An efficient acquisition and analysis method for medical film thickness measurement data

Abstract

The present invention belongs to the technical field of methods for collecting and analyzing medical membrane thickness measurement data, and specifically relates to an efficient method for collecting and analyzing medical membrane thickness measurement data. In the method, during the production process of the medical membrane, a reflected laser is measured on an intercepted medical membrane, a two-dimensional matrix of all measurement points is constructed, each element in the two-dimensional matrix is ​​the thickness data of the corresponding measurement point, at least one suspected abnormal thickness data is screened, and all the suspected abnormal thickness data are divided into scattered noise data and abnormal thickness data; the filter window size of the scattered noise data, the abnormal thickness data and other thickness data in the two-dimensional matrix is ​​adaptively adjusted, so that all the thickness data in the two-dimensional matrix are mean filtered according to the adaptive filtering window size of each thickness data in the two-dimensional matrix to obtain filtered thickness data, thereby improving the filtering effect of the mean filter and improving the collection and detection efficiency of the laser measurement data of the medical membrane thickness.

Description

Technical Field

[0001] The present invention relates to the technical field of methods for collecting and analyzing medical membrane thickness measurement data, and specifically to a method for efficiently collecting and analyzing medical membrane thickness measurement data. Background Art

[0002] Medical membrane is a polymer film material used in the medical field. It has specific physical and chemical properties and is widely used in various medical devices and medical treatment processes. Common medical membranes include breathable membranes, moisture-permeable membranes, dressing membranes, catheter membranes, surgical coverings, and drug-controlled release membranes. In the production process of medical membranes, by measuring their thickness, deviations in production can be discovered and corrected in a timely manner to ensure that product quality meets standards. Laser thickness gauges are usually used to measure the thickness of medical membranes. Since the interference of environmental factors (temperature changes, vibrations, or airflow) on the measurement cannot be strictly controlled during the thickness measurement process, it is easy for the collected thickness data to contain noise data. Therefore, it is necessary to clean the collected thickness data of medical membranes to improve the accuracy of the collected thickness data of medical membranes.

[0003] The prior art usually uses a mean filter algorithm to filter and clean the thickness data of medical films collected by a laser thickness gauge to obtain a data set for detection that is not interfered by noise. However, in the process of thickness detection of multi-layer medical films, the multi-layer interlayer structure of the medical film will cause uneven laser refractive index, resulting in strong scattering noise, which is easily identified as a defect of uneven thickness. The filter cannot filter it out, and the scattering noise will also cause distortion of adjacent data, resulting in unsatisfactory mean filtering effect. The thickness data after data cleaning is still inaccurate.

[0004] Therefore, how to improve the filtering and cleaning effect of the mean filtering algorithm on the thickness data of the medical film to ensure the accuracy of quality inspection based on the thickness data of the medical film has become an urgent problem to be solved. Summary of the invention

[0005] In view of this, the present invention proposes a method for efficiently collecting and analyzing medical film thickness measurement data to solve the problems raised in the above background technology.

[0006] An efficient collection and analysis method for medical film thickness measurement data comprises the following steps:

[0007] In the production process of the medical film, the intercepted medical film is measured by reflection laser, and the initial reflection value and the final reflection value of each measuring point are obtained respectively, and a two-dimensional matrix of all measuring points is constructed according to the spatial position of each measuring point, and each element in the two-dimensional matrix is ​​the thickness data of the corresponding measuring point, and the thickness data is the difference between the initial reflection value and the final reflection value;

[0008] All thickness data in the two-dimensional matrix are sorted from small to large to obtain a thickness data sequence, at least one suspected abnormal thickness data is screened according to data fluctuation changes in the thickness data sequence, each suspected abnormal thickness data is subjected to regional growth in the two-dimensional matrix to obtain a growth region of each suspected abnormal thickness data, and all suspected abnormal thickness data are divided into scattered noise data and abnormal thickness data according to the thickness data distribution in the growth region of each suspected abnormal thickness data;

[0009] The filtering window sizes of the scattered noise data, abnormal thickness data and other thickness data in the two-dimensional matrix are adaptively adjusted to obtain the adaptive filtering window size of each thickness data in the two-dimensional matrix, and all the thickness data in the two-dimensional matrix are mean filtered according to the adaptive filtering window size of each thickness data in the two-dimensional matrix to obtain the filtered thickness data.

[0010] Preferably, the step of obtaining at least one suspected abnormal thickness data by screening according to the data fluctuation change in the thickness data sequence comprises:

[0011] For any thickness data in the thickness data sequence, according to the difference between the thickness data and its local thickness data and the discrete degree of the thickness data distribution in the thickness data sequence, obtaining the suspected abnormality degree of the thickness data;

[0012] A preset suspected abnormality degree threshold is obtained, and if the suspected abnormality degree of the thickness data is greater than or equal to the suspected abnormality degree threshold, the thickness data is confirmed to be suspected abnormal thickness data.

[0013] Preferably, obtaining the suspected abnormality degree of the thickness data according to the difference between the thickness data and its local thickness data and the discrete degree of thickness data distribution in the thickness data sequence includes:

[0014] Obtaining the thickness data variance, maximum thickness data and minimum thickness data of the thickness data sequence, obtaining a first difference between the thickness data and the minimum thickness data, and a second difference between the maximum thickness data and the minimum thickness data, calculating a ratio between the first difference and the second difference, and obtaining a first difference absolute value between the ratio of a preset multiple and a constant 1;

[0015] Obtaining left adjacent thickness data and right adjacent thickness data of the thickness data in the thickness data sequence, obtaining absolute values ​​of differences between the thickness data and the left adjacent thickness data and the right adjacent thickness data, respectively, and obtaining corresponding averages of the absolute values ​​of the differences;

[0016] The product of the thickness data variance, the first difference absolute value and the difference absolute value mean is used as the suspected abnormality degree of the thickness data.

[0017] Preferably, the method of dividing all suspected abnormal thickness data into scattered noise data and abnormal thickness data according to the thickness data distribution in the growth region of each suspected abnormal thickness data comprises:

[0018] According to the thickness data distribution in the growth area of ​​each suspected abnormal thickness data, obtaining the abnormal degree of scattering noise corresponding to the suspected abnormal thickness data;

[0019] According to the abnormal degree of the scattering noise of each of the suspected abnormal thickness data, all the suspected abnormal thickness data are divided into scattered noise data and abnormal thickness data.

[0020] Preferably, obtaining the abnormal degree of scattering noise corresponding to each suspected abnormal thickness data according to the thickness data distribution in the growth region of each suspected abnormal thickness data comprises:

[0021] For any suspected abnormal thickness data, obtain the minimum circumscribed rectangle of the growth area of ​​the suspected abnormal thickness data to obtain the minimum circumscribed rectangle area, calculate the ratio between the area of ​​the growth area of ​​the suspected abnormal thickness data and the minimum circumscribed rectangle area, and normalize the ratio to obtain a corresponding first normalized value;

[0022] Take any thickness data in the growth area of ​​the suspected abnormal thickness data as the target thickness data, obtain the absolute value of the thickness difference and the absolute value of the suspected abnormal degree difference between the target thickness data and each thickness data in its growth area, and obtain the addition result between the absolute value of the thickness difference and the absolute value of the suspected abnormal degree difference corresponding to each thickness data in the growth area of ​​the target thickness data, and calculate the mean value of the addition result according to the addition result corresponding to each thickness data in the growth area of ​​the target thickness data;

[0023] Obtaining the mean of the addition results corresponding to each thickness data in the growth area of ​​the suspected abnormal thickness data, obtaining the average of the mean of the addition results, and normalizing the average of the mean of the addition results to obtain a corresponding second normalized value;

[0024] The addition result between the first normalized value and the second normalized value is used as the abnormal degree of scattering noise of the suspected abnormal thickness data.

[0025] Preferably, the method of dividing all suspected abnormal thickness data into scattered noise data and abnormal thickness data according to the abnormal degree of the scattered noise of each suspected abnormal thickness data comprises:

[0026] A preset scattering noise abnormality threshold is obtained. If the scattering noise abnormality of any suspected abnormal thickness data is greater than the scattering noise abnormality threshold, the suspected abnormal thickness data is determined to be scattering noise data. If the scattering noise abnormality of any suspected abnormal thickness data is less than or equal to the scattering noise abnormality threshold, the suspected abnormal thickness data is determined to be abnormal thickness data.

[0027] Preferably, the adaptively adjusting the filter window size of the scattered noise data, abnormal thickness data and other thickness data in the two-dimensional matrix to obtain the adaptive filter window size of each thickness data in the two-dimensional matrix includes:

[0028] A preset initial filter window size is obtained, and for any scattered noise data in the two-dimensional matrix, the initial filter window size is increased by a preset proportion to obtain an adaptive filter window size for the scattered noise data; for any abnormal thickness data in the two-dimensional matrix, the initial filter window size is reduced by a preset proportion to obtain an adaptive filter window size for the abnormal thickness data; for any other thickness data in the two-dimensional matrix except the scattered noise data and the abnormal thickness data, the initial filter window size is used as the adaptive filter window size for the other thickness data.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] In the production process of the medical film of the present invention, the intercepted medical film is measured by reflected laser, and the initial reflection value and the final reflection value of each measuring point are obtained respectively; a two-dimensional matrix of all measuring points is constructed according to the spatial position of each measuring point, and each element in the two-dimensional matrix is ​​the thickness data of the corresponding measuring point, and the thickness data is the difference between the initial reflection value and the final reflection value; all the thickness data in the two-dimensional matrix are sorted from small to large to obtain a thickness data sequence, and at least one suspected abnormal thickness data is screened according to the data fluctuation change in the thickness data sequence, and each of the suspected abnormal thickness data is regionally grown in the two-dimensional matrix to obtain the growth area of ​​each suspected abnormal thickness data; according to the thickness data distribution in the growth area of ​​each suspected abnormal thickness data, all the suspected abnormal thickness data are divided into scattered noise data and abnormal thickness data; the filtering window size of the scattered noise data, the abnormal thickness data and other thickness data in the two-dimensional matrix is ​​adaptively adjusted to obtain the adaptive filtering window size of each thickness data in the two-dimensional matrix, and all the thickness data in the two-dimensional matrix are mean filtered according to the adaptive filtering window size of each thickness data in the two-dimensional matrix to obtain the filtered thickness data. Among them, suspected abnormal thickness data are screened according to the thickness data of each measuring point, and then regional growth is performed on these suspected abnormal thickness data to obtain the abnormal degree of scattering noise of each suspected abnormal thickness data according to the thickness data distribution in the production area, which is used to classify the thickness data of all measuring points, and then the adaptive filtering window of each thickness data is adjusted according to the classification result, so that it is easier to filter out the scattering noise during mean filtering, and retain the abnormal thickness data of the actual uneven thickness of the medical membrane, thereby improving the filtering effect of the mean filtering and improving the collection and detection efficiency of the laser measurement data of the medical membrane thickness. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0032] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0033] Embodiments of the present disclosure are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limiting the present disclosure.

[0034] It should be noted that the terms "first", "second", etc. in the specification of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure.

[0035] In order to illustrate the technical solution of the present invention, specific embodiments are provided below for illustration.

[0036] like Figure 1 As shown, an efficient collection and analysis method for medical film thickness measurement data includes:

[0037] Step S101, during the production process of the medical film, the intercepted medical film is measured for reflected laser, and the initial reflection value and the final reflection value of each measuring point are obtained respectively, and a two-dimensional matrix of all measuring points is constructed according to the spatial position of each measuring point. Each element in the two-dimensional matrix is ​​the thickness data of the corresponding measuring point, and the thickness data is the difference between the initial reflection value and the final reflection value.

[0038] Use a laser thickness gauge to measure the thickness of the medical film in the production line. Each time, a fixed length of medical film is intercepted for measurement. There is no limit on the length of the intercepted medical film. When using a laser thickness gauge to measure the reflected laser of the intercepted medical film, lasers at multiple measuring points are used to evenly collect the thickness data corresponding to multiple measuring points on the medical film. It is worth noting that when each measuring point receives the reflected laser data, the initial reflection value and the final reflection value of the reflected laser can be obtained, which are used to reflect the return value of the laser on the surface and ground of the medical film. The absolute value of the difference between the initial reflection value and the final reflection value of each measuring point is used as the thickness data of the corresponding measuring point on the medical film, in μm. At this point, the thickness data of each measuring point on the medical film can be obtained.

[0039] According to the spatial position of each measurement point collected and the corresponding thickness data, a two-dimensional matrix of the thickness data of the medical film is constructed, wherein the rows of the two-dimensional matrix are the horizontal coordinates of the measurement points, the columns of the two-dimensional matrix are the vertical coordinates of the measurement points, and the elements in the two-dimensional matrix are the thickness data of the corresponding measurement points.

[0040] Step S102, sorting all thickness data in the two-dimensional matrix from small to large to obtain a thickness data sequence, screening to obtain at least one suspected abnormal thickness data according to the data fluctuation change in the thickness data sequence, performing regional growth on each suspected abnormal thickness data in the two-dimensional matrix to obtain the growth region of each suspected abnormal thickness data, and dividing all suspected abnormal thickness data into scattered noise data and abnormal thickness data according to the thickness data distribution in the growth region of each suspected abnormal thickness data.

[0041] The environmental noise in the process of measuring the thickness of medical membranes includes but is not limited to environmental factors such as temperature, vibration, and airflow. These environmental factors will cause slight changes in the reflection value when the medical membrane is reflected by laser measurement, that is, noise data appears in the collected thickness data. In this case, the mean filtering algorithm is usually used to filter out the noise data in the thickness data to achieve the purpose of data cleaning and ensure that the collected thickness data of the medical membrane is relatively accurate. However, considering that the multi-layer structure of the medical membrane causes a large degree of scattering noise, the filtering effect of the mean filter under the fixed window is poor; secondly, the uneven thickness defects of the medical membrane itself will also cause abnormal reflection values, and the mean filter will smooth out these abnormal values ​​that need to be further detected later. In view of the above two situations where the quality of the collected thickness data set cannot be guaranteed in the mean filtering process, it is necessary to first obtain suspected abnormal data, including thickness data caused by scattering noise and thickness data of uneven thickness defects, so as to further adjust the filter window of each thickness data in the future.

[0042] Scattering noise and thickness data of uniformity defects have a common feature, that is, the difference between the reflection value and the reflection value corresponding to the normal thickness of the medical film is large, which may be caused by the reflection value returned by the surface of the medical film being too small or too large, or the reflection value returned by the bottom of the film being too small or too large. Therefore, it is necessary to first screen the suspected abnormal thickness data in the two-dimensional matrix.

[0043] Since each element (thickness data) in the two-dimensional matrix is ​​the absolute value of the difference between the initial reflection value and the final reflection value of the corresponding measurement point, all thickness data in the two-dimensional matrix are sorted from small to large according to the absolute value of the difference to obtain a one-dimensional thickness data sequence L. The closer the thickness data is to the boundaries on both sides of the sequence, the greater the possibility of abnormality. However, when there is little or no scattering noise or uneven thickness defects on the surface of the medical film, the thickness data close to the boundary is also unlikely to be abnormal. Therefore, it is necessary to screen and obtain at least one suspected abnormal thickness data according to the data fluctuation changes in the thickness data sequence.

[0044] Preferably, at least one suspected abnormal thickness data is obtained by screening according to the data fluctuation change in the thickness data sequence, including:

[0045] (1) For any thickness data in the thickness data sequence, the suspected abnormality degree of the thickness data is obtained according to the difference between the thickness data and its local thickness data and the discrete degree of the thickness data distribution in the thickness data sequence.

[0046] Wherein, obtaining the suspected abnormality degree of the thickness data according to the difference between the thickness data and its local thickness data and the discreteness degree of the thickness data distribution in the thickness data sequence includes:

[0047] Obtaining the thickness data variance, maximum thickness data and minimum thickness data of the thickness data sequence, obtaining a first difference between the thickness data and the minimum thickness data, and a second difference between the maximum thickness data and the minimum thickness data, calculating a ratio between the first difference and the second difference, and obtaining a first difference absolute value between the ratio of a preset multiple and a constant 1;

[0048] Obtaining left adjacent thickness data and right adjacent thickness data of the thickness data in the thickness data sequence, obtaining absolute values ​​of differences between the thickness data and the left adjacent thickness data and the right adjacent thickness data, respectively, and obtaining corresponding averages of the absolute values ​​of the differences;

[0049] The product of the thickness data variance, the first difference absolute value and the difference absolute value mean is used as the suspected abnormality degree of the thickness data.

[0050] In one embodiment, taking the qth thickness data in the thickness data sequence as an example, the calculation expression of the suspected abnormality degree of the qth thickness data is:

[0051]

[0052] Among them, R q represents the suspected abnormality of the qth thickness data, σ(L) represents the thickness data variance of the thickness data sequence, L q-1 represents the q-1th thickness data in the thickness data sequence, that is, the left adjacent thickness data of the qth thickness data, L q+1 represents the q+1th thickness data in the thickness data sequence, that is, the thickness data right adjacent to the qth thickness data, L q represents the qth thickness data in the thickness data sequence, || represents the absolute value symbol, min(L) represents the minimum value in the thickness data sequence, max(L) represents the maximum value in the thickness data sequence, and 1 represents a constant.

[0053] It should be noted that the larger the thickness data variance σ(L) of the thickness data sequence is, the less concentrated the data distribution in the thickness data sequence is, the more likely there is discrete thickness data, the greater the possibility that the corresponding qth thickness data is abnormal thickness data, and the greater the suspected abnormality of the corresponding qth thickness data; The value of is used to characterize the difference between the qth thickness data and its adjacent thickness data. The larger the difference, the greater the possibility that the qth thickness is abnormally discrete. The purpose is to normalize the value of the qth thickness data to the interval [-1, 1], and then perform an absolute value so that, under the premise of the possibility of abnormality, the closer the value of the qth thickness data is to the boundary value, the greater the degree of suspected abnormality.

[0054] Similarly, obtaining the suspected abnormality degree of each thickness data in the thickness data sequence is to determine the suspected abnormality degree of each thickness data in the two-dimensional matrix.

[0055] (2) A preset suspected abnormality degree threshold is obtained. If the suspected abnormality degree of the thickness data is greater than or equal to the suspected abnormality degree threshold, the thickness data is confirmed to be suspected abnormal thickness data.

[0056] In one embodiment, the suspected abnormality degree threshold is set to 0.8. If the suspected abnormality degree of any thickness data is greater than or equal to 0.8, the thickness data is determined to be suspected abnormal thickness data. Each thickness data is traversed to obtain all suspected abnormal thickness data in the two-dimensional matrix.

[0057] Furthermore, due to the complex manufacturing process of multilayer films, the interlayer bonding is prone to unevenness. Although it does not affect the overall thickness of the film, the uneven interface or uneven distribution between layers in the multilayer film will cause the laser to scatter when reflected at the interlayer interface when the overall thickness remains unchanged, forming scattering noise, which interferes with the thickness measurement of the medical film. When the overall thickness unevenness defect of the film occurs, the defective part is usually distributed in blocks, that is, the distribution of abnormal thickness data is more concentrated, and due to the adhesion between the multilayer films, the uneven distribution of the thickness of the film changes slowly in space, that is, the thickness difference between the thickness data and the possibility of abnormality are low. In contrast, the laser scattering caused by the uneven distribution between the layers of the multilayer film is more dispersed and random in distribution. Therefore, the suspected abnormal thickness data in the two-dimensional matrix is ​​classified by the distribution of the neighborhood data of the suspected abnormal thickness data, so as to divide the suspected abnormal thickness data into scattered noise data and abnormal thickness data.

[0058] First, determine the neighborhood data of each suspected abnormal thickness data in the two-dimensional matrix, then perform regional growth on each suspected abnormal thickness data in the two-dimensional matrix to obtain the growth area of ​​each suspected abnormal thickness data. Specifically, for any suspected abnormal thickness data, the suspected abnormal thickness data is used as a seed point, and the growth range each time is the eight neighborhoods of the seed point. The growth condition is: when the absolute value of the difference between the thickness data of the seed point and the thickness data of any data point in its neighborhood is less than 0.2μm, growth is performed. When there is no point that can grow in the neighborhood of the seed point, the growth is terminated, and the growth area of ​​the suspected abnormal thickness data corresponding to the seed point is obtained. It should be noted that regional growth belongs to the prior art and will not be described in detail here. At this point, the growth area of ​​each suspected abnormal thickness data in the two-dimensional matrix is ​​obtained, which is used to characterize the distribution of its neighborhood data.

[0059] After obtaining the growth region of each suspected abnormal thickness data, all suspected abnormal thickness data can be divided into scattered noise data and abnormal thickness data according to the thickness data distribution in the growth region of each suspected abnormal thickness data. The specific process is as follows:

[0060] (1) According to the thickness data distribution in the growth area of ​​each of the suspected abnormal thickness data, the abnormal degree of scattering noise corresponding to the suspected abnormal thickness data is obtained.

[0061] Specifically, for any suspected abnormal thickness data, the minimum circumscribed rectangle of the growth region of the suspected abnormal thickness data is obtained to obtain the minimum circumscribed rectangle area, the ratio between the area of ​​the growth region of the suspected abnormal thickness data and the minimum circumscribed rectangle area is calculated, and the ratio is normalized to obtain a corresponding first normalized value;

[0062] Take any thickness data in the growth area of ​​the suspected abnormal thickness data as the target thickness data, obtain the absolute value of the thickness difference and the absolute value of the suspected abnormal degree difference between the target thickness data and each thickness data in its growth area, and obtain the addition result between the absolute value of the thickness difference and the absolute value of the suspected abnormal degree difference corresponding to each thickness data in the growth area of ​​the target thickness data, and calculate the mean value of the addition result according to the addition result corresponding to each thickness data in the growth area of ​​the target thickness data;

[0063] Obtaining the mean of the addition results corresponding to each thickness data in the growth area of ​​the suspected abnormal thickness data, obtaining the average of the mean of the addition results, and normalizing the average of the mean of the addition results to obtain a corresponding second normalized value;

[0064] The addition result between the first normalized value and the second normalized value is used as the abnormal degree of scattering noise of the suspected abnormal thickness data.

[0065] In one embodiment, taking the pth suspected abnormal thickness data in the two-dimensional matrix as an example, for the suspected abnormal thickness data corresponding to the thickness unevenness defect, the growth area should be more complete due to more similar thickness data around it, and the thickness data distribution in the growth area is more concentrated, and the difference in the suspected abnormal degree of each suspected abnormal thickness data in the growth area is smaller; for the suspected abnormal thickness data corresponding to the laser scattering noise, the distribution of the thickness data with the same characteristics in its growth area is relatively more discrete and random, and the obtained growth area is uneven in shape, which is prone to defects, and the difference between each thickness data in the growth area is relatively larger. Therefore, the minimum circumscribed rectangle of the growth area of ​​the pth suspected abnormal thickness data is obtained to obtain the minimum circumscribed rectangle area, and the scattering noise abnormality degree of the pth suspected abnormal thickness data is obtained according to the minimum circumscribed rectangle area and the difference between each thickness data in the growth area of ​​the pth suspected abnormal thickness data. Then, the calculation expression of the scattering noise abnormality degree of the pth suspected abnormal thickness data is:

[0066]

[0067] Among them, W p represents the abnormal degree of scattering noise of the pth suspected abnormal thickness data, norm() represents the normalization function, S p The minimum circumscribed rectangular area of ​​the growth area of ​​the pth suspected abnormal thickness data, n p represents the area of ​​the growth region of the pth suspected abnormal thickness data, n p represents the number of thickness data contained in the growth region of the pth suspected abnormal thickness data, n i represents the number of thickness data in the growth region of the i-th thickness data in the growth region of the p-th suspected abnormal thickness data, L i represents the i-th thickness data in the growth area of ​​the p-th suspected abnormal thickness data, L j represents the jth thickness data in the growth area of ​​the i-th thickness data in the growth area of ​​the p-th suspected abnormal thickness data, R i represents the suspected abnormality degree of the i-th thickness data in the growth area of ​​the p-th suspected abnormal thickness data, R j represents the suspected abnormality degree of the jth thickness data in the growth region of the ith thickness data in the growth region of the pth suspected abnormal thickness data, and || represents the absolute value sign.

[0068] It should be noted that It is used to characterize the proportion of the actual growth area of ​​the pth suspected abnormal thickness data to the minimum circumscribed rectangle. The larger the proportion, the more uniform and complete the regional growth shape of the pth suspected abnormal thickness data is. Otherwise, the smaller the proportion, the more consistent with the growth point distribution of the scattering noise anomaly, and the greater the degree of the corresponding scattering noise anomaly. Therefore, the inverse form is used. The value of is used to characterize the degree of neighborhood difference of the i-th thickness data in the growth area of ​​the p-th suspected abnormal thickness data. The larger the value, the greater the thickness data difference between the neighbors of the i-th thickness data. The value of is used to characterize the cumulative value of the neighborhood difference degree of all thickness data in the growth area of ​​the p-th suspected abnormal thickness data. The larger the cumulative value of the neighborhood difference degree is, the greater the difference between the overall thickness data in the growth area of ​​the p-th suspected abnormal thickness data is, the more it belongs to scattered noise data, and the greater the abnormality degree of the scattering noise corresponding to the p-th suspected abnormal thickness data is.

[0069] Similarly, the abnormal degree of scattering noise of each suspected abnormal thickness data in the two-dimensional matrix is ​​obtained.

[0070] (2) According to the abnormal degree of the scattering noise of each of the suspected abnormal thickness data, all the suspected abnormal thickness data are divided into scattered noise data and abnormal thickness data.

[0071] Specifically, a preset scattering noise abnormality level threshold is obtained. If the scattering noise abnormality level of any suspected abnormal thickness data is greater than the scattering noise abnormality level threshold, the suspected abnormal thickness data is determined to be scattering noise data. If the scattering noise abnormality level of any suspected abnormal thickness data is less than or equal to the scattering noise abnormality level threshold, the suspected abnormal thickness data is determined to be abnormal thickness data.

[0072] In one embodiment, the scattering noise abnormality threshold is set to 0.7, which can be adjusted floatingly according to the real-time scenario. When the scattering noise abnormality of any suspected abnormal thickness data is greater than 0.7, it means that the suspected abnormal thickness data is abnormal thickness data formed by the influence of scattering noise. Therefore, the suspected abnormal thickness data is determined to be scattering noise data. Conversely, when the scattering noise abnormality of any suspected abnormal thickness data is less than or equal to 0.7, the suspected abnormal thickness data is determined to be abnormal thickness data, which is thickness data corresponding to the overall thickness unevenness defect of the film.

[0073] Step S103, adaptively adjust the filter window size of the scattered noise data, abnormal thickness data and other thickness data in the two-dimensional matrix to obtain the adaptive filter window size of each thickness data in the two-dimensional matrix, and perform mean filtering on all thickness data in the two-dimensional matrix according to the adaptive filter window size of each thickness data in the two-dimensional matrix to obtain filtered thickness data.

[0074] According to step S102, the thickness data in the two-dimensional matrix is ​​divided into scattered noise data, abnormal thickness data and other thickness data (normal thickness data), and then the filter window in the mean filtering process is adaptively adjusted according to the division result to avoid smoothing out the abnormal thickness data caused by non-scattered noise, which affects the subsequent further quality detection of medical membranes. Therefore, the filter window sizes of the scattered noise data, abnormal thickness data and other thickness data in the two-dimensional matrix are adaptively adjusted to obtain the adaptive filter window size of each thickness data in the two-dimensional matrix. The specific method is:

[0075] A preset initial filter window size is obtained, and for any scattered noise data in the two-dimensional matrix, the initial filter window size is increased by a preset proportion to obtain an adaptive filter window size for the scattered noise data; for any abnormal thickness data in the two-dimensional matrix, the initial filter window size is reduced by a preset proportion to obtain an adaptive filter window size for the abnormal thickness data; for any other thickness data in the two-dimensional matrix except the scattered noise data and the abnormal thickness data, the initial filter window size is used as the adaptive filter window size for the other thickness data.

[0076] In one embodiment, the initial filter window size of the filter is set to 8 neighborhoods for each data point. For any scattered noise data in the two-dimensional matrix, the filter window size when the scattered noise data is mean filtered is set to 24 neighborhoods to increase the filtering and smoothing effect of the scattered noise data; for any abnormal thickness data in the two-dimensional matrix, the filter window size when the abnormal thickness data is mean filtered is set to 4 neighborhoods to prevent the abnormal thickness data from being over-smoothed, resulting in distortion of important detection data; for any other thickness data in the two-dimensional matrix except the scattered noise data and the abnormal thickness data, the filter window size when the other thickness data is mean filtered is kept at 8 neighborhoods. So far, the adaptive filter window size of each thickness data in the two-dimensional matrix is ​​obtained.

[0077] Furthermore, after obtaining the adaptive filtering window size of each thickness data in the two-dimensional matrix, a mean filtering algorithm with an adaptively adjustable filter window area is used to perform mean filtering on the thickness data in the two-dimensional matrix, traversing from the first thickness data in the upper left corner of the two-dimensional matrix to the last thickness data in the lower right corner, and performing mean filtering on all thickness data according to the adaptive filtering window size of each thickness data to obtain the filtered thickness data, thereby completing the data cleaning of the thickness data of the medical film. Among them, mean filtering belongs to the prior art and will not be described in detail here.

[0078] It should be noted that the thickness data after filtering and smoothing eliminates the interference of environmental noise. Combined with the thickness data distribution characteristics of different measurement points, it further eliminates the laser scattering interference noise existing in the laser thickness measurement of medical membranes. At the same time, it reduces the risk of points with uneven thickness characteristics being filtered and smoothed, so as to obtain more accurate filtering results, thereby collecting more accurate laser thickness measurement data sets of medical membranes, and improving the accuracy and efficiency of medical membrane thickness detection.

[0079] In summary, in the production process of the medical film, the embodiment of the present invention performs a reflection laser measurement on the intercepted medical film to obtain the initial reflection value and the final reflection value of each measuring point respectively, and constructs a two-dimensional matrix of all measuring points according to the spatial position of each measuring point, wherein each element in the two-dimensional matrix is ​​the thickness data of the corresponding measuring point, and the thickness data is the difference between the initial reflection value and the final reflection value; sorts all the thickness data in the two-dimensional matrix from small to large to obtain a thickness data sequence, and obtains at least one suspected abnormal thickness data according to the data fluctuation change in the thickness data sequence, performs regional growth on each of the suspected abnormal thickness data in the two-dimensional matrix to obtain the growth region of each of the suspected abnormal thickness data, and divides all the suspected abnormal thickness data into scattered noise data and abnormal thickness data according to the thickness data distribution in the growth region of each of the suspected abnormal thickness data; adaptively adjusts the filtering window size of the scattered noise data, abnormal thickness data and other thickness data in the two-dimensional matrix to obtain the adaptive filtering window size of each thickness data in the two-dimensional matrix, and performs mean filtering on all the thickness data in the two-dimensional matrix according to the adaptive filtering window size of each thickness data in the two-dimensional matrix to obtain the filtered thickness data. Among them, suspected abnormal thickness data are screened according to the thickness data of each measuring point, and then regional growth is performed on these suspected abnormal thickness data to obtain the abnormal degree of scattering noise of each suspected abnormal thickness data according to the thickness data distribution in the production area, which is used to classify the thickness data of all measuring points, and then the adaptive filtering window of each thickness data is adjusted according to the classification result, so that it is easier to filter out the scattering noise during mean filtering, and retain the abnormal thickness data of the actual uneven thickness of the medical membrane, thereby improving the filtering effect of the mean filtering and improving the collection and detection efficiency of the laser measurement data of the medical membrane thickness.

[0080] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. An efficient collection and analysis method for medical film thickness measurement data, characterized in that: include: In the production process of the medical film, the intercepted medical film is measured by reflection laser, and the initial reflection value and the final reflection value of each measuring point are obtained respectively, and a two-dimensional matrix of all measuring points is constructed according to the spatial position of each measuring point, and each element in the two-dimensional matrix is ​​the thickness data of the corresponding measuring point, and the thickness data is the difference between the initial reflection value and the final reflection value; All thickness data in the two-dimensional matrix are sorted from small to large to obtain a thickness data sequence, at least one suspected abnormal thickness data is screened according to data fluctuation changes in the thickness data sequence, each suspected abnormal thickness data is subjected to regional growth in the two-dimensional matrix to obtain a growth region of each suspected abnormal thickness data, and all suspected abnormal thickness data are divided into scattered noise data and abnormal thickness data according to the thickness data distribution in the growth region of each suspected abnormal thickness data; Adaptively adjusting the filter window sizes of the scattered noise data, abnormal thickness data and other thickness data in the two-dimensional matrix to obtain the adaptive filter window size of each thickness data in the two-dimensional matrix, and performing mean filtering on all thickness data in the two-dimensional matrix according to the adaptive filter window size of each thickness data in the two-dimensional matrix to obtain filtered thickness data; The step of obtaining at least one suspected abnormal thickness data according to the data fluctuation change in the thickness data sequence includes: For any thickness data in the thickness data sequence, according to the difference between the thickness data and its local thickness data and the discrete degree of the thickness data distribution in the thickness data sequence, obtaining the suspected abnormality degree of the thickness data; Obtaining a preset suspected abnormality degree threshold, and if the suspected abnormality degree of the thickness data is greater than or equal to the suspected abnormality degree threshold, confirming that the thickness data is suspected abnormal thickness data; The obtaining the suspected abnormality degree of the thickness data according to the difference between the thickness data and its local thickness data and the discreteness degree of the thickness data distribution in the thickness data sequence includes: Obtaining the thickness data variance, maximum thickness data and minimum thickness data of the thickness data sequence, obtaining a first difference between the thickness data and the minimum thickness data, and a second difference between the maximum thickness data and the minimum thickness data, calculating a ratio between the first difference and the second difference, and obtaining a first difference absolute value between the ratio of a preset multiple and a constant 1; Obtaining left adjacent thickness data and right adjacent thickness data of the thickness data in the thickness data sequence, obtaining absolute values ​​of differences between the thickness data and the left adjacent thickness data and the right adjacent thickness data, respectively, and obtaining corresponding averages of the absolute values ​​of the differences; taking the product of the thickness data variance, the first difference absolute value and the difference absolute value mean as the suspected abnormality degree of the thickness data; The method of dividing all suspected abnormal thickness data into scattered noise data and abnormal thickness data according to the thickness data distribution in the growth region of each suspected abnormal thickness data comprises: According to the thickness data distribution in the growth area of ​​each suspected abnormal thickness data, obtaining the abnormal degree of scattering noise corresponding to the suspected abnormal thickness data; According to the abnormal degree of scattering noise of each of the suspected abnormal thickness data, all the suspected abnormal thickness data are divided into scattering noise data and abnormal thickness data; The step of obtaining the abnormal degree of scattering noise corresponding to each suspected abnormal thickness data according to the thickness data distribution in the growth region of each suspected abnormal thickness data comprises: For any suspected abnormal thickness data, obtain the minimum circumscribed rectangle of the growth area of ​​the suspected abnormal thickness data to obtain the minimum circumscribed rectangle area, calculate the ratio between the area of ​​the growth area of ​​the suspected abnormal thickness data and the minimum circumscribed rectangle area, and normalize the ratio to obtain a corresponding first normalized value; Take any thickness data in the growth area of ​​the suspected abnormal thickness data as the target thickness data, obtain the absolute value of the thickness difference and the absolute value of the suspected abnormal degree difference between the target thickness data and each thickness data in its growth area, and obtain the addition result between the absolute value of the thickness difference and the absolute value of the suspected abnormal degree difference corresponding to each thickness data in the growth area of ​​the target thickness data, and calculate the mean value of the addition result according to the addition result corresponding to each thickness data in the growth area of ​​the target thickness data; Obtaining the mean of the addition results corresponding to each thickness data in the growth area of ​​the suspected abnormal thickness data, obtaining the average of the mean of the addition results, and normalizing the average of the mean of the addition results to obtain a corresponding second normalized value; taking the sum of the first normalized value and the second normalized value as the abnormal degree of scattering noise of the suspected abnormal thickness data; The adaptively adjusting the filter window size of the scattered noise data, abnormal thickness data and other thickness data in the two-dimensional matrix to obtain the adaptive filter window size of each thickness data in the two-dimensional matrix includes: A preset initial filter window size is obtained, and for any scattered noise data in the two-dimensional matrix, the initial filter window size is increased by a preset proportion to obtain an adaptive filter window size for the scattered noise data; for any abnormal thickness data in the two-dimensional matrix, the initial filter window size is reduced by a preset proportion to obtain an adaptive filter window size for the abnormal thickness data; for any other thickness data in the two-dimensional matrix except the scattered noise data and the abnormal thickness data, the initial filter window size is used as the adaptive filter window size for the other thickness data.

2. The method for efficiently collecting and analyzing medical film thickness measurement data according to claim 1, characterized in that: The method of dividing all suspected abnormal thickness data into scattered noise data and abnormal thickness data according to the abnormal degree of the scattering noise of each suspected abnormal thickness data comprises: A preset scattering noise abnormality threshold is obtained. If the scattering noise abnormality of any suspected abnormal thickness data is greater than the scattering noise abnormality threshold, the suspected abnormal thickness data is determined to be scattering noise data. If the scattering noise abnormality of any suspected abnormal thickness data is less than or equal to the scattering noise abnormality threshold, the suspected abnormal thickness data is determined to be abnormal thickness data.

Images