Data transmission method and system of industrial visual cloud service platform

Abstract

The invention relates to the technical field of edge data transmission, in particular to a data transmission method and system of an industrial visual cloud service platform. According to the method, the historical computing power resource condition of the cloud platform is evaluated by analyzing the frame image fuzzy condition and the modeling quality inspection time length of the inspected workpiece in the quality inspection process, reference is provided for evaluating the computing power resource condition of the current cloud platform, and then the frame image compression rate is dynamically adjusted by combining the frame image fuzzy condition of the current workpiece to be inspected in the quality inspection process, so that the quality inspection accuracy of the workpiece to be inspected is improved. The transmission bandwidth is reduced, the working condition of the cloud server is considered in real time, the transmission effect of the edge data is improved, and then the subsequent quality inspection effect is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of edge data transmission, and particularly relates to a data transmission method and system for an industrial vision cloud service platform. Background Art

[0002] Industrial vision is a technology that uses computer vision to analyze and process images collected during the industrial production process, and is widely used in fields such as production line inspection, quality control, and automated operation. However, a large amount of detection data will be generated during the quality inspection application process of industrial vision. The computing power at the edge may not be sufficient to process it in real time. Therefore, the detection data at the edge is usually compressed and transmitted to the cloud service platform for storage and processing, which can improve the quality inspection efficiency and store and share the quality inspection data at the same time.

[0003] However, the computing capacity of the cloud server is dynamically changing. An excessively high compression rate will occupy a large amount of decompression resources, which may lead to a decline in the performance stability of the cloud service platform when the computing capacity of the cloud server is low, and may also affect the accuracy of the detection data; when the computing capacity of the cloud server is high, an excessively low compression rate may lead to a waste of transmission bandwidth, and may also lead to too long transmission time or transmission failure, thus affecting the quality inspection efficiency; therefore, it is crucial to select an appropriate compression rate, and an inappropriate compression rate will seriously affect the transmission effect of the detection data at the edge. Summary of the Invention

[0004] In order to solve the technical problem of poor transmission effect of the monitoring data at the edge in the prior art, the purpose of the present invention is to provide a data transmission method and system for an industrial vision cloud service platform, and the specific technical solutions adopted are as follows:

[0005] A data transmission method for an industrial vision cloud service platform, the method includes:

[0006] Obtain the video frame sequences of each workpiece at different quality inspection angles during the quality inspection process, where the workpiece includes the current workpiece to be inspected and the inspected workpieces; obtain the modeling detection duration of each inspected workpiece on the cloud platform;

[0007] In each video frame sequence, perform matching tracking on different frame images, and obtain the relative change parameter of each frame image according to the difference of the matching pixel points between adjacent frame images; during the quality inspection process of each workpiece, obtain the frame blurring coefficient of the frame images under the same frame image number according to the relative change parameters of the frame images in different video frame sequences.

[0008] During the quality inspection process of each inspected workpiece, according to the frame image blur coefficient and the modeling detection duration under all frame image serial numbers, construct the cloud platform's cloud performance parameters; during the quality inspection process of the current workpiece to be inspected, according to the frame image blur coefficient and the preset compression rate range under each frame image serial number, and in combination with the cloud performance parameters corresponding to the quality inspection processes of all inspected workpieces, obtain the compression rate of each frame image in each video frame sequence during the current quality inspection process;

[0009] Compress the corresponding frame image according to the compression rate and transmit the compression result to the cloud platform.

[0010] Furthermore, the method for obtaining the relative change parameter includes:

[0011] In each video frame sequence, construct a rectangular coordinate system with the center of each frame image as the origin, and obtain the position coordinates of each pixel point in each frame image; between each frame image and the previous adjacent frame image, take the Euclidean distance between the position coordinates of each pair of matching pixel points as the change sub-parameter, and take the mean value of the change sub-parameters between all matching pixel points as the relative change parameter between each frame image and the previous adjacent frame image.

[0012] Furthermore, the method for obtaining the frame image blur coefficient includes:

[0013] During the quality inspection process of each workpiece, between every two video frame sequences, take the absolute value of the difference between the relative change parameters of the frame images under the same frame image serial number as the blur sub-parameter corresponding to the frame image serial number;

[0014] Under each frame image serial number, comprehensively obtain the frame image blur coefficient by integrating the blur sub-parameters between all pairs of different video frame sequences during the workpiece quality inspection process.

[0015] Furthermore, the method for obtaining the cloud performance parameters includes:

[0016] During the quality inspection process of each inspected workpiece, take the negative correlation mapping result of the sum value of the frame image blur coefficients under all frame image serial numbers as the first performance parameter, and take the negative correlation mapping result of the modeling detection duration as the second performance parameter;

[0017] Take the first performance parameter and the second performance parameter as vector elements to construct a two-dimensional vector, and take the two-dimensional vector as the cloud performance parameters of the cloud platform during the quality inspection process of the corresponding known workpiece.

[0018] Furthermore, the method for obtaining the compression rate includes:

[0019] During the quality inspection process of each inspected workpiece, fuse the first performance parameter and the second performance parameter in the cloud performance parameters of the cloud platform to obtain the calculation redundancy parameter of the cloud platform in each quality inspection process; according to the calculation redundancy parameters in the quality inspection processes of all inspected workpieces, obtain the cloud computing capacity coefficient in the quality inspection process of the current workpiece to be inspected.

[0020] Fuse the negative correlation mapping result of the cloud computing capacity coefficient in the quality inspection process of the current workpiece to be inspected with the frame map blur coefficient under each frame map serial number to obtain the frame map compression weight under each frame map serial number in all video frame sequences of the current workpiece to be inspected.

[0021] According to the frame map compression weight and the preset compression rate range, obtain the compression rate of each frame map in each video frame sequence in the current quality inspection process.

[0022] Further, the method for obtaining the cloud computing capacity coefficient includes:

[0023] Take the calculation redundancy parameter of the cloud platform in the previous adjacent quality inspection process of the current workpiece to be inspected as the current calculation redundancy parameter; according to the fluctuation deviation of the calculation redundancy parameter in the quality inspection process of each inspected workpiece relative to the average level, obtain the calculation redundancy stability parameter.

[0024] Fuse the current calculation redundancy parameter and the calculation redundancy stability parameter to obtain the cloud computing capacity coefficient in the quality inspection process of the current workpiece to be inspected.

[0025] Further, the method for obtaining the compression rate of each frame map in each video frame sequence in the current quality inspection process according to the frame map compression weight and the preset compression rate range includes:

[0026] Take the left endpoint of the interval of the preset compression rate range as the basic compression rate, use the frame map compression weight to weight the interval length of the preset compression rate range, and take the weighted result as the adjusted compression rate; add the basic compression rate and the adjusted compression rate to obtain the compression rate of the frame map under the corresponding frame map serial number.

[0027] Further, the method for matching and tracking different frame maps includes:

[0028] Obtain the workpiece area contour in each frame map and obtain all corner points on the workpiece area contour; in each video frame sequence, take any frame map as the target frame map and any pixel point in the target frame map as the target corner point, and find the nearest neighbor corner point of the target corner point in each non-target frame map based on the KD tree, and take the target corner point and its nearest neighbor corner point in each non-target frame map as a series of matching points.

[0029] Further, the method for obtaining the corner points includes:

[0030] Based on the Harris corner detection algorithm, all corner points are obtained.

[0031] The present invention also provides a data transmission system for an industrial vision cloud service platform, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of a data transmission method for an industrial vision cloud service platform are implemented.

[0032] The present invention has the following beneficial effects:

[0033] The present invention first obtains the video frame sequences of each workpiece at different quality inspection angles during the quality inspection process, and obtains the modeling detection duration of each inspected workpiece, so as to prepare for subsequent evaluation of the cloud performance parameters of the cloud platform; then in each video frame sequence, according to the difference of the matching pixel points between adjacent frame images, the relative change parameter of each frame image is evaluated, and further the relative change parameters of the frame images with the same frame image number in all video frame sequences during the quality inspection process of each workpiece are comprehensively evaluated to evaluate the frame image blur coefficient of the frame images under each frame image number. The frame image blur coefficient will affect the subsequent adjustment of the compression ratio to a certain extent; further during the quality inspection process of each inspected workpiece, the cloud performance parameters of the cloud platform are constructed to provide a historical reference for subsequent analysis of the computing power of the cloud platform; then during the quality inspection process of the current workpiece to be inspected, according to the frame image blur coefficient and the preset compression ratio range under each frame image number, combined with the cloud performance parameters corresponding to the quality inspection processes of all inspected workpieces, the compression ratio of each frame image in each video frame sequence during the current quality inspection process is obtained; finally, the corresponding frame images are compressed and transmitted according to the compression ratio. By analyzing the frame image blur situation and the modeling quality inspection duration of the inspected workpieces during the quality inspection process, the present invention evaluates the historical computing power resources of the cloud platform as a reference for evaluating the current computing power resources of the cloud platform, and then dynamically adjusts the frame image compression ratio in combination with the frame image blur situation of the current workpiece to be inspected during the quality inspection process. While reducing the transmission bandwidth, it also takes into account the working conditions of the cloud server in real time, improves the transmission effect of the edge data, and further improves the subsequent quality inspection effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0035] Figure 1 It is a flowchart of a data transmission method for an industrial vision cloud service platform provided by an embodiment of the present invention;

[0036] Figure 2 Flowchart of a method for obtaining a compression ratio provided by an embodiment of the present invention. Detailed implementation manners

[0037] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following describes in detail the specific implementation manners, structures, features, and effects of a data transmission method and system for an industrial vision cloud service platform proposed according to the present invention in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0038] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0039] The following specifically describes the specific solutions of a data transmission method and system for an industrial vision cloud service platform provided by the present invention in conjunction with the accompanying drawings.

[0040] Please refer to Figure 1 , which shows a flowchart of a data transmission method for an industrial vision cloud service platform provided by an embodiment of the present invention, specifically including:

[0041] Step S1, obtain the video frame sequences of each workpiece at different quality inspection angles during the quality inspection process, where the workpieces include the current workpiece to be inspected and the inspected workpieces; obtain the modeling inspection duration of each inspected workpiece on the cloud platform.

[0042] In order to efficiently transmit quality inspection data such as workpiece surface images collected at the automatic workpiece surface quality inspection device (hereinafter referred to as the quality inspection terminal) to the cloud platform for modeling quality inspection; an embodiment of the present invention first collects the workpiece surface videos of each workpiece at different quality inspection angles during the quality inspection process, where the workpieces include the current workpiece to be inspected and the inspected workpieces, and different quality inspection angles refer to different shooting perspectives of the workpiece. All quality inspection angles should preferably cover the entire outer contour of the workpiece for subsequent modeling quality inspection.

[0043] It should be noted that an embodiment of the present invention mainly focuses on the surface quality inspection of the engine turbocharger housing workpiece; in other embodiments, the implementer also conducts surface quality inspections on other types of workpieces, and the compression transmission method and surface quality inspection method of the quality inspection data of different workpieces are the same.

[0044] It should be noted that it is a well-known technology to collect videos of workpieces at different shooting perspectives through existing quality inspection terminals, and thus it will not be elaborated here.

[0045] In an embodiment of the present invention, after obtaining the videos of each workpiece at different quality inspection angles during the quality inspection process, each video is framed at a preset frame rate of 120fps to obtain all the frame images in each video, and then the frame images are sorted according to the acquisition time sequence to construct a video frame sequence corresponding to each video; the video frame sequence collected by the quality inspection terminal and the parameters involved in subsequent analysis are used as edge-end data, and then compressed and transmitted to the cloud platform.

[0046] It should be noted that the lengths of the videos at different quality inspection angles may be different, and the lengths of the corresponding video frame sequences may also be different. For the convenience of subsequent analysis, the length of the video frame sequence is defined as the shortest length among all video frame sequences. Without affecting the modeling effect, the frame images beyond the shortest length are discarded; the implementer can also define the preset frame rate by himself, but at least it is necessary to ensure that the frame rates of the video frame sequences at different quality inspection angles collected synchronously are the same.

[0047] In an embodiment of the present invention, after obtaining all the video frame sequences of each inspected workpiece, these video frame sequences can be uploaded to the cloud server for 3D modeling of the inspected workpiece, providing an analysis and comparison basis for subsequent quality control, defect detection, surface roughness analysis, etc.; further obtain the modeling detection duration during the 3D modeling process to prepare for subsequent evaluation of the cloud performance parameters of the cloud platform; so as to adaptively adjust the quality inspection data transmission and quality inspection process according to the cloud performance parameters of the cloud platform without affecting the quality inspection effect.

[0048] It should be noted that 3D modeling of workpieces based on video frame sequences from different angles is a well-known technology and will not be elaborated here.

[0049] Step S2, in each video frame sequence, perform matching tracking on different frame images, and obtain the relative change parameters of each frame image according to the difference of the matching pixel points between adjacent frame images; during the quality inspection process of each workpiece, obtain the frame image blur coefficient of each frame image number according to the relative change parameters of the frame images with the same frame image number in different video frame sequences.

[0050] Considering that when the robotic arm grabs and transfers the workpiece, the workpiece moves with the robotic arm, resulting in edge blurring of the video frame images in the captured video, and since the contour of the workpiece itself may be an irregular shape, it will further deepen the severity of frame image blurring; the higher the blurring degree of the frame image, the smaller the contribution to subsequent modeling. Therefore, the compression ratio can be appropriately increased to balance the transmission pressure in combination with the cloud performance of the cloud platform;

[0051] Also considering that during the process of the robotic arm grasping and transferring workpieces, the displacement amount is relatively limited. Therefore, during the quality inspection process of each workpiece, the change between adjacent frame images in each video frame sequence should also be relatively limited. If each frame image changes significantly compared to its adjacent frame image, it indicates that there may be blurring.

[0052] Based on this, in the embodiment of the present invention, first, in each video frame sequence, different frame images are matched and tracked, and then the relative change parameters of each frame image are obtained according to the difference of the matching pixel points between adjacent frame images. The relative change parameters reflect the inter-frame change situation of the workpiece during the transfer process. Further, by comprehensively considering the inter-frame change situations in the video frame sequences at all quality inspection angles during the quality inspection process of each workpiece, that is, by comprehensively considering the relative change parameters of the frame images with the same frame image serial number, the frame blurring coefficient of the frame images at each frame image serial number during the quality inspection process is evaluated. The frame blurring coefficient reflects the motion blurring situation at the acquisition moment corresponding to each frame image serial number, and prepares for adjusting the compression and transmission method in combination with the cloud performance situation of the cloud platform in the future.

[0053] Preferably, in an embodiment of the present invention, considering that there may be many corner points in the workpiece, directly performing matching and tracking will consume a large amount of computing resources. Also considering that the contour edge is most likely to be blurred, providing a certain reference value for blurring by comparing the relative changes of the corner points on the contour edge. Therefore, first, the corner points on the contour edge of the workpiece in each frame image are obtained, and then they are matched and tracked to evaluate the inter-frame change situation in the future. It is also considered that the KD tree (K-Dimensional Tree) can efficiently process high-dimensional data and perform fast searches in multi-dimensional spaces, which can improve the matching speed of corner points between different frame images. Based on this, the method for matching and tracking different frame images includes:

[0054] Obtain the workpiece area contour in each frame image and all the corner points on the workpiece area contour. In each video frame sequence, take any frame image as the target frame image and any pixel point in the target frame image as the target corner point. Based on the KD tree, find the nearest neighbor corner point of the target corner point in each non-target frame image, and use the target corner point and its nearest neighbor corner point in each non-target frame image as a series of matching points.

[0055] As an example, first, based on a pre-trained contour annotation model, annotate the workpiece area contour, and at the same time, use the Hessian corner detection algorithm to obtain all the corner points on the workpiece area contour, and obtain the descriptor of each corner point. Then, based on the descriptors of all the corner points in each non-target frame image, construct a KD tree, and query the nearest neighbor corner point of the target corner point in the target frame image in the KD tree corresponding to each non-target frame image, so as to obtain the matching corner point of the target corner point in each non-target frame image. The nearest neighbor distance ratio of the KD tree is set to 0.75, and the implementer can also define it by himself.

[0056] It should be noted that the training and application of the contour annotation model, the acquisition of Hessian corner points and descriptors, and the feature matching based on the KD tree are all prior arts and will not be elaborated here; in other examples, the implementer can also obtain all the edge contours in each frame image through an edge detection algorithm such as the canny algorithm, and further use the SIFT algorithm to obtain the corner points and corresponding descriptors on the edge contours; the implementer can also adjust the feature matching algorithm in combination with the computing power of the quality inspection terminal server, that is, the edge server. For example, directly use the SIFT algorithm for feature matching, or use the optical flow method to track and match pixel points, which are all well-known techniques and will not be elaborated further.

[0057] After obtaining the matching points between different frame images in each video frame sequence, the relative change parameters of each frame image can be calculated.

[0058] Preferably, in an embodiment of the present invention, considering that the displacement amount of the robotic arm is relatively limited and the frame rate is set relatively high, the position difference of the matching corner points between adjacent frame images in each video frame sequence reflects its change situation to a certain extent, and indirectly reflects the blurring effect it suffers; therefore, the method for obtaining the relative change parameters includes:

[0059] In each video frame sequence, a rectangular coordinate system is constructed with the center of each frame image as the origin, and the position coordinates of each pixel point in each frame image are obtained; between each frame image and the previous adjacent frame image, the Euclidean distance between the position coordinates of each pair of matching pixel points is used as the change sub-parameter, and the mean value of the change sub-parameters between all matching pixel points is used as the relative change parameter between each frame image and the previous adjacent frame image.

[0060] In another embodiment of the present invention, between each frame image and the previous adjacent frame image, the absolute value of the difference between the gray values of each pair of matching pixel points is further calculated, and the fusion result of the absolute value of the gray value difference and the Euclidean distance, such as the product, sum value or weighted sum value, is further used as the change sub-parameter, and the relative change parameter is further obtained.

[0061] After obtaining the relative change parameters of each frame image in each video frame sequence, the frame blurring coefficient of each frame image under each frame image serial number during the quality inspection of each workpiece can be further obtained.

[0062] Preferably, in an embodiment of the present invention, the method for obtaining the frame blurring coefficient includes:

[0063] During the quality inspection of each workpiece, between every two video frame sequences, the absolute value of the difference between the relative change parameters of the frame images with the same frame image serial number is used as the blurring sub-parameter corresponding to the frame image serial number;

[0064] Under each frame number, the frame blur coefficients are obtained by synthesizing the blur sub-parameters between all pairs of different video frame sequences during the quality inspection process of the workpiece.

[0065] As an example, during the quality inspection process of each workpiece, first obtain the blur sub-parameters under each frame number between every two video frame sequences; further accumulate the blur sub-parameters under each frame number between all pairwise combinations of video frame sequences, and use the linearly normalized sum value of the accumulated sum as the frame blur coefficient under the corresponding frame number; the greater the difference between pairwise combinations indicates the greater the change difference of the frames at the same acquisition moment under any two quality inspection angles, and the more likely there is an edge blur situation.

[0066] In another embodiment of the present invention, during the quality inspection process of each workpiece, the implementer can also calculate the variance of the relative change parameters of the frames with the same frame number in all video frame sequences, normalize the variance, and obtain the frame blur coefficient under the corresponding frame number; the greater the normalized value of the variance indicates the greater the volatility and consistency of the frame change situations at the same acquisition moment under any different quality inspection angles, and the more likely there is an edge blur situation.

[0067] It should be noted that the implementer can also adopt other normalization means, and can also use other discrete metric parameters such as standard deviation instead of variance, which are all prior arts and will not be elaborated here.

[0068] Step S3, during the quality inspection process of each inspected workpiece, construct the cloud platform's cloud performance parameters according to the frame blur coefficients and the modeling detection duration under all frame numbers; during the quality inspection process of the current workpiece to be inspected, according to the frame blur coefficients and the preset compression rate range under each frame number, combined with the cloud performance parameters corresponding to the quality inspection processes of all inspected workpieces, obtain the compression rate of each frame in each video frame sequence during the current quality inspection process.

[0069] Considering that during each quality inspection process, the shorter the modeling detection duration indicates the stronger the computing power of the cloud platform, and the compression rate of subsequent workpiece video frames can be appropriately reduced to improve the modeling quality inspection accuracy; on the contrary, the longer the modeling detection time indicates the worse the computing power of the cloud platform, and the compression rate of workpiece video frames can be appropriately increased to improve the data transmission efficiency and the efficiency between subsequent decompression processing and modeling; also considering that during each quality inspection process, the frames with larger frame blur coefficients contribute less to subsequent modeling, so the compression rate can be appropriately increased to improve the transmission efficiency; however, increasing the compression rate will cause a large amount of cloud platform computing resources to be consumed during subsequent decompression; therefore, it is necessary to comprehensively evaluate the cloud platform's cloud performance parameters during the quality inspection process of each inspected workpiece from the above two aspects, so as to prepare for determining the compression rate of the frames subsequently.

[0070] Preferably, in an embodiment of the present invention, considering that during the quality inspection process of each inspected workpiece, the larger the frame blurring coefficient under each frame number, the higher the degree of compression of the frame map that may be performed to balance the transmission pressure during this quality inspection, and the greater the computing pressure on the cloud platform of the cloud, the smaller the cloud performance parameter; the longer the modeling detection time, it indicates that the computing power pressure on the cloud platform during this quality inspection process is not sufficient to perform rapid modeling, and the cloud performance parameter is also smaller; therefore, the method for obtaining the cloud performance parameter includes:

[0071] During the quality inspection process of each inspected workpiece, the negative correlation mapping result of the sum value of the frame blurring coefficients under all frame numbers is used as the first performance parameter, and the negative correlation mapping result of the modeling detection duration is used as the second performance parameter; the first performance parameter and the second performance parameter are used as vector elements to construct a two-dimensional vector, and the two-dimensional vector is used as the cloud performance parameter of the cloud platform during the quality inspection process of the corresponding known workpiece.

[0072] As an example, specifically during the quality inspection process of each inspected workpiece, the sum value of the frame blurring coefficients under all frame numbers is inverted to obtain the first performance parameter; the larger the frame blurring coefficient, the higher the compression rate should be increased, and the greater the computing pressure on the subsequent cloud platform of the cloud, so the first performance parameter is smaller; then the modeling detection time is also inverted to obtain the second performance parameter; the longer the modeling detection time, it indicates that the computing power of the current cloud platform of the cloud is weaker, so the second performance parameter is also smaller.

[0073] Considering the cloud performance parameters corresponding to the quality inspection processes of all inspected workpieces, it can provide a certain historical reference for the cloud performance situation of the cloud platform during the quality inspection process of the current workpiece to be inspected, and then help to aggregate the degree of adjustment reference for evaluating the compression rate based on the frame blurring coefficients under each frame number during the quality inspection process of the current workpiece to be inspected, and finally determine the compression rate of each frame in each video frame sequence during the current quality inspection process in combination with the preset compression rate range.

[0074] Preferably, in an embodiment of the present invention, the method for obtaining the compression rate includes:

[0075] Please refer to Figure 2 , which shows a flowchart of a method for obtaining a compression rate provided by an embodiment of the present invention, specifically including:

[0076] Step S201, fuse the first performance parameter and the second performance parameter in the cloud performance parameter of the cloud platform during the quality inspection process of each inspected workpiece to obtain the computing redundancy parameter of the cloud platform in each quality inspection process; based on the computing redundancy parameters in the quality inspection processes of all inspected workpieces, obtain the cloud computing capacity coefficient during the quality inspection process of the current workpiece to be inspected.

[0077] Considering that the larger the cloud performance parameter is, the stronger the computing redundancy ability of the cloud platform in the corresponding quality inspection process. Based on this, first fuse the first performance parameter and the second performance parameter to obtain the corresponding computing redundancy parameter. As an example, specifically add the first performance parameter and the second performance parameter for fusion. In other examples, the implementer can also use basic mathematical operations such as multiplication or weighted fusion or related mapping means to fuse the two, which will not be elaborated here.

[0078] Also considering that the computing redundancy parameter of the cloud platform in the previous adjacent process of the current workpiece to be inspected can reflect the computing ability of the cloud platform in the short term. If the previous adjacent computing redundancy parameter is larger, it indicates that the computing redundancy parameter of the cloud platform in the current quality inspection process may also be larger. Also considering the fluctuation characteristics of the computing redundancy parameters corresponding to all quality inspection processes, which reflects the performance stability of the cloud platform from the side. When the change difference of the computing redundancy parameter in the historical quality inspection process is smaller, it indicates that the performance of the cloud platform is more stable. When the computing redundancy parameter of the cloud platform in the current quality inspection process is larger and at the same time the computing redundancy stability coefficient is also larger, it indicates that the possibility of the stable performance of the cloud platform in the current quality inspection process is also larger.

[0079] Based on this, in a preferred embodiment of the present invention, the method for obtaining the cloud computing capacity coefficient includes:

[0080] Take the computing redundancy parameter of the cloud platform in the previous adjacent quality inspection process of the current workpiece to be inspected as the current computing redundancy parameter; obtain the computing redundancy stability parameter according to the fluctuation deviation of the computing redundancy parameter in the quality inspection process of each inspected workpiece relative to the average level.

[0081] Fuse the current computing redundancy parameter and the computing redundancy stability parameter to obtain the cloud computing capacity coefficient in the quality inspection process of the current workpiece to be inspected.

[0082] As an example, the calculation formula for the cloud computing capacity coefficient is:

[0083] Wherein, R is the cloud computing capacity coefficient in the quality inspection process of the current workpiece to be inspected; j is the serial number of the quality inspection process of the inspected workpiece; n j is the serial number of the previous adjacent quality inspection process of the current workpiece to be inspected, and is also the total number of the quality inspection processes of the inspected workpiece; is the computing redundancy parameter of the cloud platform in the previous adjacent quality inspection process of the current workpiece to be inspected, and is also the current computing redundancy parameter; D j is the computing redundancy parameter of the cloud platform in the quality inspection process of the jth inspected workpiece; μ is the mean value of the computing redundancy parameters of the cloud platform in the quality inspection processes of all inspected workpieces; || is the absolute value symbol; exp() is the exponential function with the natural constant e as the base; is the computing redundancy stability parameter of the cloud platform.

[0084] In the above formula, by calculating the absolute value of the difference between each calculation redundancy parameter and the mean value, the fluctuation deviation of its relative average level is evaluated. The larger the cumulative sum of the absolute values of the differences, the more fluctuating the calculation redundancy parameter is. Therefore, its negative correlation is mapped into the exponential function to adjust the logic and normalize it, so that the smaller the calculation redundancy stable parameter is; then the calculation redundancy stable parameter is multiplied and combined with the current calculation redundancy parameter to obtain the cloud computing capacity coefficient in the quality inspection process of the current workpiece to be inspected.

[0085] In other examples, the implementer can also evaluate the fluctuation situation based on the variance or standard deviation of all calculation redundancy parameters, and then perform negative correlation normalization to obtain the calculation redundancy stable parameter; the implementer can also adopt other negative correlation normalization methods, such as taking the reciprocal, etc., which will not be elaborated here.

[0086] Step S202: Integrate the negative correlation mapping result of the cloud computing capacity coefficient in the quality inspection process of the current workpiece to be inspected with the frame blurring coefficient under each frame number to obtain the frame compression weight under each frame number in all video frame sequences of the current workpiece to be inspected.

[0087] Considering that if the cloud computing capacity coefficient in the quality inspection process of the current workpiece to be inspected is larger, the compression ratio can be appropriately increased to improve the transmission efficiency. At this time, the cloud computing capacity is sufficient for subsequent decompression and modeling; also considering that if the frame blurring coefficient of the frame in the quality inspection process of the current workpiece to be inspected is higher, it means that its contribution to subsequent modeling is also lower, and the compression ratio can also be appropriately increased; based on this, the frame compression weight under each frame number in all video frame sequences of the current workpiece to be inspected can be obtained.

[0088] As an example, the calculation formula for the frame compression weight is:

[0089] where m is the frame number in all video frame sequences of the current workpiece to be inspected; is the frame compression weight under the m-th frame number in all video frame sequences of the current workpiece to be inspected; P m is the frame blurring coefficient under the m-th frame number in all video frame sequences of the current workpiece to be inspected; R is the cloud computing capacity coefficient in the quality inspection process of the current workpiece to be inspected.

[0090] In the above formula, specifically, the reciprocal operation is performed after adding 1 to the operation calculation capacity coefficient, and then the reciprocal is multiplied and integrated with the frame blurring coefficient to obtain the frame compression weight; the frame compression weight reflects the adjustment range of the compression ratio of the frame in the future. The larger it is, the more the compression ratio should be appropriately increased, and vice versa; in other examples, other negative correlation mapping means can also be adopted, which will not be elaborated here.

[0091] It should be noted that since the value range of the frame image blur coefficient obtained in the foregoing steps is 0-1, the value range of the frame image compression weight is also 0-1, preparing for obtaining the compression ratio subsequently.

[0092] Step S203: Obtain the compression ratio of each frame image in each video frame sequence in the current quality inspection process according to the frame image compression weight and the preset compression ratio range.

[0093] In a preferred embodiment of the present invention, the method for obtaining the compression ratio of each frame image includes:

[0094] Take the left endpoint of the interval of the preset compression ratio range as the basic compression ratio, use the frame image compression weight to weight the interval length of the preset compression ratio range, and take the weighted result as the adjusted compression ratio; add the basic compression ratio and the adjusted compression ratio to obtain the compression ratio of the frame image corresponding to the frame image serial number.

[0095] As an example, the calculation formula of the compression ratio is:

[0096] where m is the frame image serial number in all video frame sequences of the current workpiece to be inspected; L m is the compression ratio of the frame image corresponding to the m-th frame image serial number in all video frame sequences of the current workpiece to be inspected; is the frame image compression weight corresponding to the m-th frame image serial number in all video frame sequences of the current workpiece to be inspected; a is the left endpoint of the interval of the preset compression ratio range and also the basic compression ratio; b is the right endpoint of the interval of the preset compression ratio range; b-a is the interval length of the preset compression ratio range; is the adjusted compression ratio of the frame image corresponding to the m-th frame image serial number in all video frame sequences of the current workpiece to be inspected.

[0097] It should be noted that the preset compression ratio range needs to be determined in combination with the accuracy requirements for the frame image during subsequent modeling. In this example, it is set to 80%-90%, and the implementer can also define it by himself.

[0098] Step S4: Compress the corresponding frame image according to the compression ratio and transmit the compression result to the cloud platform.

[0099] After obtaining the frame image under each frame image serial number, the corresponding frame image can be compressed to obtain a compressed image; then the compressed images are sorted according to the requirements of the frame images to obtain all compressed video frame sequences, where the compressed video frame sequences and the frames in the video frame sequences Figure 1One correspondence; then, based on 5G communication technology and TCP / IP protocol, all compressed video frame sequences of the current workpiece to be inspected are transmitted to the cloud server, and then decompressed, and then 3D modeling of the current workpiece to be inspected is performed, and the surface quality of the current workpiece to be inspected is evaluated based on the 3D model; it should be noted that both 5G communication transmission and 3D modeling are well-known technologies and will not be elaborated here.

[0100] The present invention also proposes a data transmission system for an industrial vision cloud service platform, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of a data transmission method for an industrial vision cloud service platform.

[0101] In summary, the present invention obtains the video frame sequences of each workpiece at different quality inspection angles during the quality inspection process, and obtains the modeling inspection duration of each inspected workpiece on the cloud platform; then, during the quality inspection process of each workpiece, analyzes and evaluates the relative change parameters of each frame diagram in each video frame sequence, further obtains the frame diagram blur coefficient under each frame diagram serial number, and constructs the cloud performance parameters of the cloud platform in combination with the modeling inspection duration; then, according to the frame diagram blur coefficient under each frame diagram serial number and the preset compression rate range, in combination with the cloud performance parameters corresponding to the quality inspection processes of all inspected workpieces, obtains the compression rate of each frame diagram in each video frame sequence during the current quality inspection process, and then compresses and transmits the corresponding frame diagram. The present invention evaluates the historical computing power resources of the cloud platform by analyzing the frame diagram blur situation and modeling quality inspection duration of the inspected workpieces during the quality inspection process, serves as a reference for evaluating the computing power resources of the current cloud platform, and then dynamically adjusts the frame diagram compression rate in combination with the frame diagram blur situation of the current workpiece to be inspected during the quality inspection process, while reducing the transmission bandwidth, taking into account the working conditions of the cloud server in real time, improving the transmission effect of edge data, and further improving the subsequent quality inspection effect.

[0102] It should be noted that: the above sequence of embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0103] Each embodiment in this specification is described in a progressive manner. The same or similar parts between each embodiment can be referred to each other, and the key points of each embodiment are the differences from other embodiments.

Claims

1. A data transmission method for an industrial visual cloud service platform, characterized in that: The method comprises: Obtain a video frame sequence of each workpiece at different quality inspection angles during the quality inspection process, wherein the workpiece includes the current workpiece to be inspected and the inspected workpiece; obtain the modeling inspection time of each inspected workpiece on the cloud platform; In each video frame sequence, different frames are matched and tracked, and the relative change parameters of each frame are obtained according to the difference of matching pixels between adjacent frames; in the quality inspection process of each workpiece, the frame fuzzy coefficient under each frame sequence number is obtained according to the relative change parameters of the frames under the same frame sequence number in different video frame sequences; In the quality inspection process of each inspected workpiece, the cloud performance parameters of the cloud platform are constructed according to the frame image fuzziness coefficients under all frame image numbers and the modeling detection duration; in the quality inspection process of the current workpiece to be inspected, the compression rate of each frame image in each video frame sequence in the current quality inspection process is obtained according to the frame image fuzziness coefficient under each frame image number and the preset compression rate range, combined with the cloud performance parameters corresponding to the quality inspection process of all inspected workpieces; The corresponding frame image is compressed according to the compression rate, and the compression result is transmitted to the cloud platform.

2. The data transmission method of an industrial visual cloud service platform according to claim 1 is characterized in that: The method for obtaining the relative change parameter includes: In each video frame sequence, a rectangular coordinate system is constructed with the center of each frame as the origin to obtain the position coordinates of each pixel in each frame; between each frame and the previous adjacent frame, the Euclidean distance between the position coordinates of each pair of matching pixel points is used as a change sub-parameter, and the average of the change sub-parameters between all matching pixel points is used as the relative change parameter between each frame and the previous adjacent frame.

3. The data transmission method of an industrial visual cloud service platform according to claim 1 is characterized in that: The method for obtaining the frame image fuzzy coefficient includes: In the quality inspection process of each workpiece, between each two video frame sequences, the absolute value of the difference between the relative change parameters of the frames with the same frame number is used as the fuzzy sub-parameter under the corresponding frame number; Under each frame image sequence number, the fuzzy sub-parameters between all two different video frame sequences in the workpiece quality inspection process are integrated to obtain the frame image fuzzy coefficient.

4. The data transmission method of an industrial visual cloud service platform according to claim 1 is characterized in that: The method for obtaining the cloud performance parameters includes: In the quality inspection process of each inspected workpiece, the negative correlation mapping result of the sum of the frame image fuzzy coefficients under all frame image sequence numbers is used as the first performance parameter, and the negative correlation mapping result of the modeling detection time is used as the second performance parameter; The first performance parameter and the second performance parameter are used as vector elements to construct a two-dimensional vector, and the two-dimensional vector is used as a cloud performance parameter of the cloud platform in the quality inspection process of the corresponding known workpiece.

5. The data transmission method of an industrial visual cloud service platform according to claim 4 is characterized in that: The method for obtaining the compression rate includes: Integrate the first performance parameter and the second performance parameter of the cloud platform's cloud performance parameters during the quality inspection process of each inspected workpiece to obtain a computing redundancy parameter of the cloud platform during each quality inspection process; obtain a cloud computing capacity coefficient during the quality inspection process of the current workpiece to be inspected based on the computing redundancy parameters during the quality inspection process of all inspected workpieces; The negative correlation mapping result of the cloud computing capacity coefficient in the quality inspection process of the current workpiece to be inspected is integrated with the frame image fuzziness coefficient under each frame image sequence number to obtain the frame image compression weight under each frame image sequence number in all video frame sequences of the current workpiece to be inspected; According to the frame image compression weight and the preset compression rate range, the compression rate of each frame image in each video frame sequence in the current quality inspection process is obtained.

6. The data transmission method of an industrial visual cloud service platform according to claim 5, characterized in that: The method for obtaining the cloud computing capacity coefficient includes: The computing redundancy parameters of the cloud platform in the previous adjacent quality inspection process of the current workpiece to be inspected are used as the current computing redundancy parameters; and the computing redundancy stability parameters are obtained according to the fluctuation deviation of the computing redundancy parameters in the quality inspection process of each inspected workpiece relative to the average level; The current computing redundancy parameter and the computing redundancy stability parameter are integrated to obtain the cloud computing capacity coefficient in the quality inspection process of the current workpiece to be inspected.

7. The data transmission method of an industrial visual cloud service platform according to claim 5 is characterized in that: The method for obtaining the compression rate of each frame image in each video frame sequence in the current quality inspection process according to the frame image compression weight and the preset compression rate range includes: The left end point of the interval of the preset compression rate range is taken as the basic compression rate, the interval length of the preset compression rate range is weighted using the frame image compression weight, and the weighted result is taken as the adjusted compression rate; the basic compression rate is added to the adjusted compression rate to obtain the compression rate of the frame image under the corresponding frame image sequence number.

8. The data transmission method of an industrial visual cloud service platform according to claim 1, characterized in that: The methods for matching and tracking different frames include: The workpiece area contour in each frame image is obtained, and all corner points on the workpiece area contour are obtained; in each video frame sequence, any frame image is taken as the target frame image, and any pixel point in the target frame image is taken as the target corner point, and the nearest neighbor corner point of the target corner point in each non-target frame image is found based on the KD tree, and the target corner point and its nearest neighbor corner point in each non-target frame image are used as a series of matching points.

9. The data transmission method of an industrial visual cloud service platform according to claim 8, characterized in that: The method for obtaining the corner points includes: Based on the Hessian corner detection algorithm, all corner points are obtained.

10. A data transmission system for an industrial visual cloud service platform, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of a data transmission method for an industrial visual cloud service platform as described in any one of claims 1 to 9.

Images