An image matching method and system for processing data images

Abstract

The application discloses a kind of image matching's mode processing data image's interpretation method and system, including automatically segmenting actuator data, obtaining actuator displacement angle data and actuator action angle data;Displacement angle scatter diagram and action angle scatter diagram are respectively drawn;Displacement angle scatter diagram and action angle scatter diagram are respectively transformed into displacement servo curve and action servo curve;Displacement servo curve and action servo curve are grayed, obtain gray scale diagram;Gray scale diagram is binarized, and the correlation matching of normalized image is carried out, according to several normalized matching results, determine relevant peak threshold;If normalized matching result is greater than relevant peak threshold, then actuator displacement angle data and actuator action angle data registration is successful, and composite image is drawn;If less than relevant peak threshold, then registration is not successful, composite image is drawn, and troubleshooting is carried out.The man-power, time consumption when mass production is avoided, and the correctness of interpretation is guaranteed.

Description

TECHNICAL FIELD

[0001] The present application relates to a method and system for processing data interpretation in a manner of image matching, and belongs to the technical field of data analysis. BACKGROUND

[0002] When products are batch-produced, the traditional manual data analysis method cannot meet the requirements of batch production. In the traditional manual data analysis, whether the actuator displacement (angle) is consistent with the actuator action angle is mainly determined by manually drawing a graph, and the person analyzes the relevant information to finally determine whether the angles are consistent. The workload is large, and the personnel experience is high, which cannot meet the progress requirements under the background of mass production of models. In the continuous work, it is easy to cause misjudgment and wrong judgment. Therefore, it is necessary to introduce an automatic interpretation method.

[0003] In the development process of automatic interpretation, the interpretation of actuator action angle needs to be solved, and the command and response of the actuator are compared. Since the command is fixed data and the response is actual feedback data, comparison from the data cannot obtain the correct result. Therefore, people generally use manual drawing to determine the conclusion by visual judgment. However, the manual drawing tool cannot be embedded into the automatic interpretation software, and therefore an automatic interpretation method that can be programmed needs to be found. SUMMARY

[0004] The technical problem of the present application is to overcome the shortcomings of the prior art, and to provide a method and system for processing data interpretation in a manner of image matching, which can avoid the consumption of manpower and time in mass production, ensure the correctness of interpretation, and improve the efficiency.

[0005] The technical solution of the present application is:

[0006] The present application discloses a method for processing data interpretation in a manner of image matching, comprising:

[0007] Step 1: automatically segmenting the actuator data to obtain actuator displacement angle data and actuator action angle data;

[0008] Step 2: drawing a displacement angle scatter plot and an action angle scatter plot according to the actuator displacement angle data and the actuator action angle data, respectively;

[0009] Step 3: transforming the displacement angle scatter plot and the action angle scatter plot into a displacement servo curve and an action servo curve, respectively;

[0010] Step 4: performing gray-scale processing on the displacement servo curve and the action servo curve to obtain a servo actuator gray-scale image;

[0011] Step 5: performing binaryzation on the servo actuator gray-scale image to obtain a servo actuator binary image.

[0012] Step 6: Normalize the servo actuator binary image to obtain a normalized matching result;

[0013] Step 7: Determine a correlation peak threshold value according to a plurality of normalized matching results;

[0014] Step 8: Perform threshold judgment on the correlation peak. If the normalized matching result is greater than the correlation peak threshold value, the actuator displacement angle data and actuator action angle data are successfully registered, and a composite image is drawn. If it is less than the correlation peak threshold value, the registration is unsuccessful, a composite image is drawn, and troubleshooting is performed.

[0015] In the above interpretation method, the actuator data after software analysis is automatically segmented to obtain actuator displacement angle data and actuator action angle data. The specific method is to intercept the effective data from the ignition time to the flight end power-off time.

[0016] In the above interpretation method, the angle scatter plot is an image without coordinate axes.

[0017] In the above interpretation method, the displacement angle scatter plot and the action angle scatter plot are enlarged by the same scale, and the BGR mode of the displacement angle scatter plot and the action angle scatter plot is converted to RGB mode.

[0018] In the above interpretation method, the servo actuator binary image is normalized to obtain a normalized correlation coefficient matching result. The specific method is:

[0019]

[0020] Where T is the template image, I is the image to be matched, (x, y) is the x-direction distance and y-direction distance of the upper left corner point of the template from the upper left corner point of the image, and x' and y' are the coordinates of the template moving with (x, y) as the origin.

[0021] In the above interpretation method, a correlation peak threshold value is determined according to a plurality of normalized matching results. The specific method is:

[0022] The matching mean and standard deviation are obtained from the normalized matching results of a plurality of test images.

[0023] The difference between the mean and n times the standard deviation is taken as the threshold value, where n is an integer greater than 1.

[0024] In the above interpretation method, the matching mean and standard deviation are obtained from the normalized matching results of a plurality of test images. The specific method is:

[0025]

[0026] Among them, X i Here, n represents the number of relevant peak values.

[0027]

[0028] Where S is the standard deviation, X i Here, n represents the number of relevant peak values. This is the mean.

[0029] In the above interpretation method, the threshold value is the difference between the mean and n times the standard deviation. The specific method is as follows:

[0030]

[0031] Where X is the final threshold obtained. S is the mean, and S is the standard deviation.

[0032] In the above interpretation method, the composite image is drawn using the following method:

[0033] The initial and final positions of the image plotted from the actuator displacement angle data and the image plotted from the actuator motion angle data are aligned according to time points to generate a composite image.

[0034] In the above interpretation method, n = 2 or 3.

[0035] This invention discloses a data interpretation system that processes data using image matching, comprising: a data acquisition module, an image preprocessing module, and an image matching module, wherein:

[0036] Data acquisition module: automatically segments actuator data to obtain actuator displacement angle data and actuator motion angle data; sends the actuator displacement angle data and actuator motion angle data to the image preprocessing module;

[0037] Image preprocessing module: Based on the actuator displacement angle data and actuator motion angle data sent by the data acquisition module, it plots displacement angle scatter plots and motion angle scatter plots respectively; it transforms the displacement angle scatter plots and motion angle scatter plots into displacement servo curves and motion servo curves respectively; it converts the displacement servo curves and motion servo curves into grayscale to obtain a grayscale image of the servo actuator; it binarizes the grayscale image of the servo actuator to obtain a binarized image of the servo actuator; and it sends the binarized image of the servo actuator to the image matching module.

[0038] Image matching module: Performs normalized correlation matching on the binarized image of the servo actuator sent by the image preprocessing module to obtain a normalized matching result; determines a correlation peak threshold based on several of the normalized matching results; performs threshold judgment on the correlation peak, if the normalized matching result is greater than the correlation peak threshold, the actuator displacement angle data and actuator motion angle data are successfully registered, and a composite image is drawn; if it is less than the correlation peak threshold, the registration is unsuccessful, a composite image is drawn, and troubleshooting is performed.

[0039] The advantages of this invention over the prior art are as follows:

[0040] (1) This invention solves the problem that the traditional manual data analysis method cannot meet the batch production requirements when the batch production task is large, by automatically interpreting the servo actuation data. It avoids the problem of large consumption of manpower and time during mass production, and at the same time ensures the accuracy of interpretation to a certain extent.

[0041] (2) This invention preprocesses the servo actuation data, plots the data as a curve image, and then solves the image preparation work before servo actuation curve registration by removing the coordinate axis plotting, scaling up, converting BGR mode to RGB mode, grayscale, binarization and other image processing methods, thereby improving efficiency.

[0042] (3) This invention solves the image threshold problem in automatic registration by matching the correlation coefficient of the image data after normalization and then determining the correlation peak threshold by the statistical Chebyshev rule, thus ensuring the best registration effect without distortion.

[0043] (4) The present invention can dynamically adjust the matching method of images to achieve the best registration effect for different needs, thus solving the problem of the extensibility of the invention. Attached Figure Description

[0044] Figure 1 This is a flowchart of the image matching method for interpreting data images according to the present invention.

[0045] Figure 2 This invention is used to register the displacement angle data curve image of the actuator before the invention.

[0046] Figure 3 The image shows the motion angle data curve of the actuator before registration in this invention. Detailed Implementation

[0047] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0048] This invention provides a method for interpreting data processed by image matching, comprising:

[0049] The actuator data parsed by the software is automatically segmented to obtain actuator displacement angle data and actuator motion angle data; valid data from the ignition moment to the end of flight and power-off moment is extracted.

[0050] Based on the actuator displacement angle data and actuator motion angle data, scatter plots of displacement angle and motion angle are plotted respectively; the angle scatter plots are images with the coordinate axes removed. The displacement angle scatter plots and motion angle scatter plots are enlarged proportionally, and the BGR mode of the displacement angle scatter plots and motion angle scatter plots are converted to RGB mode.

[0051] Transform the displacement angle scatter plot and motion angle scatter plot into displacement servo curves and motion servo curves, respectively.

[0052] The displacement servo curve and motion servo curve are converted to grayscale to obtain a grayscale image of the servo actuator.

[0053] The grayscale image of the servo actuator is binarized to obtain a binarized image of the servo actuator.

[0054] The binarized image of the servo actuator is normalized for correlation matching to obtain the normalized matching result. The specific method is as follows:

[0055]

[0056] Where T is the template image, I is the image to be matched, (x,y) is the x-direction distance and y-direction distance of the top left corner of the template from the top left corner of the image, and x' and y' are the coordinates of the template moving with (x,y) as the origin.

[0057] Based on several normalized matching results, the relevant peak thresholds are determined;

[0058] A threshold judgment is performed on the relevant peaks. If the normalized matching result is greater than the relevant peak threshold, the actuator displacement angle data and actuator motion angle data are successfully registered, and a composite image is drawn. If it is less than the relevant peak threshold, the registration is unsuccessful, a composite image is drawn, and troubleshooting is performed. The initial and ending positions of the image drawn from the actuator displacement angle data and the image drawn from the actuator motion angle data are aligned according to the time points to generate a composite image.

[0059] The mean and standard deviation of the matching were obtained by normalizing the matching results of several experimental images. The specific method is as follows:

[0060]

[0061] Among them, X i Here, n represents the number of relevant peak values.

[0062]

[0063] Where S is the standard deviation, X i Here, n represents the number of relevant peak values. This is the mean.

[0064] The threshold is determined by taking the difference between the mean and n times the standard deviation. The specific method is as follows:

[0065] Where X is the final threshold obtained. Let S be the mean and S be the standard deviation. Here, n is an integer greater than 1, where n = 2 or 3.

[0066] This invention discloses a data interpretation system that processes data using image matching, comprising: a data acquisition module, an image preprocessing module, and an image matching module, wherein:

[0067] Data acquisition module: automatically segments actuator data to obtain actuator displacement angle data and actuator motion angle data; sends the actuator displacement angle data and actuator motion angle data to the image preprocessing module;

[0068] Image preprocessing module: Based on the actuator displacement angle data and actuator motion angle data sent by the data acquisition module, it plots displacement angle scatter plots and motion angle scatter plots respectively; it transforms the displacement angle scatter plots and motion angle scatter plots into displacement servo curves and motion servo curves respectively; it converts the displacement servo curves and motion servo curves into grayscale to obtain a grayscale image of the servo actuator; it binarizes the grayscale image of the servo actuator to obtain a binarized image of the servo actuator; and it sends the binarized image of the servo actuator to the image matching module.

[0069] Image matching module: Performs normalized correlation matching on the binarized image of the servo actuator sent by the image preprocessing module to obtain a normalized matching result; determines the correlation peak threshold based on several normalized matching results; performs threshold judgment on the correlation peak. If the normalized matching result is greater than the correlation peak threshold, the actuator displacement angle data and actuator motion angle data are successfully registered, and a composite image is drawn; if it is less than the correlation peak threshold, the registration is unsuccessful, a composite image is drawn, and troubleshooting is performed.

[0070] Example 1

[0071] like Figure 1 As shown, this embodiment provides a method for interpreting data images by image matching, comprising the following steps:

[0072] 1. Automatically segment the actuator data after software parsing, and extract the useful data from the total data.

[0073] The specific method is as follows: Since the data parsed by the analysis software is total data, it is necessary to first automatically process the total data and extract the valid registration data. If the data is not extracted correctly, it will cause curve errors. The data is extracted according to the moment when the flight time is 0, specifically the data from the ignition moment (when the flight time is 0) to the moment when the power is cut off at the end of the flight.

[0074] 2. Based on the actuator displacement angle data and actuator motion angle data, draw an angle scatter plot.

[0075] The specific method is as follows: use the matplotlib and numpy libraries in Python to draw the scatter plot of the servo actuation curve. The axis function is used to remove the coordinate axis of the scatter plot, otherwise the coordinate axis will have a huge impact on the matching of the curve graph.

[0076] 3. Enlarge the two images proportionally (keeping the enlargement level consistent).

[0077] The specific method is to use the matplotlib library and the figure function of the numpy library in Python to enlarge the image.

[0078] 4. Convert the two images from BGR mode to RGB mode.

[0079] The specific method is to convert the image from BGR mode to RGB mode using the cvtColor function of Python's OpenCV library.

[0080] 5. Convert the servo curve image to grayscale to obtain the grayscale image of the servo actuator.

[0081] The specific method is as follows: the image grayscale mode is converted using the cvtColor function of the Python OpenCV library, which uses...

[0082] 6. Binarize the grayscale image to obtain the binarized image of the servo actuator.

[0083] The specific method is as follows: Binarization of the grayscale image is performed using the `threshold` function from Python's OpenCV library, employing the Otsu thresholding method. For example... Figure 2 and 3 As shown.

[0084] 7. Perform normalized correlation coefficient matching on the grayscale image of the servo actuator to obtain the normalized matching result.

[0085] The specific method is as follows:

[0086]

[0087] Where T represents the template image, I represents the image to be matched, (x,y) represents the x-direction distance and y-direction distance of the top left corner of the template from the top left corner of the image, and x' and y' are the coordinates of the template moving with (x,y) as the origin.

[0088] 8. Determine the relevant peak threshold based on several normalized matching results.

[0089] The specific method is as follows: the registration mean and standard deviation are obtained through the registration results of a large number of various experimental images. According to Chebyshev's rule in statistics, the difference between the mean and three times the standard deviation is taken as the threshold. (where X is the final threshold, Let S be the mean and S be the standard deviation. The mean is... Among them, X i Let n be the number of relevant peaks. The standard deviation is... Among them, X i Here, n represents the number of relevant peak values. (This is the mean.)

[0090] 9. Methods for drawing composite images.

[0091] The specific method is as follows: First, align the reference time start and end points of the two images, and then use Python's matplotlib library to plot the two curves in the same image using different colors.

[0092] Although the present invention has been described in detail through the preferred embodiments above, it should be understood that the above description should not be considered as a limitation of the present invention. Various modifications and substitutions to the present invention will be apparent to those skilled in the art after reading the above description. Therefore, the scope of protection of the present invention should be defined by the appended claims.

[0093] The contents not described in detail in this specification are common knowledge to those skilled in the art.

Claims

1. A method for interpreting data processed by image matching, characterized in that, include: The actuator data is automatically segmented to obtain actuator displacement angle data and actuator motion angle data; Based on the actuator displacement angle data and actuator motion angle data, draw scatter plots of displacement angle and motion angle respectively; The displacement angle scatter plot and motion angle scatter plot are transformed into displacement servo curves and motion servo curves, respectively. The displacement servo curve and motion servo curve are converted to grayscale to obtain a grayscale image of the servo actuator. The grayscale image of the servo actuator is binarized to obtain a binarized image of the servo actuator. The binarized image of the servo actuator is subjected to normalized correlation matching to obtain the normalized matching result. Based on the normalized matching results, the relevant peak thresholds are determined; A threshold judgment is performed on the relevant peaks. If the normalized matching result is greater than the relevant peak threshold, the actuator displacement angle data and actuator motion angle data are successfully registered, and a composite image is drawn. If it is less than the relevant peak threshold, the registration is unsuccessful, a composite image is drawn, and troubleshooting is performed.

2. The method for interpreting data processed by image matching according to claim 1, characterized in that: The method for automatically segmenting actuator data to obtain actuator displacement angle data and actuator motion angle data is as follows: extracting valid data from the ignition moment to the end of flight and the power-off moment.

3. The method for interpreting data processed by image matching according to claim 1, characterized in that: The angular scatter plot is an image with the coordinate axes removed.

4. The method for interpreting data processed by image matching according to claim 1, characterized in that: The displacement angle scatter plot and motion angle scatter plot are enlarged proportionally, and the BGR mode of the displacement angle scatter plot and motion angle scatter plot are converted to RGB mode.

5. The method for interpreting data processed by image matching according to claim 1, characterized in that: The binarized image of the servo actuator is normalized, and the correlation coefficient is matched to obtain the normalized correlation coefficient matching result. The specific method is as follows: Where T is the template image, I is the image to be matched, (x,y) is the x-direction distance and y-direction distance of the top left corner of the template from the top left corner of the image, and x' and y' are the coordinates of the template moving with (x,y) as the origin.

6. The method for interpreting data processed by image matching according to claim 1, characterized in that: Based on several normalized matching results, the relevant peak thresholds are determined using the following method: The mean and standard deviation of the matching were obtained from the normalized matching results of several experimental images. The threshold is the difference between the mean and n times the standard deviation, where n is an integer greater than 1.

7. The method for interpreting data processed by image matching according to claim 6, characterized in that: The method for obtaining the matching mean and standard deviation from the normalized matching results of several test images is as follows: Among them, X i Here, n represents the number of relevant peak values. Where S is the standard deviation, X i Here, n represents the number of relevant peak values. This is the mean.

8. The method for interpreting data processed by image matching according to claim 6, characterized in that: The threshold is determined by taking the difference between the mean and n times the standard deviation. The specific method is as follows: Where X is the final threshold obtained. S is the mean, and S is the standard deviation.

9. The method for interpreting data processed by image matching according to claim 1, characterized in that: The specific method for drawing a composite image is as follows: The initial and final positions of the image plotted from the actuator displacement angle data and the image plotted from the actuator motion angle data are aligned according to time points to generate a composite image.

10. The method for interpreting data processed by image matching according to claim 6, characterized in that: n = 2 or 3.

11. A data interpretation system for image matching, characterized in that, include: The module consists of a data acquisition module, an image preprocessing module, and an image matching module, among which: Data acquisition module: automatically segments actuator data to obtain actuator displacement angle data and actuator motion angle data; sends the actuator displacement angle data and actuator motion angle data to the image preprocessing module; Image preprocessing module: Based on the actuator displacement angle data and actuator motion angle data sent by the data acquisition module, it plots displacement angle scatter plots and motion angle scatter plots respectively; it transforms the displacement angle scatter plots and motion angle scatter plots into displacement servo curves and motion servo curves respectively; it converts the displacement servo curves and motion servo curves into grayscale to obtain a grayscale image of the servo actuator; it binarizes the grayscale image of the servo actuator to obtain a binarized image of the servo actuator; and it sends the binarized image of the servo actuator to the image matching module. Image matching module: Performs normalized correlation matching on the binarized image of the servo actuator sent by the image preprocessing module to obtain a normalized matching result; determines a correlation peak threshold based on several of the normalized matching results; performs threshold judgment on the correlation peak, if the normalized matching result is greater than the correlation peak threshold, the actuator displacement angle data and actuator motion angle data are successfully registered, and a composite image is drawn; if it is less than the correlation peak threshold, the registration is unsuccessful, a composite image is drawn, and troubleshooting is performed.

Images