Standardized image acquisition method, device and equipment

Abstract

The invention discloses a standardized image acquisition method, device and equipment. The method comprises the following steps: acquiring key element training data of a sample gear; inputting the key element training data into different types of classifiers for training, and screening out a target classifier from the trained classifiers; inputting the measurement parameters of the gear to be detected into the target classifier, and obtaining shooting light source parameters output by the target classifier; and controlling the optical acquisition assembly to perform standardized image acquisition on the to-be-detected gear based on the shooting light source parameters. Targeted training is carried out on different types of gears to obtain a target classifier, and unified shooting light source parameters are output for the same type of gears through the target classifier, so that unified standardization of an optical environment of the same type of gears during image acquisition is ensured; therefore, the uniformity and the accuracy of the detection effect of the AI algorithm are improved.

Description

Technical Field

[0001] The present invention relates to the field of detection technology, and in particular to a method, device and equipment for standardizing image acquisition. Background Art

[0002] Gears, as key transmission components, can suffer from inherent quality issues such as blackened skin, missing parts, scratches, cracks, and uneven surfaces. These issues can cause vibration, noise, or even breakage during operation, severely impacting mechanical performance, service life, and operational safety. Currently, gear appearance inspection typically relies on computer vision technology. This involves capturing an image of the gear surface using a high-resolution camera and transmitting it to a host computer. Traditional machine vision algorithms or AI algorithms are deployed on the host computer to infer the image and detect surface defects.

[0003] However, during image acquisition, the different gears in the delivery scene vary greatly in size and shape. Due to the differences in size and shape, the optical environment changes. Therefore, every time a new gear is encountered for inspection, the light source needs to be manually debugged, resulting in inconsistent shooting optical environments for each part, thus affecting the uniformity and accuracy of the AI ​​algorithm's detection effect. Summary of the Invention

[0004] The present invention provides a method for standardizing image acquisition to realize standardization acquisition of images.

[0005] According to a first aspect of the present invention, there is provided an image standardization acquisition method, comprising: acquiring key element training data of a sample gear, wherein the key element training data includes sample gear parameters, sample gear shooting light source parameters, and sample gear image parameters;

[0006] Inputting the key element training data into different types of classifiers for training, and screening out a target classifier from the trained classifiers;

[0007] Inputting the measurement parameters of the gear to be inspected into the target classifier, and obtaining the shooting light source parameters output by the target classifier;

[0008] The optical acquisition component is controlled to perform standardized image acquisition on the gear to be detected based on the shooting light source parameters.

[0009] Optionally, obtain training data on key elements of sample gears, including:

[0010] Obtain the accumulated data of the historical experimental process of the sample gear, and use the specified algorithm to perform data association analysis to extract key element data;

[0011] The extracted key element data are divided according to a ratio to obtain the key element training data and key element test data.

[0012] Optionally, the key element training data is input into different types of classifiers for training, and a target classifier is selected from the trained classifiers, including:

[0013] Inputting the key element training data into different types of classifiers so that each classifier is synchronously trained based on the input key element training data, wherein the classifiers include a logistic regression classifier, a random forest classifier, a naive Bayes classifier, and a K-nearest neighbor classifier;

[0014] Input the key factor test data into each trained classifier, and score the classifier according to the output results of each classifier according to the specified evaluation indicators;

[0015] The scoring results of each classifier are obtained, and the classifier with the highest scoring result is used as the target classifier.

[0016] Optionally, inputting the measurement parameters of the gear to be inspected into the target classifier and obtaining the shooting light source parameters output by the target classifier includes:

[0017] Inputting measurement parameters of the gear to be inspected into the target classifier, wherein the measurement parameters include tilt angle, tooth height, tooth width and diameter;

[0018] The target classifier determines a gear type based on the measurement parameters, and outputs the photographing light source parameters matching the gear type.

[0019] Optionally, controlling the optical acquisition component to perform standardized image acquisition on the gear to be inspected based on the shooting light source parameters includes:

[0020] Configuring the shooting light source parameters to the optical acquisition component, and moving the optical acquisition component to a target position by a robotic arm;

[0021] The optical acquisition component is controlled to acquire a standardized image of the gear to be detected at the target position.

[0022] Optionally, controlling the optical acquisition component to acquire a standardized image of the gear to be inspected at the target position includes:

[0023] Controlling the rotary clamping mechanism to rotate the gear to be detected to determine different placement angles of the gear to be detected;

[0024] The optical acquisition component is controlled to perform standardized image acquisition on the gears to be detected at different placement angles at the target position.

[0025] Optionally, after controlling the optical acquisition component to acquire a standardized image of the gear to be inspected based on the shooting light source parameters, the method further includes:

[0026] Perform defect detection on standardized images of gears to be inspected;

[0027] A defect detection report is generated based on the defect detection results, and the defect detection report is visually displayed.

[0028] According to another aspect of the present invention, there is provided an image standardization acquisition device, comprising: a key element training data acquisition module, configured to acquire key element training data of a sample gear, wherein the key element training data includes sample gear parameters, sample gear shooting light source parameters, and sample gear image parameters;

[0029] A target classifier screening module is used to input the key element training data into different types of classifiers for training, and to screen out target classifiers from the trained classifiers;

[0030] a shooting light source parameter acquisition module, configured to input the measurement parameters of the gear to be detected into the target classifier and acquire the shooting light source parameters output by the target classifier;

[0031] The standardized image acquisition module is used to control the optical acquisition component to perform standardized image acquisition on the gear to be detected based on the shooting light source parameters.

[0032] According to another aspect of the present invention, an electronic device is provided, comprising:

[0033] at least one processor; and

[0034] a memory communicatively connected to the at least one processor; wherein,

[0035] The memory stores a computer program that can be executed by the at least one processor. The computer program is executed by the at least one processor to enable the at least one processor to perform the method according to any embodiment of the present invention.

[0036] According to another aspect of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the method described in any embodiment of the present invention when executed.

[0037] The technical solution of the embodiment of the present invention obtains a target classifier by conducting targeted training on different types of gears, and outputs unified shooting light source parameters for gears of the same type through the target classifier, thereby ensuring the uniform standardization of the optical environment of gears of the same type during image acquisition, thereby improving the uniformity and accuracy of the AI ​​algorithm detection effect.

[0038] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0040] Figure 1 This is a flowchart of an image standardization acquisition method provided in accordance with the first embodiment of the present invention;

[0041] Figure 2 This is a structural diagram of an optical collection component provided according to Embodiment 1 of the present invention;

[0042] Figure 3 This is a schematic structural diagram of a detection device provided according to the first embodiment of the present invention;

[0043] Figure 4 This is a flowchart of an image standardization acquisition method provided according to the second embodiment of the present invention;

[0044] Figure 5 This is a structural diagram of an image annotation acquisition device provided according to a third embodiment of the present invention;

[0045] Figure 6 It is a structural diagram of an electronic device provided by the fourth embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0047] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention 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 numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0048] Example 1

[0049] Figure 1 A flowchart of a method for image standardization acquisition is provided for the first embodiment of the present invention. This embodiment is applicable to the case of performing image standardization acquisition on gears. The method can be executed by an image standardization acquisition device, which can be implemented in the form of hardware and / or software. Figure 1 As shown, the method includes:

[0050] Step S101: Acquire key element training data of sample gears.

[0051] Optionally, key element training data of the sample gear is obtained, including: obtaining accumulated data of the historical experimental process of the sample gear, and using a specified algorithm to perform data association analysis to extract key element data; dividing the extracted key element data according to proportion to obtain key element training data and key element test data.

[0052] Specifically, in this embodiment, the accumulated data of the historical experimental process of the sample gear is extracted based on past project experience, and then the accumulated data of the historical experimental process is preprocessed by cleaning and completion. Then, a specified algorithm, such as the apriori operator and the Eclat operator, is used to perform correlation analysis on the data to extract key element data. The following Table 1 shows an example of the extracted key element data:

[0053] Table 1

[0054]

[0055] Among them, the key element data extracted include sample gear parameters such as tilt angle, tooth height, tooth width and diameter, sample gear shooting light source parameters such as light source angle and light source brightness, and sample gear image parameters such as image grayscale value, image contrast and defect grayscale value. In this embodiment, corresponding standard ranges are pre-set for image grayscale value, image contrast and defect grayscale value. Therefore, the meaning of True in the last three columns of Table 1 is that the parameters of the acquired gear image are within the corresponding standard range, and the meaning of False is that the parameters of the acquired gear image are not within the corresponding standard range. For example, the pre-set standard range for image grayscale value is [50 220], the pre-set standard range for image contrast is [0.2 0.7], and the pre-set standard range for defect grayscale value is [150 200]. Of course, this embodiment is only an example and does not limit the standard range corresponding to each parameter. The user can pre-set it according to the requirements of detection accuracy. When the image grayscale value of the gear image captured for assembly Gear-1 is 65, the corresponding image grayscale value output result is True. Of course, this embodiment uses image grayscale values ​​as an example for illustration. The methods for determining the output results of image contrast and defect grayscale values ​​are generally similar and will not be further elaborated in this embodiment. Of course, this embodiment is merely illustrative and does not limit the specific types of key element data. Furthermore, the sample gear image parameters are determined jointly by an AI model and a machine vision algorithm, eliminating traditional manual visual judgment and ensuring an objective, scientific, and rational determination method.

[0056] It is worth mentioning that in this embodiment, the acquired key element data will be divided. For example, when it is determined that there are 1,000 pieces of data information shown in Table 1 above, 80% of the data will be used as key element training data, that is, 800 pieces, and 20% of the data will be used as key element test data, that is, 200 pieces. Of course, this embodiment is only an example for illustration and does not limit the specific rules for data division.

[0057] Step S102: input the key element training data into different types of classifiers for training, and select a target classifier from the trained classifiers.

[0058] Optionally, the key element training data is input into different types of classifiers for training, and a target classifier is screened out from the trained classifiers, including: inputting the key element training data into different types of classifiers so that each classifier is synchronously trained based on the input key element training data, wherein the classifiers include logistic regression classifiers, random forest classifiers, naive Bayes classifiers, and K-nearest neighbor classifiers; inputting the key element test data into each trained classifier, and scoring the classifiers according to the output results of each classifier according to the specified evaluation indicators; obtaining the scoring results of each classifier, and taking the classifier with the highest scoring result as the target classifier.

[0059] Specifically, in this embodiment, the 800 key element training data obtained above are input into different types of classifiers for training, such as logistic regression classifier, random forest classifier, naive Bayes classifier, and K-nearest neighbor classifier. Among them, the regular term type of the logistic regression classifier is selected as L2, whether to use the dual formula is selected as false, and the maximum number of iterations is selected as 80; the data purity index of the random forest classifier is selected as gini, the maximum depth is 30, the pruning coefficient is selected as 2, and the maximum number of leaf nodes is selected as 18; the mode of the naive Bayes classifier is selected as MultinomialNB; the nearest neighbor algorithm of the K-nearest neighbor classifier is auto, the leaf node size is 30, the distance metric parameter is 5, and the weight function is uniform. Of course, this embodiment is only an example, and does not limit the type of classifier or the specific working parameters of each classifier.

[0060] Among them, each classifier will be trained synchronously based on the input key element training data, and after the training is completed, the key element test data will be input into each classifier, and the output results of each classifier will be scored according to the specified evaluation indicators, such as confusion matrix or model evaluation indicators. The specific type of the specified evaluation indicator is not limited in this embodiment. And after obtaining the scoring results of each classifier, for example, logistic regression classifier -10 points, random forest classifier -9 points, naive Bayes classifier -8 points, K nearest neighbor classifier -7 points, the separator with the highest scoring result, that is, the logistic regression classifier, will be used as the target classifier. Of course, this embodiment is only an example and does not limit the specific type of the target classifier.

[0061] Step S103: input the measurement parameters of the gear to be inspected into the target classifier, and obtain the shooting light source parameters output by the target classifier.

[0062] Optionally, the measurement parameters of the gear to be detected are input into a target classifier, and the shooting light source parameters output by the target classifier are obtained, including: inputting the measurement parameters of the gear to be detected into the target classifier, wherein the measurement parameters include the inclination angle, tooth height, tooth width and diameter; determining the gear type based on the measurement parameters through the target classifier, and outputting shooting light source parameters matching the gear type.

[0063] Among them, in this embodiment, through continuous parameter adjustment and test verification during the training process, the gears are mainly solidified into several categories, and the corresponding optical environment is designed according to the classifier classification. Therefore, when there is a new detection task, it is only necessary to extract the corresponding gear parameters, such as the inclination angle, tooth height, tooth width and diameter, etc., and perform type identification through the classifier, and output the shooting light source parameters matching the gear type according to the identified type, and the same shooting light source parameters will be output for the same type of gears, thereby ensuring the consistency of the optical environment of the same type of gears.

[0064] For example, the inclination angle of gear 1 is 45 degrees, the tooth height is 6 mm, the tooth width is 3.5 mm, and the diameter is 155 mm. The inclination angle of gear 2 is 33 degrees, the tooth height is 5 mm, the tooth width is 3 mm, and the diameter is 134 mm. Although the parameters of the two parts are different, the classifier determines that the two parts are of the same type after identification. At this time, the same shooting light source parameters will be output for the above two parts.

[0065] Specifically, the classifier in this embodiment classifies different gears into one category, thereby outputting the same shooting light source parameters. The advantage is that there is no need to design shooting parameters for each gear separately, which improves efficiency. New gears can automatically match existing environmental parameters through classification. However, the problem is that classification errors may lead to unreasonable parameter sharing, resulting in unreasonable matching shooting light source parameters. In order to solve this problem, the present application adopts an output verification method, that is, when the classifier is scored by evaluation using the aforementioned 200 test data during the training phase, a method is added, namely input feature optimization, that is, standardization of parameters of different dimensions to eliminate dimensional differences and avoid the model being dominated by features with a large numerical range. In addition, only the parameter values ​​are standardized, and no artificial screening of high-discrimination parameters is performed for feature processing. The original parameters are not artificially transformed or combined to generate new features that can better reflect the differences. Only the parameter values ​​are processed data-wise to eliminate dimensional differences, thereby avoiding the influence of human factors on the model output.

[0066] Optionally, before inputting the key element training data into different types of classifiers, it also includes: obtaining the type of each key element in the key element training data, and determining the data normalization method that matches each key element, wherein the key elements of continuous parameters are matched with the standard score normalization method, and the key elements of non-continuous parameters are matched with the quantile normalization method; processing each key element in the key element training data according to the matched normalization method to obtain the normalized key element training data.

[0067] Optionally, each key element in the key element training data is processed in a matching normalization manner to obtain the normalized key element training data, including: when the key element is determined to be a continuous parameter, calculating the difference between the key element and the median of the key element, and taking the ratio of the difference to the interquartile range of the key element as the normalized key element training data, wherein the continuous parameter at least includes the brightness of the light source; when the key element is determined to be a non-continuous parameter, calculating the ranking of the key element in the training data, calculating the ratio of the ranking to the total number of training samples plus 1, and taking the calculation result of the inverse function of the standard normal distribution of the ratio as the normalized key element training data, wherein the non-continuous parameters at least include tooth height, tooth width and diameter.

[0068] Specifically, this embodiment adopts a hybrid strategy: Z-score normalization (i.e., standard score normalization) is used for continuous parameters such as light source brightness; quantile normalization is used for tooth height, tooth width, or diameter; and sine / cosine transformation is used instead of normalization for angles. The following formula (1) shows an example of Z-score normalization:

[0069] (1)

[0070] Where Median represents the median, X represents the original eigenvalue, that is, a single data point that is not normalized, IQR represents the interquartile range, and X new Represents the new eigenvalue after normalization.

[0071] The following formula (2) shows an example of quantile normalization:

[0072] (2)

[0073] Among them, X represents the original eigenvalue, that is, a single data point that has not been normalized, rank(X) represents the ranking of the current data X in the data set, that is, the position number after ascending order, N represents the total number of samples in the data set, rank(X) represents the quantile, Φ -1 Represents the inverse function of the standard normal distribution.

[0074] In addition, the sine can be specifically used To express it, cosine is expressed as Of course, this embodiment is only an example, and does not limit the specific normalization method used for different types of data. It should be noted that in the preprocessing stage, before model training, the statistics of all training data are pre-calculated, such as mean / standard deviation, minimum / maximum value, etc., while the sine / cosine transformation does not require the storage of additional statistics. In the inference stage, each new sample needs to be converted using the statistics calculated in the training stage, and if the eigenvalues ​​encountered in the inference stage exceed the range of the training set, for Min-Max normalization, it is necessary to force truncation to the [0,1] interval to retain the data.

[0075] Step S104: Control the optical acquisition component to acquire a standardized image of the gear to be inspected based on the shooting light source parameters.

[0076] Optionally, the optical acquisition component is controlled to perform standardized image acquisition on the gear to be detected based on the shooting light source parameters, including: configuring the shooting light source parameters to the optical acquisition component, and moving the optical acquisition component to the target position through the robotic arm; controlling the optical acquisition component to perform standardized image acquisition on the gear to be detected at the target position.

[0077] Optionally, controlling the optical acquisition component to perform standardized image acquisition on the gear to be detected at the target position includes: controlling the rotating clamping mechanism to rotate the gear to be detected to determine different placement angles of the gear to be detected; controlling the optical acquisition component to perform standardized image acquisition on the gear to be detected at different placement angles at the target position.

[0078] In this embodiment, the shooting light source parameters output by the classifier are configured on the optical acquisition component, such as Figure 2 The figure shows the structure of the optical acquisition component, which includes an area array camera, a line array camera and three light sources. The shooting light source parameters are specifically configured for the three light sources, and the optical acquisition component is specifically installed on the detection equipment through a robotic arm, such as Figure 3 The figure shows a schematic diagram of the structure of the detection equipment, which mainly includes an optical acquisition component, a robotic arm connected to the optical acquisition component, a rotating clamping mechanism for fixing the gear, and a host computer for receiving and performing image processing, etc. Of course, this embodiment is only an example and does not limit the specific structure of the detection equipment.

[0079] Specifically, when the acquisition and detection equipment is inspecting the gear, it is specifically through manual loading, and then the gear is pressed by the rotating clamping mechanism. The rotating clamping mechanism is controlled to rotate the gear to be inspected to rotate the gear to be inspected to different placement angles. After each placement position is determined, the optical acquisition component is moved to the target position by the robotic arm, and the gear with the determined placement position is photographed at the target position. In this embodiment, the placement angle can specifically be the axial surface, tooth surface and end surface, so that the gear can be photographed from different angles, thereby obtaining multi-angle images of the gear to ensure the accuracy of subsequent defect detection. Because the specific designed optical environment can unify the imaging standards of gears of the same type, it achieves compatibility of the optical environment of gears of the same type and AI algorithm compatibility. Combined with the above-mentioned tooling and robotic arm method, it achieves compatibility in mechanical structure. And when encountering new market needs, it only needs to extract the corresponding gear parameters and input the parameters into the classifier to obtain the corresponding group and the corresponding AI model. Therefore, it solves the gear compatibility problem in the market, greatly improves the efficiency of equipment research and development, and reduces development costs.

[0080] In the embodiment of the present application, a target classifier is obtained by conducting targeted training for different types of gears, and a unified shooting light source parameter is output for the same type of gears through the target classifier, thereby ensuring the uniform standardization of the optical environment of the same type of gears during image acquisition, thereby improving the uniformity and accuracy of the AI ​​algorithm detection effect.

[0081] Example 2

[0082] Figure 4 The second embodiment of the present invention provides a flowchart of a method for standardizing image acquisition. This embodiment is based on the above embodiment. After controlling the optical acquisition component to acquire a standardized image of the gear to be inspected based on the shooting light source parameters, it also includes: performing defect detection on the standardized image of the gear to be inspected, generating a defect detection report based on the defect detection results, and visually displaying the defect detection report. Figure 4 As shown, the method includes:

[0083] Step S201: Acquire key element training data of sample gears.

[0084] Optionally, key element training data of the sample gear is obtained, including: obtaining accumulated data of the historical experimental process of the sample gear, and using a specified algorithm to perform data association analysis to extract key element data; dividing the extracted key element data according to proportion to obtain key element training data and key element test data.

[0085] Step S202: input the key element training data into different types of classifiers for training, and select a target classifier from the trained classifiers.

[0086] Optionally, the key element training data is input into different types of classifiers for training, and a target classifier is screened out from the trained classifiers, including: inputting the key element training data into different types of classifiers so that each classifier is synchronously trained based on the input key element training data, wherein the classifiers include logistic regression classifiers, random forest classifiers, naive Bayes classifiers, and K-nearest neighbor classifiers; inputting the key element test data into each trained classifier, and scoring the classifiers according to the output results of each classifier according to the specified evaluation indicators; obtaining the scoring results of each classifier, and taking the classifier with the highest scoring result as the target classifier.

[0087] Step S203: inputting the measurement parameters of the gear to be inspected into the target classifier, and obtaining the shooting light source parameters output by the target classifier.

[0088] Optionally, the measurement parameters of the gear to be detected are input into a target classifier, and the shooting light source parameters output by the target classifier are obtained, including: inputting the measurement parameters of the gear to be detected into the target classifier, wherein the measurement parameters include the inclination angle, tooth height, tooth width and diameter; determining the gear type based on the measurement parameters through the target classifier, and outputting shooting light source parameters matching the gear type.

[0089] Step S204: Control the optical acquisition component to acquire a standardized image of the gear to be inspected based on the shooting light source parameters.

[0090] Optionally, the optical acquisition component is controlled to perform standardized image acquisition on the gear to be detected based on the shooting light source parameters, including: configuring the shooting light source parameters to the optical acquisition component, and moving the optical acquisition component to the target position through the robotic arm; controlling the optical acquisition component to perform standardized image acquisition on the gear to be detected at the target position.

[0091] Optionally, controlling the optical acquisition component to perform standardized image acquisition on the gear to be detected at the target position includes: controlling the rotating clamping mechanism to rotate the gear to be detected to determine different placement angles of the gear to be detected; controlling the optical acquisition component to perform standardized image acquisition on the gear to be detected at different placement angles at the target position.

[0092] Step S205 , performing defect detection on the standardized image of the gear to be inspected, generating a defect detection report according to the defect detection result, and visually displaying the defect detection report.

[0093] Among them, in this embodiment, defect detection will be performed on the standardized images taken of the gears to be inspected. Specifically, the standardized images are inferred through a built-in algorithm to identify defects such as black skin, missing processing, scratches, cracks and bumps in the images. Since the same type of parts use the same light source in this embodiment, the uniformity of the captured images is guaranteed. Since the specific method of defect detection on the image is not the focus of this application, it will not be repeated in this embodiment.

[0094] It should be noted that in this embodiment, after defect detection is performed on the image, a defect detection report will be generated based on the defect detection results. The defect detection report mainly includes the time of defect detection, the results of defect detection, and when multiple gears are detected at the same time, it also includes the defect rate, etc. Of course, this embodiment is only an example and does not limit the specific content contained in the defect detection report. In this embodiment, the defect detection report will be visualized on the human-computer interaction interface to facilitate users to obtain the gear defect detection results intuitively and quickly. When the defect rate value in the defect detection report exceeds the specified value, it means that the overall quality of the gears produced in the current batch is poor. At this time, an alarm prompt will be generated and the alarm prompt will be displayed in the form of voice or image. This makes it convenient for gear manufacturers to urgently terminate the production of the current batch of gears according to the alarm prompt, and to check the problems that arise in the production process to ensure that the quality of the gears produced meets the standards.

[0095] In the implementation mode of the present application, a target classifier is obtained by conducting targeted training for different types of gears, and a unified shooting light source parameter is output for the same type of gears through the target classifier, thereby ensuring the uniform standardization of the optical environment of the same type of gears during image acquisition, thereby improving the uniformity and accuracy of the AI ​​algorithm detection effect.

[0096] Example 3

[0097] Figure 5 The third embodiment of the present invention provides a schematic diagram of the structure of an image standardization acquisition device. Figure 5 As shown, the device includes: a key element training data acquisition module 310, a target classifier screening module 320, a shooting light source parameter acquisition module 330 and a standardized image acquisition module 340.

[0098] The key element training data acquisition module 310 is used to acquire key element training data of the sample gear, wherein the key element training data includes sample gear parameters, sample gear shooting light source parameters and sample gear image parameters;

[0099] The target classifier screening module 320 is used to input key element training data into different types of classifiers for training, and screen out target classifiers from the trained classifiers;

[0100] The shooting light source parameter acquisition module 330 is used to input the measurement parameters of the gear to be detected into the target classifier and obtain the shooting light source parameters output by the target classifier;

[0101] The standardized image acquisition module 340 is used to control the optical acquisition component to acquire a standardized image of the gear to be inspected based on the shooting light source parameters.

[0102] Optional, key element training data acquisition module, used to obtain the accumulated data of the historical experimental process of the sample gear, and use the specified algorithm to perform data association analysis to extract key element data;

[0103] The extracted key element data are divided according to proportion to obtain key element training data and key element test data.

[0104] Optionally, a target classifier screening module is used to input key element training data into different types of classifiers so that each classifier is trained synchronously based on the input key element training data, wherein the classifiers include logistic regression classifiers, random forest classifiers, naive Bayes classifiers, and K-nearest neighbor classifiers;

[0105] Input the key element test data into each trained classifier, and score the classifier according to the output results and the specified evaluation indicators;

[0106] Get the scoring results of each classifier and use the classifier with the highest scoring result as the target classifier.

[0107] Optionally, a shooting light source parameter acquisition module is used to input measurement parameters of the gear to be inspected into the target classifier, wherein the measurement parameters include tilt angle, tooth height, tooth width and diameter;

[0108] The target classifier determines the gear type based on the measurement parameters and outputs the shooting light source parameters that match the gear type.

[0109] Optionally, a standardized image acquisition module is used to configure the shooting light source parameters to the optical acquisition component and move the optical acquisition component to the target position through the robotic arm;

[0110] The optical acquisition component is controlled to perform standardized image acquisition on the gear to be detected at a target position.

[0111] Optionally, the standardized image acquisition module is further used to control the rotary clamping mechanism to rotate the gear to be inspected to determine different placement angles of the gear to be inspected;

[0112] The optical acquisition component is controlled to perform standardized image acquisition on the gears to be inspected at different placement angles at the target position.

[0113] Optionally, the device further comprises a defect detection module for performing defect detection on the standardized image of the gear to be inspected;

[0114] Generate a defect detection report based on the defect detection results and display the defect detection report visually.

[0115] The image standardization acquisition device provided in the embodiment of the present invention can execute the image standardization acquisition method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0116] Example 4

[0117] Figure 6 The following is a schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention. The electronic device is intended to represent various forms of digital computers. The electronic device 10 includes at least one processor 11 and memory, such as a read-only memory (ROM) 12 and a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer programs stored in the ROM 12 or loaded from the storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0118] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0119] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, device, or apparatus. A computer-readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or apparatus, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0120] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0121] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A method for standardizing image acquisition, characterized in that: include: Acquire key element training data of a sample gear, wherein the key element training data includes sample gear parameters, sample gear shooting light source parameters, and sample gear image parameters; Inputting the key element training data into different types of classifiers for training, and screening out a target classifier from the trained classifiers; Inputting the measurement parameters of the gear to be inspected into the target classifier, and obtaining the shooting light source parameters output by the target classifier; The optical acquisition component is controlled to perform standardized image acquisition on the gear to be detected based on the shooting light source parameters.

2. The method according to claim 1, characterized in that The step of obtaining the key element training data of the sample gear includes: Obtain the accumulated data of the historical experimental process of the sample gear, and use the specified algorithm to perform data association analysis to extract key element data; The extracted key element data are divided according to a ratio to obtain the key element training data and key element test data.

3. The method according to claim 2, characterized in that The step of inputting the key element training data into different types of classifiers for training, and selecting a target classifier from the trained classifiers, comprises: Inputting the key element training data into different types of classifiers so that each classifier is synchronously trained based on the input key element training data, wherein the classifiers include a logistic regression classifier, a random forest classifier, a naive Bayes classifier, and a K-nearest neighbor classifier; Input the key factor test data into each trained classifier, and score the classifier according to the output results of each classifier according to the specified evaluation indicators; The scoring results of each classifier are obtained, and the classifier with the highest scoring result is used as the target classifier.

4. The method according to claim 3, characterized in that Before inputting the key element training data into different types of classifiers, the method further includes: Obtaining the type of each key element in the key element training data, and determining a data normalization method that matches each key element, wherein the key elements of continuous parameters are matched with a standard score normalization method, and the key elements of discontinuous parameters are matched with a quantile normalization method; Each key element in the key element training data is processed according to a matching normalization method to obtain the normalized key element training data.

5. The method according to claim 4, characterized in that The step of processing each key element in each key element training data according to a matching normalization method to obtain the normalized key element training data includes: When it is determined that the key factor is a continuous parameter, the difference between the key factor and the key factor median is calculated, and the ratio of the difference to the key factor interquartile range is used as the normalized key factor training data, wherein the continuous parameter includes at least the light source brightness; When it is determined that the key factor is a non-continuous parameter, the ranking of the key factor in the training data is calculated, the ratio of the ranking to the total number of training samples plus 1 is calculated, and the calculation result of the inverse function of the standard normal distribution of the ratio is used as the normalized key factor training data, wherein the non-continuous parameters include at least tooth height, tooth width and diameter.

6. The method according to claim 1, wherein The step of inputting the measurement parameters of the gear to be detected into the target classifier and obtaining the shooting light source parameters output by the target classifier includes: Inputting measurement parameters of the gear to be inspected into the target classifier, wherein the measurement parameters include at least tilt angle, tooth height, tooth width, diameter, surface material and roughness; The target classifier determines a gear type based on the measurement parameters, and outputs the photographing light source parameters matching the gear type.

7. The method according to claim 1, characterized in that The controlling optical acquisition component to perform standardized image acquisition on the gear to be detected based on the shooting light source parameters includes: Configuring the shooting light source parameters to the optical acquisition component, and moving the optical acquisition component to a target position by a robotic arm; The optical acquisition component is controlled to acquire a standardized image of the gear to be detected at the target position.

8. The method according to claim 7, characterized in that The controlling the optical acquisition component to acquire a standardized image of the gear to be detected at the target position includes: Controlling the rotary clamping mechanism to rotate the gear to be detected to determine different placement angles of the gear to be detected; The optical acquisition component is controlled to perform standardized image acquisition on the gears to be detected at different placement angles at the target position.

9. An image standardization acquisition device, characterized in that: include: A key element training data acquisition module is used to acquire key element training data of a sample gear, wherein the key element training data includes sample gear parameters, sample gear shooting light source parameters, and sample gear image parameters; A target classifier screening module is used to input the key element training data into different types of classifiers for training, and to screen out target classifiers from the trained classifiers; a shooting light source parameter acquisition module, configured to input the measurement parameters of the gear to be detected into the target classifier and acquire the shooting light source parameters output by the target classifier; The standardized image acquisition module is used to control the optical acquisition component to perform standardized image acquisition on the gear to be detected based on the shooting light source parameters.

10. An electronic device, characterized in that: The electronic device comprises: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program to be executed by the at least one processor, where the computer program is executed by the at least one processor so as to enable the at least one processor to perform the method according to any one of claims 1 to 8.

Images