Data processing method and system applied to product quality assessment model

Abstract

An embodiment of the present invention provides a data processing method and system for a product quality assessment model. The initial product quality assessment model is optimized based on a first reference product acquisition image sequence containing labeled product defect information and a second reference product acquisition image sequence not carrying labeled product defect information. The product quality assessment of each reference product acquisition image in a second reference product acquisition image sequence for another product purpose can be learned under the condition of a first reference product acquisition image sequence for one product purpose. The model is iteratively optimized based on a global error value determined based on a first training error value between reference product acquisition images with the same product defect (including the same product defect with the same product purpose and different purposes) and a second training error value between reference product acquisition images with different product defects, thereby improving the model performance of the product quality assessment model.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a data processing method and system applied to a product quality assessment model. Background Art

[0002] Artificial intelligence can improve production efficiency and quality and reduce labor costs and time waste through intelligent control and automation technology. With the rapid development of the artificial intelligence era, this technology has been sought after by many businesses. Product quality assessment is the most important link in enterprise production and an important decisive factor affecting enterprise development. Product quality assessment is of great significance to business operators. In the production process, how to improve the reliability of product quality assessment is a technical problem that needs to be solved urgently in this field. Summary of the invention

[0003] In view of this, the purpose of the embodiment of the present invention is to provide a data processing method and system for a product quality assessment model, which optimizes the initial product quality assessment model based on a first reference product acquisition image sequence containing annotated product defect information and a second reference product acquisition image sequence that does not carry annotated product defect information, and can learn the product quality assessment of each reference product acquisition image in a second reference product acquisition image sequence for another product use under the condition of a first reference product acquisition image sequence for one product use. And based on the first training error value between reference product acquisition images of the same product defect (including the same product defect for the same product use and different uses), the difference between reference product acquisition images of the same product defect can be continuously reduced. Based on the second training error value between reference product acquisition images of different product defects, the difference between reference product acquisition images of different product defects can be continuously increased, thereby reducing the mutual influence between reference product acquisition images of different product defects and improving the product quality assessment accuracy of product acquisition images across product uses. Based on the estimated error value between the prior product defect distribution and the estimated product defect information of each first reference product acquisition image in the first reference product acquisition image sequence, iterative model optimization is performed based on the estimated error value, the second training error value and the global error value determined by the first training error value, thereby improving the model performance of the product quality assessment model.

[0004] According to one aspect of an embodiment of the present invention, a data processing method and system for a product quality assessment model are provided, the method comprising:

[0005] Acquire product quality learning data, the product quality learning data comprising a first reference product acquisition image sequence and a second reference product acquisition image sequence, wherein each first reference product acquisition image in the first reference product acquisition image sequence carries annotated product defect information, the annotated product defect information characterizes a priori product defect distribution of the first reference product acquisition image, and the first reference product acquisition image sequence and the second reference product acquisition image sequence are reference product acquisition image sequences for different purposes;

[0006] Determining initial product defect information corresponding to each second reference product captured image in the second reference product captured image sequence;

[0007] Loading the product quality learning data into an initial product quality assessment model to obtain estimated product defect information of each reference product captured image, and determining an estimated error value based on a priori product defect distribution and the estimated product defect information of each first reference product captured image;

[0008] Determine a first training error value based on reference product images with the same purpose and the same product defect, and reference product images with the same product defect for different purposes, and determine a second training error value based on reference product images with the same purpose and different product defects, and reference product images with different purposes and different product defects;

[0009] Determining a global error value based on the estimated error value, the first training error value, and the second training error value;

[0010] Based on the global error value, the model parameters of the initial product quality assessment model are updated until the global error value converges, and the initial product quality assessment model corresponding to the convergence is output as a product quality assessment model for cross-product purposes;

[0011] Among them, for each round of model parameter updating, if the global error value does not converge, the product defect information corresponding to each second reference product captured image and the estimated product defect information corresponding to each second reference product captured image is adjusted to obtain new product defect information, and used for the next round of model parameter updating.

[0012] In an alternative implementation, the determining the initial product defect information corresponding to each second reference product captured image in the second reference product captured image sequence includes:

[0013] Based on each of the first reference product captured images and the corresponding annotated product defect information, determining the first image feature cluster center corresponding to each product defect;

[0014] For any second reference product acquisition image, determining a feature deviation value between the any second reference product acquisition image and each of the first image feature cluster centers;

[0015] Based on the a priori product defect distribution corresponding to the first image feature cluster center with the smallest feature deviation value of the any one of the second reference product captured images, initial product defect information of the any one of the second reference product captured images is determined.

[0016] In an alternative implementation, determining the first training error value based on the captured images of reference products with the same purpose and the same product defect, and the captured images of reference products with the same purpose and the same product defect, includes:

[0017] Determining a first cluster internal distance between each first reference product captured image of each product defect based on each first reference product captured image and its corresponding annotated product defect information;

[0018] Determining a second cluster internal distance between each second reference product captured image of each product defect based on each second reference product captured image and its corresponding product defect information;

[0019] Determine a third cluster internal distance between each first reference product acquisition image and each second reference product acquisition image of each product defect based on each of the first reference product acquisition images and their corresponding annotated product defect information, and each of the second reference product acquisition images and their corresponding product defect information;

[0020] A first training error value is determined based on the first cluster internal distance, the second cluster internal distance, and the third cluster internal distance.

[0021] In an alternative implementation, determining the second training error value based on the reference product images with different product defects for the same purpose and the reference product images with different product defects for different purposes includes:

[0022] Determining a first cluster external distance between first reference product captured images of different product defects based on the first reference product captured images and their corresponding annotated product defect information;

[0023] Based on each of the second reference product captured images and their corresponding product defect information, determine the second cluster external distance between each of the second reference product captured images of different product defects; based on each of the first reference product captured images and their corresponding annotated product defect information, and each of the second reference product captured images and their corresponding product defect information, determine the third cluster external distance between each of the first reference product captured images of different product defects and each of the second reference product captured images; based on the first cluster external distance, the second cluster external distance and the third cluster external distance, determine the second training error value.

[0024] In an alternative implementation, adjusting the product defect information corresponding to each second reference product captured image to obtain new product defect information based on each second reference product captured image and the estimated product defect information corresponding to each second reference product captured image includes:

[0025] Based on each of the second reference product captured images and the corresponding product defect information, determining the second image feature cluster center corresponding to each product defect;

[0026] For any second reference product captured image, determining a feature deviation value between the any second reference product captured image and each of the second image feature cluster centers;

[0027] Based on the a priori product defect distribution corresponding to the second image feature cluster center closest to the any one of the second reference product acquisition images, the product defect information corresponding to the any one of the second reference product acquisition images is adjusted.

[0028] In an alternative implementation, after each updating of the product defect information corresponding to each of the second reference product captured images to obtain new product defect information, the method further includes:

[0029] Determine the adjustment quantity of the product defect information corresponding to each of the second reference product captured images; if the adjustment quantity is equal to the set quantity, use the product defect information after the adjustment as the final product defect information of each of the second reference product captured images.

[0030] In an alternative implementation, determining a global error value based on the estimated error value, the first training error value, and the second training error value includes:

[0031] Obtaining a first importance coefficient corresponding to the estimated error value and a second importance coefficient corresponding to a cluster distance error value, wherein the cluster distance error value includes the first training error value and the second training error value;

[0032] A global error value is determined based on the estimated error value and its corresponding first importance coefficient, and the cluster distance error value and its corresponding second importance coefficient.

[0033] In an alternative embodiment, the method further comprises:

[0034] Acquire a target product captured image, where the target product captured image is a product captured image belonging to a first product purpose, or is a product captured image belonging to a second product purpose; load the target product captured image into a cross-product purpose product quality assessment model, and determine a product quality assessment result of the target product captured image based on an output of the cross-product purpose product quality assessment model.

[0035] According to another aspect of an embodiment of the present invention, a data processing method and system for a product quality assessment model are provided, the system comprising:

[0036] According to another aspect of an embodiment of the present invention, there is provided a data processing system applied to a product quality assessment model, comprising: a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; the memory is used to store a computer program; the processor is used to implement any one of the above steps of the data processing method applied to a product quality assessment model when executing the computer program.

[0037] According to another aspect of an embodiment of the present invention, a readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the above-described data processing methods applied to a product quality assessment model can be executed.

[0038] In order to make the above-mentioned objects, features and advantages of the embodiments of the present invention more obvious and easy to understand, the embodiments will be described in detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A schematic diagram of components of a data processing system applied to a product quality assessment model provided by an embodiment of the present invention is shown;

[0041] Figure 2A flow chart of a data processing method and system for a product quality assessment model provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0042] In order to enable students in the technical field to better understand the scheme of the present invention, the technical scheme in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all the embodiments. According to the embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] The terms "first", "second", "third", etc. (if any) in the specification 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 data 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, for example. 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 that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0044] Figure 1An exemplary component schematic diagram of a data processing 100 applied to a product quality assessment model is shown. The data processing 100 applied to the product quality assessment model may include one or more processors 104, such as one or more central processing units (CPUs), each of which may implement one or more hardware threads. The data processing 100 applied to the product quality assessment model may also include any storage medium 106, which is used to store any kind of information such as code, settings, data, etc. Non-limiting, for example, the storage medium 106 may include any one or more combinations of the following: any type of RAM, any type of ROM, flash memory device, hard disk, optical disk, etc. More generally, any storage medium may use any technology to store information. Further, any storage medium may provide volatile or non-volatile retention of information. Further, any storage medium may represent a fixed or removable component of the data processing 100 applied to the product quality assessment model. In one case, when the processor 104 executes an associated instruction stored in any storage medium or a combination of storage media, the data processing 100 applied to the product quality assessment model may perform any operation of the associated instruction. The data processing 100 applied to the product quality assessment model also includes one or more drive units 108 for interacting with any storage medium, such as a hard disk drive unit, an optical disk drive unit, etc.

[0045] The data processing 100 applied to the product quality assessment model also includes an input / output 110 (I / O) for receiving various inputs (via input unit 112) and for providing various outputs (via output unit 114). A specific output mechanism may include a presentation device 116 and an associated graphical user interface (GUI) 118. The data processing 100 applied to the product quality assessment model may also include one or more network interfaces 120, which are used to exchange data with other devices via one or more communication units 122. One or more communication buses 124 couple the components described above together.

[0046] The communication unit 122 may be implemented in any manner, for example, through a local area network, a wide area network (e.g., the Internet), a point-to-point connection, etc., or any combination thereof. The communication unit 122 may include any combination of hardwired links, wireless links, routers, gateway functions, data processing 100, etc., governed by any protocol or combination of protocols.

[0047] Figure 2 The flowchart of the data processing method and system for product quality assessment model provided by the embodiment of the present invention is shown. The data processing method and system for product quality assessment model can be Figure 1The data processing 100 applied to the product quality assessment model shown in is performed, and the detailed steps of the data processing method and system applied to the product quality assessment model are introduced as follows.

[0048] Step S110, obtaining product quality learning data, wherein the product quality learning data includes a first reference product acquisition image sequence and a second reference product acquisition image sequence, wherein each first reference product acquisition image in the first reference product acquisition image sequence carries annotated product defect information, and the annotated product defect information represents a priori product defect distribution of the first reference product acquisition image, and the first reference product acquisition image sequence and the second reference product acquisition image sequence are reference product acquisition image sequences for different purposes;

[0049] Step S120, determining initial product defect information corresponding to each second reference product captured image in the second reference product captured image sequence;

[0050] Step S130, loading the product quality learning data into an initial product quality assessment model, obtaining estimated product defect information of each reference product captured image, and determining an estimated error value based on a priori product defect distribution and the estimated product defect information of each first reference product captured image;

[0051] Step S140, determining a first training error value based on reference product images with the same purpose and the same product defect, and reference product images with the same product defect for different purposes, and determining a second training error value based on reference product images with the same purpose and different product defects, and reference product images with different purposes and different product defects;

[0052] Step S150, determining a global error value based on the estimated error value, the first training error value, and the second training error value;

[0053] Step S160, updating the model parameters of the initial product quality assessment model based on the global error value until the global error value converges, and outputting the initial product quality assessment model corresponding to the convergence as a cross-product use product quality assessment model;

[0054] Step S170, wherein, for each round of model parameter update, if the global error value does not converge, then based on each of the second reference product captured images and the estimated product defect information corresponding to each of the second reference product captured images, the product defect information corresponding to each of the second reference product captured images is adjusted to obtain new product defect information, and used for the next round of model parameter update.

[0055] According to the above steps, this embodiment optimizes the initial product quality assessment model based on the first reference product acquisition image sequence containing the labeled product defect information and the second reference product acquisition image sequence without the labeled product defect information, and can learn the product quality assessment of each reference product acquisition image in the second reference product acquisition image sequence for another product use under the condition of the first reference product acquisition image sequence for one product use. And according to the first training error value between the reference product acquisition images with the same product defect (including the same product defect with the same product use and different uses), the difference between the reference product acquisition images with the same product defect can be continuously reduced. According to the second training error value between the reference product acquisition images with different product defects, the difference between the reference product acquisition images with different product defects can be continuously increased, thereby reducing the mutual influence between the reference product acquisition images with different product defects and improving the product quality assessment accuracy of the product acquisition images across product uses. According to the estimated error value between the prior product defect distribution and the estimated product defect information of each first reference product acquisition image in the first reference product acquisition image sequence, iterative model optimization is performed according to the estimated error value, the second training error value and the global error value determined by the first training error value, thereby improving the model performance of the product quality assessment model

[0056] In an alternative implementation, the determining the initial product defect information corresponding to each second reference product captured image in the second reference product captured image sequence includes:

[0057] Based on each of the first reference product captured images and the corresponding annotated product defect information, determining the first image feature cluster center corresponding to each product defect;

[0058] For any second reference product acquisition image, determining a feature deviation value between the any second reference product acquisition image and each of the first image feature cluster centers;

[0059] Based on the a priori product defect distribution corresponding to the first image feature cluster center with the smallest feature deviation value of the any one of the second reference product captured images, initial product defect information of the any one of the second reference product captured images is determined.

[0060] In an alternative implementation, determining the first training error value based on the captured images of reference products with the same purpose and the same product defect, and the captured images of reference products with the same purpose and the same product defect, includes:

[0061] Determining a first cluster internal distance between each first reference product captured image of each product defect based on each first reference product captured image and its corresponding annotated product defect information;

[0062] Determining a second cluster internal distance between each second reference product captured image of each product defect based on each second reference product captured image and its corresponding product defect information;

[0063] Determine a third cluster internal distance between each first reference product acquisition image and each second reference product acquisition image of each product defect based on each of the first reference product acquisition images and their corresponding annotated product defect information, and each of the second reference product acquisition images and their corresponding product defect information;

[0064] A first training error value is determined based on the first cluster internal distance, the second cluster internal distance, and the third cluster internal distance.

[0065] In an alternative implementation, determining the second training error value based on the reference product images with different product defects for the same purpose and the reference product images with different product defects for different purposes includes:

[0066] Determining a first cluster external distance between first reference product captured images of different product defects based on the first reference product captured images and their corresponding annotated product defect information;

[0067] Based on each of the second reference product captured images and their corresponding product defect information, determine the second cluster external distance between each of the second reference product captured images of different product defects; based on each of the first reference product captured images and their corresponding annotated product defect information, and each of the second reference product captured images and their corresponding product defect information, determine the third cluster external distance between each of the first reference product captured images of different product defects and each of the second reference product captured images; based on the first cluster external distance, the second cluster external distance and the third cluster external distance, determine the second training error value.

[0068] In an alternative implementation, adjusting the product defect information corresponding to each second reference product captured image to obtain new product defect information based on each second reference product captured image and the estimated product defect information corresponding to each second reference product captured image includes:

[0069] Based on each of the second reference product captured images and the corresponding product defect information, determining the second image feature cluster center corresponding to each product defect;

[0070] For any second reference product captured image, determining a feature deviation value between the any second reference product captured image and each of the second image feature cluster centers;

[0071] Based on the a priori product defect distribution corresponding to the second image feature cluster center closest to the any one of the second reference product acquisition images, the product defect information corresponding to the any one of the second reference product acquisition images is adjusted.

[0072] In an alternative implementation, after each updating of the product defect information corresponding to each of the second reference product captured images to obtain new product defect information, the method further includes:

[0073] Determine the adjustment quantity of the product defect information corresponding to each of the second reference product captured images; if the adjustment quantity is equal to the set quantity, use the product defect information after the adjustment as the final product defect information of each of the second reference product captured images.

[0074] In an alternative implementation, determining a global error value based on the estimated error value, the first training error value, and the second training error value includes:

[0075] Obtaining a first importance coefficient corresponding to the estimated error value and a second importance coefficient corresponding to a cluster distance error value, wherein the cluster distance error value includes the first training error value and the second training error value;

[0076] A global error value is determined based on the estimated error value and its corresponding first importance coefficient, and the cluster distance error value and its corresponding second importance coefficient.

[0077] In an alternative embodiment, the method further comprises:

[0078] Acquire a target product captured image, where the target product captured image is a product captured image belonging to a first product purpose, or is a product captured image belonging to a second product purpose; load the target product captured image into a cross-product purpose product quality assessment model, and determine a product quality assessment result of the target product captured image based on an output of the cross-product purpose product quality assessment model.

[0079] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0080] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the present invention can be implemented in other specific forms without departing from the spirit or essential features of the present invention. Therefore, the embodiments should be considered exemplary and non-restrictive in all respects, and the scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the present invention.

Claims

1. A data processing method applied to a product quality assessment model, characterized in that: The method comprises: Acquire product quality learning data, the product quality learning data comprising a first reference product acquisition image sequence and a second reference product acquisition image sequence, wherein each first reference product acquisition image in the first reference product acquisition image sequence carries annotated product defect information, the annotated product defect information represents a priori product defect distribution of the first reference product acquisition image, and the first reference product acquisition image sequence and the second reference product acquisition image sequence are reference product acquisition image sequences for different purposes; Determining initial product defect information corresponding to each second reference product captured image in the second reference product captured image sequence; Loading the product quality learning data into an initial product quality assessment model to obtain estimated product defect information of each reference product captured image, and determining an estimated error value based on a priori product defect distribution and the estimated product defect information of each first reference product captured image; Determine a first training error value based on reference product images with the same purpose and the same product defect, and reference product images with the same product defect for different purposes, and determine a second training error value based on reference product images with the same purpose and different product defects, and reference product images with different purposes and different product defects; Determining a global error value based on the estimated error value, the first training error value, and the second training error value; Based on the global error value, the model parameters of the initial product quality assessment model are updated until the global error value converges, and the initial product quality assessment model corresponding to the convergence is output as a product quality assessment model for cross-product purposes; Wherein, for each round of model parameter update, if the global error value does not converge, then based on each second reference product captured image and the estimated product defect information corresponding to each second reference product captured image, the product defect information corresponding to each second reference product captured image is adjusted to obtain new product defect information, and used for the next round of model parameter update; The determining of the first training error value based on the captured images of reference products with the same purpose and the same product defect, and the captured images of reference products with the same purpose and the same product defect, comprises: Determining a first cluster internal distance between each first reference product captured image of each product defect based on each first reference product captured image and its corresponding annotated product defect information; Determining a second cluster internal distance between each second reference product captured image of each product defect based on each second reference product captured image and its corresponding product defect information; Determine a third cluster internal distance between each first reference product acquisition image and each second reference product acquisition image of each product defect based on each of the first reference product acquisition images and their corresponding annotated product defect information, and each of the second reference product acquisition images and their corresponding product defect information; Determine a first training error value based on the first cluster internal distance, the second cluster internal distance, and the third cluster internal distance; The determining of the second training error value based on the reference product images with different product defects for the same purpose and the reference product images with different product defects for different purposes includes: Determining a first cluster external distance between first reference product captured images of different product defects based on the first reference product captured images and their corresponding annotated product defect information; Based on each of the second reference product captured images and their corresponding product defect information, determine the second cluster external distance between each of the second reference product captured images of different product defects; Based on each of the first reference product captured images and their corresponding annotated product defect information, and each of the second reference product captured images and their corresponding product defect information, determine the third cluster external distance between each of the first reference product captured images and each of the second reference product captured images of different product defects; Based on the first cluster external distance, the second cluster external distance and the third cluster external distance, determine the second training error value.

2. The method according to claim 1, characterized in that: The determining of initial product defect information corresponding to each second reference product captured image in the second reference product captured image sequence includes: Based on each of the first reference product captured images and the corresponding annotated product defect information, determining the first image feature cluster center corresponding to each product defect; For any second reference product acquisition image, determining a feature deviation value between the any second reference product acquisition image and each of the first image feature cluster centers; Based on the a priori product defect distribution corresponding to the first image feature cluster center with the smallest feature deviation value of the any one of the second reference product captured images, initial product defect information of the any one of the second reference product captured images is determined.

3. The method according to claim 1, characterized in that: The adjusting the product defect information corresponding to each of the second reference product captured images to obtain new product defect information based on each of the second reference product captured images and the estimated product defect information corresponding to each of the second reference product captured images includes: Based on each of the second reference product captured images and the corresponding product defect information, determining the second image feature cluster center corresponding to each product defect; For any second reference product captured image, determining a feature deviation value between the any second reference product captured image and each of the second image feature cluster centers; Based on the a priori product defect distribution corresponding to the second image feature cluster center closest to the any one of the second reference product acquisition images, the product defect information corresponding to the any one of the second reference product acquisition images is adjusted.

4. The method according to claim 3, characterized in that: After each updating of the product defect information corresponding to each of the second reference product captured images to obtain new product defect information, the method further includes: Determine the adjustment quantity of the product defect information corresponding to each of the second reference product captured images; if the adjustment quantity is equal to the set quantity, use the product defect information after the adjustment as the final product defect information of each of the second reference product captured images.

5. The method according to claim 1, characterized in that: The determining a global error value based on the estimated error value, the first training error value, and the second training error value comprises: Obtaining a first importance coefficient corresponding to the estimated error value and a second importance coefficient corresponding to a cluster distance error value, wherein the cluster distance error value includes the first training error value and the second training error value; A global error value is determined based on the estimated error value and its corresponding first importance coefficient, and the cluster distance error value and its corresponding second importance coefficient.

6. The method according to claim 1, characterized in that: The method further comprises: Acquire a target product captured image, where the target product captured image is a product captured image belonging to a first product purpose, or a product captured image belonging to a second product purpose; The target product captured image is loaded into a cross-product usage product quality assessment model, and a product quality assessment result of the target product captured image is determined based on an output of the cross-product usage product quality assessment model.

7. A data processing system applied to a product quality assessment model, characterized in that: include: A processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus; The memory is used to store computer programs; the processor is used to implement the steps of the data processing method applied to the product quality assessment model described in any one of claims 1 to 6 when executing the computer program.

8. A readable storage medium, characterized in that: The readable storage medium stores a computer program, which, when executed by a processor, can execute the steps of the data processing method applied to a product quality assessment model as described in any one of claims 1 to 6.

Images