drawing difference detection method according to claim 3, characterized in that: The simulated annealing algorithm is used to optimize the dual transmission combination scheme sample and output a matching drawing transmission mode combination, the method comprising: Decomposing the plurality of drawing difference indicators and outputting feature difference labels; According to the feature difference labels, a matching degree calculation is performed on the feature retention label groups corresponding to each dual transmission combination scheme, so as to obtain a matching degree score for each dual transmission combination scheme; The dual transmission combination scheme is disturbed and updated according to the matching score by using a simulated annealing algorithm until a global optimization result is obtained, and the output is a matching drawing transmission mode combination.

5. The intelligent drawing difference detection method according to claim 4, characterized in that: Obtaining an optimization result of the global optimization, identifying whether a matching score of the optimization result is less than a preset matching score, and if it is less than the preset matching score, generating a sample of three transmission combination solutions; Wherein, the three transmission combination scheme samples are obtained by combining three different drawing transmission methods; The simulated annealing algorithm is used to optimize the three transmission combination scheme samples and output an updated matching drawing transmission mode combination.

6. The intelligent drawing difference detection method according to claim 1, characterized in that: The format conversion differences include the loss of features of special line types, layer structures, and drawing symbols; the dimension annotation differences include the offset, font changes, and scale distortion of the annotation text; and the position offset differences include the plane displacement differences and space displacement differences caused by different transmission methods.

7. Drawing difference intelligent detection system, characterized by: The system is used to execute the drawing difference intelligent detection method according to any one of claims 1 to 6, and the system comprises: A transmission mode reading unit, the transmission mode reading unit is used to obtain a drawing transmission module, wherein the drawing transmission module includes a plurality of drawing transmission modes, each drawing transmission mode is different from another; A drawing information reading unit, the drawing information reading unit is used to read information items of the drawing to be transmitted, the information items of the drawing to be transmitted include drawing geometric features, annotation information and file format information; A drawing difference prediction unit, the drawing difference prediction unit is used to cooperate with the drawing transmission module to predict the drawing difference of the information item of the drawing to be transmitted by using a drawing difference prediction model, and output a plurality of drawing difference indicators corresponding to the plurality of drawing transmission modes respectively; A combined transmission matching unit is used to perform combined matching of drawing transmission modes according to the multiple drawing difference indicators, output a matching drawing transmission mode combination, and perform combined transmission of the drawings to be transmitted using the matching drawing transmission mode combination.

Images