Measurement accuracy assessment method based on big data

Abstract

The invention discloses a measurement accuracy evaluation method based on big data, which is realized based on a control terminal and intelligent scanning equipment, and comprises the following specific steps: S1, the control terminal sends a scanning starting instruction to the intelligent scanning equipment, and sets scanning parameters at the same time; s2, the intelligent scanning device receives the scanning starting instruction, starts to execute a scanning task, and obtains three-dimensional point cloud data of the object; and S3, the intelligent scanning equipment transmits the scanned three-dimensional point cloud data to the control terminal. Through a big data technology and an advanced algorithm model, deep mining and accurate analysis of a large amount of measurement data can be realized, interference of human factors is effectively avoided, the accuracy and reliability of an evaluation result are improved, three-dimensional point cloud data transmitted by intelligent scanning equipment can be received and processed in real time or at regular time, and the accuracy and reliability of the evaluation result are improved. The method greatly shortens the evaluation period, improves the evaluation efficiency, and reduces the operation difficulty and labor cost through an integrated software interface and an intelligent data processing algorithm.

Description

Technical Field

[0001] The present application relates to the field of measurement and evaluation technology, and in particular to a measurement accuracy evaluation method based on big data. Background Art

[0002] Object measurement is a crucial link in many fields such as manufacturing, quality inspection, and product design. Measurement accuracy, as a key indicator for evaluating the consistency between measurement results and actual values, is directly related to product quality, cost control, and the reliability of the production process. With the advancement of science and technology, especially the development of intelligent manufacturing and big data technology, the accuracy requirements for object measurement are increasing. Therefore, how to scientifically and efficiently evaluate measurement accuracy has become a problem that needs to be urgently solved in the current field of measurement technology.

[0003] However, existing measurement accuracy assessment methods still have many defects:

[0004] On the one hand, traditional methods mostly rely on manual measurement and manual data analysis, which is not only time-consuming and labor-intensive, but also easily affected by human factors, which greatly reduces the accuracy and reliability of the evaluation results.

[0005] On the other hand, although some automated evaluation methods have improved evaluation efficiency, due to the simplicity of the algorithm model and insufficient data processing capabilities, it is difficult to fully explore and utilize the potential information in big data, and the accuracy and comprehensiveness of the evaluation results still need to be improved;

[0006] Therefore, there is an urgent need for a more scientific, efficient and accurate measurement accuracy evaluation method based on big data to overcome the shortcomings of existing technologies. Summary of the invention

[0007] To this end, the present application provides a measurement accuracy evaluation method based on big data to solve the problems that the accuracy and reliability of evaluation results of traditional methods in the prior art are insufficient, and the accuracy and comprehensiveness of evaluation results of some automated evaluation methods still need to be improved.

[0008] In order to achieve the above objectives, this application provides the following technical solutions:

[0009] A measurement accuracy evaluation method based on big data is implemented based on a control terminal and an intelligent scanning device. The specific steps are as follows:

[0010] S1 control terminal sends scanning start instruction to intelligent scanning device and sets scanning parameters at the same time;

[0011] The S2 intelligent scanning device receives the scanning start command, starts to execute the scanning task, and obtains the 3D point cloud data of the object;

[0012] The S3 intelligent scanning device transmits the scanned 3D point cloud data to the control terminal;

[0013] The S4 control terminal cleans the received 3D point cloud data, automatically aligns the 3D point cloud data with the original design model of the object, and creates a deviation color map based on the aligned 3D point cloud data and design model;

[0014] The S5 control terminal measures the key dimensions and tolerances of the object according to the deviation color map and compares them with the theoretical values ​​in the design model to obtain deviation data;

[0015] S6 conducts in-depth analysis of deviation data, dimension data, and tolerance data to evaluate the measurement accuracy of the intelligent scanning device and generate an evaluation report.

[0016] Optionally, in step S1, when generating instructions, the control terminal needs to set a series of scanning parameters according to the characteristics and requirements of the scanned object. These parameters include scanning accuracy, scanning speed and scanning range.

[0017] Optionally, in step S2, the scanning task includes moving the scanning head to cover the entire scanning area, adjusting the intensity and angle of the light source to obtain the best scanning effect, and processing the sensor data in real time to generate three-dimensional point cloud data.

[0018] Optionally, in step S4, data cleaning includes removing useless data such as noise points and duplicate points and filling in missing points.

[0019] Optionally, in step S4, before data alignment, the control terminal extracts feature points from the three-dimensional point cloud data and the design model, and uses a matching algorithm to match, optimize and adjust the feature points in the three-dimensional point cloud data with the feature points in the design model.

[0020] Optionally, in step S5, the control terminal is installed with 3D detection software, and the 3D point cloud data is aligned with the CAD design model. Once the alignment is completed, the software will generate a deviation color map based on the difference between the point cloud data and the CAD model, and obtain the deviation data by identifying the deviation color map.

[0021] Optionally, in step S6, the deviation data, dimension data and tolerance data obtained from multiple measurements are imported into a big data analysis platform, and the measurement data are deeply mined and compared and analyzed using big data analysis technology to evaluate the measurement accuracy of the intelligent scanning device.

[0022] Optionally, the intelligent scanning device is connected to the control terminal via a wired or wireless connection to receive instructions from the control terminal.

[0023] Compared with the prior art, this application has at least the following beneficial effects:

[0024] (i) Through big data technology and advanced algorithm models, this method can achieve in-depth mining and precise analysis of a large amount of measurement data, effectively avoiding the interference of human factors and improving the accuracy and reliability of the evaluation results;

[0025] (ii) This method adopts an automated evaluation process, which can receive and process the three-dimensional point cloud data transmitted by the intelligent scanning equipment in real time or at a fixed time, greatly shortening the evaluation cycle and improving the evaluation efficiency. At the same time, through the integrated software interface and intelligent data processing algorithm, it reduces the difficulty of operation and labor costs;

[0026] (III) In the evaluation process, this method fully considers the impact of the performance differences of the measuring equipment itself and the changes in the measuring environment on the accuracy. By optimizing the algorithm model and adjusting the scanning parameters, a comprehensive and objective evaluation of the measurement accuracy is achieved;

[0027] (IV) This method not only provides the evaluation results of measurement accuracy, but also optimizes the algorithm model of the intelligent scanning device according to the evaluation results, thereby improving the measurement accuracy and stability. At the same time, by collecting user feedback and usage experience, it continuously iterates and optimizes product performance to meet the needs of different application scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more intuitively illustrate the prior art and the present application, exemplary drawings are given below. It should be understood that the specific shapes and structures shown in the drawings should not generally be regarded as limiting conditions for implementing the present application; for example, those skilled in the art are capable of easily making conventional adjustments or further optimizations to the addition / reduction / attribution division, specific shapes, positional relationships, connection methods, dimensional ratios, etc. of certain units (components) based on the technical concepts and exemplary drawings disclosed in the present application.

[0029] Figure 1 A schematic diagram of the flow chart of the measurement accuracy evaluation method based on big data provided in this application. DETAILED DESCRIPTION

[0030] The present application is further described below in detail through specific embodiments in conjunction with the accompanying drawings.

[0031] like Figure 1 As shown, a measurement accuracy evaluation method based on big data is implemented based on a control terminal and an intelligent scanning device. The method is as follows:

[0032] The control terminal sends a scan start command to the intelligent scanning device and sets the scanning parameters (such as scanning accuracy, scanning speed, scanning range, etc.). After receiving the scan start command, the intelligent scanning device starts to execute the scanning task, obtains the 3D point cloud data of the object, and transmits the scanned 3D point cloud data to the control terminal in real time or at a fixed time for subsequent processing and analysis. The specific contents are as follows:

[0033] 1. On a control terminal (such as a computer, tablet or smart phone), the user or operator enters scanning instructions through a dedicated software interface. These instructions include the start and end of scanning, the definition of the scanning area, the setting of scanning accuracy, etc.;

[0034] 2. When generating instructions, the user needs to set a series of scanning parameters according to the characteristics and requirements of the scanned object, including but not limited to scanning accuracy, scanning speed, and scanning range;

[0035] 3. When the command and parameter setting are completed, the control terminal sends the command to the intelligent scanning device through a wired or wireless connection (such as USB, Wi-Fi, Bluetooth, etc.);

[0036] 4. After receiving the instructions and parameters from the control terminal, the built-in microprocessor or controller of the intelligent scanning device will parse and verify to ensure the correctness and feasibility of the instructions;

[0037] 5. Perform a series of initialization operations before formal scanning, such as calibrating the sensor, preheating the light source (for devices using structured light or laser scanning), etc. Then, according to the received instructions and parameters, the intelligent scanning device starts to perform the scanning task;

[0038] Scanning tasks include moving the scan head to cover the entire scanning area, adjusting the light source intensity and angle to obtain the best scanning effect, and processing sensor data in real time to generate a 3D point cloud.

[0039] 6. During the scanning process, the intelligent scanning device continuously records the acquired 3D point cloud data and stores the data in the internal memory or external storage device in a specific file format (such as PLY, OBJ, STL, etc.);

[0040] Once the scan is completed, the intelligent scanning device (a high-precision, high-stability 3D scanner or intelligent photogrammetry device) will initially organize and process the recorded data, such as removing noise points and filling in missing data, to improve the quality and availability of the data. A data transmission protocol will be established between the intelligent scanning device and the control terminal to ensure accurate, fast and reliable data transmission, involving mechanisms such as data compression, encryption, and verification. Then, the intelligent scanning device will transmit the processed 3D point cloud data to the control terminal in real time or at a fixed time according to the protocol.

[0041] Real-time transmission is suitable for application scenarios that require immediate feedback, while scheduled transmission is suitable for situations where the amount of data is large or the transmission speed is limited;

[0042] After receiving the data, the control terminal will perform operations such as decompression, decryption and verification to ensure the integrity and accuracy of the data. Then, the data will be stored in a designated location for subsequent processing and analysis;

[0043] The control terminal cleans the received 3D point cloud data, removes useless data such as noise points and duplicate points, improves data quality, and automatically aligns the scanned 3D point cloud data with the original CAD design model of the object to ensure the spatial consistency between the two.

[0044] 1. Data cleaning is the first step of data preprocessing, which aims to remove useless or erroneous information in the received 3D point cloud data to improve the accuracy and efficiency of subsequent analysis. In the 3D scanning process, due to various factors (such as equipment accuracy, environmental interference, object surface characteristics, etc.), noise points, duplicate points, missing points and other problems may occur. Therefore, data cleaning becomes particularly important. The specific contents of data cleaning are as follows:

[0045] 1. Noise points are usually caused by random errors or environmental interference during the scanning process. These points usually have significant deviations from the surrounding point cloud data and can be removed by statistical filtering, radius filtering, etc.

[0046] Statistical filtering analyzes the distance distribution between each point and the surrounding points, and removes the points beyond a certain range as noise points. Radius filtering determines whether each point is a noise point based on the number of neighboring points within a certain radius.

[0047] 2. Duplicate points are generated due to overlapping scans or insufficient equipment accuracy during scanning. These points usually have the same coordinates or very close coordinates. Algorithms such as clustering algorithms or spatial hash tables can be used to detect and remove duplicate points.

[0048] The clustering algorithm divides the point cloud data into multiple clusters, and the points in each cluster are considered to be duplicates; the spatial hash table maps the point cloud data into a hash table through a hash function, thereby quickly detecting duplicate points.

[0049] 3. In some cases, due to the surface characteristics of the object (such as reflective, transparent, etc.) or the limitations of the scanning equipment (such as scanning angle, scanning distance, etc.), data in some areas may be missing, and the missing points need to be filled. This can be done through interpolation algorithms (such as linear interpolation, nearest neighbor interpolation, etc.) or machine learning-based methods;

[0050] The interpolation algorithm estimates the coordinates and attribute values ​​of the missing points based on the coordinates and attribute values ​​of the known points; the machine learning-based method predicts the attribute values ​​of the missing points by training the model;

[0051] 2. Data alignment is the process of automatically aligning the scanned 3D point cloud data with the original CAD design model of the object. The purpose of alignment is to ensure the spatial consistency of the two so as to carry out subsequent deviation analysis, dimension measurement and other tasks. The specific contents are as follows:

[0052] 1. Before data alignment, feature points or feature lines need to be extracted from 3D point cloud data and CAD design models. These feature points or feature lines usually have significant geometric features (such as corners, edges, etc.) or semantic features (such as holes, slots, etc.). Feature extraction can be achieved through geometry-based methods (such as corner detection, edge detection, etc.) or machine learning-based methods (such as deep learning networks);

[0053] 2. After extracting features, a matching algorithm is needed to match the feature points in the 3D point cloud data with the feature points in the CAD design model. The matching algorithm can be divided into geometry-based matching algorithms (such as ICP algorithm, NDT algorithm, etc.) and feature-based matching algorithms (such as SIFT, SURF, etc.);

[0054] The ICP algorithm minimizes the error between point cloud data and the model by iterating the closest point;

[0055] The NDT algorithm converts point cloud data into a probability density function and finds the best alignment parameters by optimizing the objective function.

[0056] Feature-based matching algorithms find matching points by comparing the descriptors of feature points;

[0057] 3. After the matching is completed, optimization and adjustment are required to ensure the accuracy and stability of the alignment. Optimization can minimize the error by adjusting the alignment parameters (such as rotation matrix, translation vector, etc.), and adjustment can improve the alignment results through manual intervention or the introduction of other constraints. In some cases, multiple iterations and optimizations may be required to obtain a satisfactory alignment result.

[0058] The control terminal (installed with 3D detection software) creates a deviation color map based on the aligned 3D point cloud data and CAD design model. Different colors in the map represent different deviation ranges, thereby intuitively displaying the shape deviation of the object.

[0059] Based on the deviation color map, measure the critical dimensions and tolerances (GD&T) of the object and compare them with the theoretical values ​​in the CAD design model;

[0060] Import the measured deviation data, dimension data, tolerance data, etc. into the big data analysis platform for in-depth mining and analysis. By comparing and analyzing the measurement data of different objects, the measurement accuracy of the intelligent scanning equipment can be evaluated.

[0061] According to the results of big data analysis, the object shape measurement accuracy assessment report is automatically generated. The report contains information such as statistical analysis of measurement data, deviation distribution diagram, and measurement accuracy assessment results;

[0062] The specific contents are as follows:

[0063] 1. Create a deviation color map

[0064] First, you need to choose a professional 3D inspection software that has the functions of processing 3D point cloud data, aligning with CAD design models, and generating deviation color maps. Then import the pre-processed 3D point cloud data and CAD design model into the 3D inspection software to ensure that the data format is compatible so that the software can read and process it correctly.

[0065] Then, the software aligns the 3D point cloud data with the CAD design model. Once the alignment is completed, the software will generate a deviation color map based on the differences between the point cloud data and the CAD model. Different colors in the map represent different deviation ranges. For example, red may indicate a positive deviation (exceeding the design size), blue indicates a negative deviation (less than the design size), and green may indicate that the deviation is within an acceptable range.

[0066] Finally, by identifying the deviation color map, we can intuitively understand the shape deviation of the object. The color distribution, deviation range, and deviation concentration area in the map are of great significance for subsequent dimensional measurement and tolerance analysis.

[0067] 2. Measuring dimensions and tolerances

[0068] First, determine the key dimensions that need to be measured based on the design requirements and actual application scenarios of the object. These dimensions are usually closely related to the function, performance, and assembly requirements of the object.

[0069] Then, according to the design requirements, set a reasonable tolerance range for each key dimension. The tolerance range should take into account factors such as errors in the manufacturing process, material deformation, and the accuracy of the measuring equipment;

[0070] Based on the deviation color map, the key dimensions of the object are measured using 3D detection software or dedicated measurement tools to ensure the accuracy and repeatability of the measurement process;

[0071] Finally, the measured dimension data is compared with the theoretical value in the CAD design model to check whether the dimension is within the set tolerance range. For dimensions that exceed the tolerance, the cause needs to be further analyzed and corresponding corrective measures taken;

[0072] 3. Big Data Analysis

[0073] First, the deviation data, dimension data, and tolerance data obtained from multiple measurements are imported into the big data analysis platform to ensure the integrity and accuracy of the data, and the collected data is cleaned, sorted, and standardized for subsequent analysis and mining;

[0074] Then, use big data analysis techniques (such as cluster analysis, association rule mining, trend prediction, etc.) to conduct in-depth mining of the measurement data to discover hidden patterns, associations, and trend changes in the data;

[0075] Finally, the measurement data of different objects are compared and analyzed to evaluate the measurement accuracy of the intelligent scanning equipment. By comparing the measurement data of different batches, different models or different manufacturing conditions, the stability, consistency and limitations of the equipment performance can be found.

[0076] 4. Generate an evaluation report

[0077] First, according to the evaluation requirements, design an evaluation report template that includes information such as measurement data statistical analysis, deviation distribution diagram, and measurement accuracy evaluation results;

[0078] Then, fill the mining results, comparative analysis results and other information obtained from the big data analysis platform into the report template. At the same time, conduct in-depth analysis and interpretation of the data in the report to provide valuable insights and suggestions;

[0079] Finally, after the report is generated, it is carefully reviewed and proofread to ensure the accuracy and completeness of the report content, and the report is issued to relevant personnel or departments so that they can understand the measurement accuracy of the intelligent scanning equipment and take corresponding improvement measures;

[0080] Based on the evaluation results, the control terminal optimizes the algorithm model of the intelligent scanning device to improve the measurement accuracy and stability. According to the actual application requirements, the control terminal adjusts the scanning parameters of the intelligent scanning device (such as scanning accuracy, scanning speed, etc.) to achieve the best measurement effect. The control terminal also incorporates the evaluation results and user feedback into the iterative development process of the intelligent scanning device to continuously improve and optimize product performance. The specific contents are as follows:

[0081] 1. Model Optimization

[0082] First, we need to conduct an in-depth analysis of the algorithm models currently used by smart scanning devices, including understanding the model's architecture, working principles, and performance in different scenarios, and identify potential problems and room for improvement in the model through analysis.

[0083] In order to improve the generalization ability of the model, the training data can be enhanced, including increasing the diversity of the data, simulating the data changes under different scanning conditions, and introducing noise data, which will help the model better adapt to various actual scanning scenarios and improve the measurement accuracy and robustness;

[0084] Based on the results of algorithm analysis and data enhancement, the model is adjusted and optimized, which involves changing the model's parameter settings, adding new layers or nodes, introducing regularization techniques to reduce overfitting, etc. The adjusted model needs to be fully tested and verified to ensure its performance improvement;

[0085] 2. Parameter adjustment

[0086] Adjust the scanning accuracy of the intelligent scanning device according to the evaluation results and user needs. For scenes that require high-precision measurement, the density and number of scanning points can be increased. For scenes that do not require high accuracy, the scanning accuracy can be reduced to increase the scanning speed.

[0087] Under the premise of ensuring measurement accuracy, optimizing scanning speed is the key to improving the performance of intelligent scanning equipment. This can be achieved by improving scanning algorithms, increasing data processing speed, and optimizing equipment hardware. The optimized equipment can complete scanning tasks in a shorter time and improve work efficiency.

[0088] In addition to scanning accuracy and speed, other scanning parameters can be adjusted according to actual needs, such as scanning range, light source intensity, exposure time, etc. The adjustment of these parameters helps the device better adapt to different scanning objects and scenes, and improves the accuracy and stability of measurement;

[0089] 3. Feedback and iteration

[0090] Actively collect user feedback and usage experience, which can be achieved through questionnaires, user interviews, online reviews, etc. User feedback is an important way to understand device performance and user needs, and helps to discover potential problems and improvement directions;

[0091] Conduct in-depth analysis of user feedback, identify the root causes of the problems and propose solutions;

[0092] Incorporate evaluation results and user feedback into the iterative development process of smart scanning equipment. Through continuous iteration and optimization, gradually improve the performance and quality of the equipment. During the iterative development process, it is necessary to maintain communication and exchanges with users in order to timely understand user needs and market changes, and provide strong support for continuous product improvement.

[0093] In summary, the present invention can realize in-depth mining and precise analysis of a large amount of measurement data through big data technology and advanced algorithm models, effectively avoid the interference of human factors, and improve the accuracy and reliability of evaluation results; adopt an automated evaluation process, and can receive and process three-dimensional point cloud data transmitted by intelligent scanning equipment in real time or at a scheduled time, which greatly shortens the evaluation cycle and improves the evaluation efficiency. At the same time, through the integrated software interface and intelligent data processing algorithm, the operation difficulty and labor cost are reduced; the performance differences of the measuring equipment itself and the impact of changes in the measuring environment on the accuracy are fully considered, and a comprehensive and objective evaluation of the measurement accuracy is achieved by optimizing the algorithm model and adjusting the scanning parameters; not only the evaluation results of the measurement accuracy are provided, but also the algorithm model of the intelligent scanning equipment is optimized according to the evaluation results, thereby improving the measurement accuracy and stability. At the same time, by collecting user feedback and usage experience, the product performance is continuously iterated and optimized to meet the needs of different application scenarios.

[0094] The technical features of the above embodiments may be arbitrarily combined (as long as there is no contradiction in the combination of these technical features). To make the description concise, not all possible combinations of the technical features in the above embodiments are described; these embodiments that are not explicitly written should also be considered to be within the scope of this specification.

Claims

1. A measurement accuracy evaluation method based on big data, implemented based on a control terminal and an intelligent scanning device, characterized in that: The specific steps are as follows: S1 control terminal sends scanning start instruction to intelligent scanning device and sets scanning parameters at the same time; The S2 intelligent scanning device receives the scanning start command, starts to execute the scanning task, and obtains the 3D point cloud data of the object; The S3 intelligent scanning device transmits the scanned 3D point cloud data to the control terminal; The S4 control terminal cleans the received 3D point cloud data, automatically aligns the 3D point cloud data with the original design model of the object, and creates a deviation color map based on the aligned 3D point cloud data and design model; The S5 control terminal measures the key dimensions and tolerances of the object according to the deviation color map and compares them with the theoretical values ​​in the design model to obtain deviation data; S6 conducts in-depth analysis of deviation data, dimension data, and tolerance data to evaluate the measurement accuracy of the intelligent scanning device and generate an evaluation report.

2. The measurement accuracy evaluation method based on big data according to claim 1, characterized in that: In step S1, when generating instructions, the control terminal needs to set a series of scanning parameters according to the characteristics and requirements of the scanned object. These parameters include scanning accuracy, scanning speed and scanning range.

3. The measurement accuracy evaluation method based on big data according to claim 1, characterized in that: In step S2, the scanning task includes moving the scanning head to cover the entire scanning area, adjusting the light source intensity and angle to obtain the best scanning effect, and processing the sensor data in real time to generate three-dimensional point cloud data.

4. The measurement accuracy evaluation method based on big data according to claim 1, characterized in that: In step S4, data cleaning includes removing useless data such as noise points and duplicate points and filling in missing points.

5. The measurement accuracy evaluation method based on big data according to claim 1, characterized in that: In step S4, before data alignment, the control terminal extracts feature points from the three-dimensional point cloud data and the design model, and uses a matching algorithm to match, optimize and adjust the feature points in the three-dimensional point cloud data with the feature points in the design model.

6. The measurement accuracy evaluation method based on big data according to claim 1, characterized in that: In step S5, the control terminal is installed with 3D detection software, and the 3D point cloud data is aligned with the CAD design model. Once the alignment is completed, the software will generate a deviation color map based on the difference between the point cloud data and the CAD model, and obtain the deviation data by identifying the deviation color map.

7. The method for evaluating measurement accuracy based on big data according to claim 1, characterized in that: In step S6, the deviation data, dimension data and tolerance data obtained from multiple measurements are imported into the big data analysis platform, and the measurement data are deeply mined and compared using big data analysis technology to evaluate the measurement accuracy of the intelligent scanning equipment.

8. The method for evaluating measurement accuracy based on big data according to claim 1, characterized in that: The intelligent scanning device is connected to the control terminal via wired or wireless communication to receive instructions from the control terminal.

Images