WO2024032491A1

WO2024032491A1 - Detection system and image processing method

Info

Publication number: WO2024032491A1
Application number: PCT/CN2023/111198
Authority: WO
Inventors: 胥飞龙; 伍强; 王伟
Original assignee: 江苏时代新能源科技有限公司
Priority date: 2022-08-08
Filing date: 2023-08-04
Publication date: 2024-02-15
Also published as: CN115063286A; CN115063286B

Abstract

The present application relates to the technical field of computers, and relates to a detection system and an image processing method. The detection system comprises: a front-end computer and N back-end computers. The front-end computer is used for distributing image scanning data of a sample to be detected; each of the N back-end computers is connected to the front-end computer, and each back-end computer is used for performing image processing on the image scanning data distributed by the front-end computer, N being an integer greater than or equal to 2. The N back-end computers are used to perform image processing in parallel on the image scanning data of different samples to be detected, such that the problem of image scanning data accumulation caused by long image processing time of a single back-end computer can be solved, and the overall detection efficiency is greatly improved.

Description

A detection system and image processing method

Cross-references to related applications

This application claims priority to Chinese patent application 202210944513.2 titled "A detection system and image processing method" submitted on August 8, 2022. The entire content of this application is incorporated herein by reference.

Technical field

The present application relates to the field of computer technology, and specifically to a detection system and an image processing method.

Background technique

This section provides merely background information related to the present disclosure and is not necessarily prior art.

Computed Tomography (Computed Tomography) technology is a method of reconstructing specific aspects of an object based on the projection data of certain physical quantities (such as X-ray light intensity, electron beam intensity, etc.) obtained around the object without destroying the structure of the object. The two-dimensional image or three-dimensional image technology is widely used in the fields of medical projection and industrial non-destructive testing.

In the industrial CT testing of power batteries, it is usually divided into manual loading (such as placing the battery sample to be tested on the workbench of the CT equipment), scanning detection (CT equipment scans the image of the battery sample to be tested on the workbench), There are three steps of image processing (the back-end machine processes the scanned image data). The principle is shown in Figure 1. Among them, manual loading and scanning inspection need to be completed serially, while manual loading + scanning inspection and image processing can be completed in parallel. With the continuous increase in power battery production capacity, the demand for industrial CT inspection is increasing day by day. Improving the overall inspection rhythm of CT and expanding inspection capacity have become the trends in industrial CT applications.

Contents of the invention

In view of the above problems, the purpose of this application is to provide a detection system and an image processing method to improve the existing detection system's problem of untimely processing of image scan data, resulting in accumulation of image scan data and affecting detection efficiency.

In the first aspect, embodiments of the present application provide a detection system, including: a front-end machine and N back-end machines; the front-end machine is used to distribute the image scan data of the sample to be detected; the N back-end machines are Each back-end machine is connected to the front-end machine, and each back-end machine is used to perform image processing on the image scan data allocated by the front-end machine, and N is an integer greater than or equal to 2.

In the embodiment of the present application, by increasing the number of back-end machines in the image processing part, N back-end machines can perform image processing on the image scanning data of different samples to be detected in parallel, thereby solving the problem of a single back-end machine. The long image processing time leads to the problem of image scanning data accumulation, which greatly improves the overall efficiency of detection.

In conjunction with a possible implementation of the embodiment of the first aspect, the value of N is not less than the quotient of the time required to process one image scan data and the time required to image scan one of the samples to be detected to obtain image scan data.

In the embodiment of the present application, when the value of N is not less than the quotient of the time required to process an image scan data and the time required to image scan a sample to be detected to obtain image scan data, the data accumulation problem can be alleviated to the greatest extent, especially When the number of samples to be tested is large enough, the effect becomes more obvious.

In conjunction with a possible implementation manner of the embodiment of the first aspect, a monitoring module is deployed in each of the back-end machines, and the monitoring module is used to monitor the processing progress of the image scan data processed by the back-end machine; the front-end machine , specifically used to allocate the image scanning data of the sample to be detected to the target back-end machine among the N back-end machines according to the feedback data of the monitoring module deployed on each of the back-end machines.

In the embodiment of this application, by deploying a monitoring module in the back-end machine, the front-end machine can distribute the image scanning data of the sample to be detected to N back-end machines based on the feedback data of the monitoring module deployed on each back-end machine. The target back-end machine in the system can achieve reasonable allocation of resources and effectively prevent data duplication and leakage processing when multiple back-end machines are used.

In conjunction with a possible implementation of the embodiment of the first aspect, the N back-end machines include a main back-end machine; the main back-end machine is used to scan the image according to the image data processed by all back-end machines and scan the image. The defect detection results of manual defect calibration are used to generate detection reports and save them.

In the embodiment of this application, the main back-end machine is used to summarize the image scanning data processed by all back-end machines and the defect detection results of manual defect calibration for the image scanning data, thereby generating a detection report and saving it for By reading the test report, you can quickly and intuitively learn the test results.

In conjunction with a possible implementation of the embodiment of the first aspect, the detection report includes: any combination of product type, detection quantity, defect quantity, defect type, defect rate, and detection excellence rate.

In the embodiment of this application, since the inspection report includes core information such as product type, inspection quantity, defect quantity, defect type, defect rate, inspection excellence rate, etc., the core information can be clearly understood by reading the inspection report.

With reference to a possible implementation manner of the embodiment of the first aspect, the detection system further includes: a CT device connected to the front-end machine, the CT device is used to perform a test on the sample to be detected placed on its own workbench. Image scanning is used to obtain image scanning data of the sample to be detected.

In a second aspect, embodiments of the present application also provide an image processing method, including: a front-end machine acquires image scanning data of a sample to be detected, and distributes the image scanning data of the sample to be detected to N backends connected thereto The target backend machine in the machine, N is an integer greater than or equal to 2; the target backend machine performs image processing on the allocated image scan data.

In conjunction with a possible implementation of the embodiment of the second aspect, a monitoring module is deployed in each of the back-end machines, and the monitoring module is used to monitor the processing progress of the image scan data processed by the back-end machine; Distributing the image scan data of the detection sample to the target back-end machine among the N back-end machines connected thereto includes: determining the current idle state based on the feedback data of the monitoring module deployed on each of the back-end machines. The target back-end machine; distributes the image scan data of the sample to be detected to the target back-end machine.

In conjunction with a possible implementation of the embodiment of the second aspect, the target back-end machine performs image processing on the allocated image scan data, including: the target back-end machine uses three-dimensional visualization software to import the image scan data for three-view display , and respond to the user's manual calibration operation on the existing defect types in each view direction to obtain the corresponding defect detection results; the target back-end machine performs slicing processing in each view direction to convert the three-dimensional view into a two-dimensional view. View saved.

In the embodiment of the present application, the image scan data is displayed in three views, so that the operator can browse from different view directions and calibrate the places where defects exist. In response to the user's manual calibration of the existing defect types in each view direction, the corresponding defect detection results can be obtained. In addition, slicing processing can be performed to convert the three-dimensional view into a two-dimensional view and save it for later viewing.

In conjunction with a possible implementation of the embodiment of the second aspect, the N back-end machines include a main back-end machine, and the method further includes: the main back-end machine scans the image data processed by all back-end machines and Based on the defect detection results of manual defect calibration based on the image scan data, a detection report is generated and saved.

Other features and advantages of the present application will be set forth in the subsequent description. The objectives and other advantages of the present application may be realized and attained by the structure particularly pointed out in the written description and accompanying drawings.

Description of drawings

Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are for the purpose of illustrating preferred embodiments only and are not to be construed as limiting the application. Also throughout the drawings, the same reference characters are used to designate the same components. In the attached picture:

Figure 1 is a schematic diagram of the principle of CT detection in the prior art.

Figure 2 shows a schematic structural diagram of a detection system provided by an embodiment of the present application.

Figure 3 shows a schematic diagram of the detection principle of a detection system provided by an embodiment of the present application.

FIG. 4 shows a schematic flowchart of an image processing method provided by an embodiment of the present application.

Detailed ways

The embodiments of the technical solution of the present application will be described in detail below with reference to the accompanying drawings. The following examples are only used to illustrate the technical solution of the present application more clearly, and are therefore only used as examples and cannot be used to limit the protection scope of the present application.

Unless otherwise defined, all technical and scientific terms used herein have the same meanings as commonly understood by those skilled in the technical field belonging to this application; the terms used herein are for the purpose of describing specific embodiments only and are not intended to be used in Limitation of this application; the terms "including" and "having" and any variations thereof in the description and claims of this application and the above description of the drawings. Shape, intended to cover non-exclusive inclusion.

In the description of the embodiments of this application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying the relative importance or implicitly indicating the quantity or specificity of the indicated technical features. Sequence or priority relationship. In the description of the embodiments of this application, "plurality" means two or more, unless otherwise explicitly and specifically limited.

Reference herein to "an embodiment" means that a particular feature, structure or characteristic described in connection with the embodiment can be included in at least one embodiment of the present application. The appearances of this phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those skilled in the art understand, both explicitly and implicitly, that the embodiments described herein may be combined with other embodiments.

In the description of the embodiments of this application, the term "and/or" is only an association relationship describing associated objects, indicating that there can be three relationships, such as A and/or B, which can mean: A exists, and A and A exist simultaneously. B, there are three situations B. In addition, the character "/" in this article generally indicates that the related objects are an "or" relationship.

In the description of the embodiments of this application, the term "multiple" refers to more than two (including two). Similarly, "multiple groups" refers to two or more groups (including two groups), and "multiple pieces" refers to It is more than two pieces (including two pieces).

In the description of the embodiments of this application, the technical terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "back", "left", "right" and "vertical" The orientation or positional relationships indicated by "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. are based on those shown in the accompanying drawings. The orientation or positional relationship is only for the convenience of describing the embodiments of the present application and simplifying the description. It does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the implementation of the present application. Example limitations.

In the description of the embodiments of this application, unless otherwise clearly stated and limited, technical terms such as "installation", "connection", "connection" and "fixing" should be understood in a broad sense. For example, it can be a fixed connection or a removable connection. It can be disassembled and connected, or integrated; it can also be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be an internal connection between two elements or an interaction between two elements. For those of ordinary skill in the art, The specific meanings of the above terms in the embodiments of this application can be understood according to specific circumstances.

At present, judging from the development of the market situation, the application of power batteries is becoming more and more extensive. Power batteries are not only used in energy storage power systems such as hydropower, thermal power, wind power and solar power stations, but are also widely used in electric vehicles such as electric bicycles, electric motorcycles and electric cars, as well as in many fields such as military equipment and aerospace. . As the application fields of power batteries continue to expand, their market demand is also constantly expanding.

Among related technologies, the bottleneck in the industrial CT (Computed Tomography) inspection of power batteries lies in the long image processing time, which is about 9 minutes, while manual loading (about 1 minute) + scanning inspection (about 2 minutes ) is 3 minutes. In this case, when a single back-end machine is used to process the image scanning data, the image scanning data will accumulate and affect the detection rhythm. Therefore, this application provides a detection system that can improve the CT detection efficiency of power batteries. It should be noted that the defects in the above solutions are all the results obtained by the inventor after practice and careful study. Therefore, the discovery process of the above problems and the solutions to the above problems proposed by the embodiments of the present invention below are All solutions should be the contribution made by the inventor to the present invention in the process of the invention.

In order to improve the efficiency of CT detection of power batteries, this application adds multiple back-end machines to the image processing part based on the existing detection system architecture to process image scan data in parallel, which can effectively alleviate the problem of data accumulation.

The detection system provided by the embodiment of the present application will be described below with reference to Figure 2. The detection system provided by the embodiment of this application includes a front-end machine and N back-end machines. Each of the N back-end machines is independent of each other, and N is an integer greater than or equal to 2. For example, N can be an integer greater than or equal to 2, such as 2, 3, 4, 5, 10, etc.

The front-end machine is used to distribute the image scanning data of the samples to be tested. For example, after the front-end machine obtains the image scanning data of different samples to be tested, it sends it to N back-end machines in a certain order for image processing. Each back-end machine among the N back-end machines is connected to the front-end machine. For example, each back-end machine is connected to the front-end machine through network equipment (such as routers, switches, etc.). Each back-end machine is used to perform image processing on the image scan data assigned by the front-end machine. The principle is shown in Figure 3. By increasing the number of back-end machines in the image processing part, N back-end machines can process the image scanning data of different samples to be detected in parallel, thereby solving the problem of image processing time of a single back-end machine. It is long, which leads to the problem of accumulation of image scanning data and greatly increases the overall tempo of CT detection.

In one embodiment, the front-end machine can obtain image scanning data of the sample to be detected from the CT device in real time. Optionally, the detection system also includes: a CT device connected to the front-end machine. The CT device is used to scan the image of the sample to be detected placed on its own workbench, and to transmit the scanned image scan data of the sample to be detected. to the front-end machine. For example, in the manual loading stage, the sample 1 to be tested is placed on the workbench of the CT equipment. In the scanning detection stage, the CT equipment scans the image of the first sample to be tested placed on its own workbench, and the obtained The image scan data of the first sample to be tested is transmitted to the front-end machine; then the material is reloaded and the second sample to be tested is placed on the workbench of the CT equipment. The image is scanned, and the obtained image scan data of the second sample to be detected is transmitted to the front-end machine, and so on, until the image scan data of all samples to be detected is completed.

In another embodiment, the front-end machine may not obtain the image scanning data of the sample to be detected from the CT device in real time. For example, the image scanning data of the sample to be detected obtained by the CT device may be stored first, such as in a disk, and then the image scanning data of the sample to be detected is obtained from the disk for subsequent image processing.

Optionally, the front-end machine can obtain the image scan data of the sample to be detected from the CT equipment, and reasonably and efficiently distribute the acquired image scan data of the sample to be detected, reducing the number of samples to be detected due to the increase in back-end machines. Problems such as repeated processing and/or missing processing of image scan data occur. At the same time, through reasonable distribution of image scanning data, N back-end machines can process the data efficiently to further improve processing efficiency.

After acquiring the image scan data of the sample to be detected, the front-end machine distributes the acquired image scan data of the sample to be detected to the target back-end machines among the N back-end machines. In order to reasonably and efficiently distribute the image scan data of the sample to be detected, each back-end machine is deployed with a monitoring module for monitoring the processing progress of the image scan data processed by the back-end machine. The front-end machine is specifically used to distribute the acquired image scanning data of the sample to be detected to the target back-end machine that is currently idle based on the feedback data from the monitoring module deployed on each back-end machine. The monitoring module is used to monitor the processing progress of image scanning data in real time, and through information interaction, the data is reasonably distributed according to the processing progress of each back-end machine.

It should be noted that in addition to monitoring the processing progress of the back-end machine, you can also monitor the scanning process. Monitor the progress of scanning and testing, that is, a monitoring module can be deployed in the CT equipment to monitor the progress of the CT equipment's image scanning of the sample to be tested. In addition, a monitoring module can also be deployed in the front-end machine to monitor the progress of task allocation in the front-end machine.

The monitoring module mainly conducts real-time monitoring of scanning detection, task allocation (i.e. data allocation), image processing and other processes, and builds an information transmission bridge for each step. Automatically and reasonably allocate resources and issue tasks according to the processing progress of each back-end machine, so that operators of each back-end machine know their respective task status, effectively preventing data duplication and missing processing when multiple back-end machines are used. situation occurs. Since each different sample to be tested has a unique identification (such as a QR code), real-time monitoring is carried out by deploying monitoring modules in the CT equipment, front-end machines, and each back-end machine, and through information interaction, such as CT equipment, The monitoring module of the back-end machine will feed back the monitored data to the front-end machine, so that the front-end machine can analyze the feedback data of each monitoring module and reasonably allocate resources to effectively prevent data duplication and leakage when using multiple back-end machines. Handle situations as they arise.

When the back-end machine performs image processing on the image scan data, the process can be to use three-dimensional visualization software to import the image scan data and display it in three views (such as the main view, top view, and right view) so that the operator can perform operations in each view direction. Browse and calibrate the types of defects that exist. The back-end machine can also perform slicing processing in each view direction, such as slicing according to the set slice size, and convert the three-dimensional view into a two-dimensional view for storage. In addition, the back-end machine will also respond to the user's manual calibration of the existing defect types in each view direction, obtain the corresponding defect detection results, and perform manual defect calibration on the acquired image scan data based on the defect detection results and images. Scan the data to generate a detection report and save it. It should be noted that, in one implementation, each back-end machine can generate a detection report based on the image scan data processed by itself and the defect detection results of manual defect calibration for the image scan data, and save it.

In another embodiment, one of the main back-end machines may collect the image scanning data processed by all back-end machines and the defect detection results of manual defect calibration based on the image scanning data to generate a detection report and save it. Optionally, the N back-end machines include a main back-end machine, which is used to generate inspection reports based on the image scanning data processed by all back-end machines and the defect detection results of manual defect calibration for the image scanning data, and save. Among them, the main back-end machine can be designated by a person, for example, one back-end machine is designated as the main back-end machine from N back-end machines; it can also be generated by N back-end machines through an election mechanism, which is not limited here.

The main back-end machine will process the image scan data processed by each back-end machine and manually The defect detection results of defect calibration are summarized, and a detection report is generated based on the summarized image scanning data processed by all back-end machines and the defect detection results of manual defect calibration based on the image scanning data, and saved.

In one implementation, a database module can be deployed in each back-end machine, and the database modules deployed in different back-end machines can interact with each other through information to realize image scanning data processed by all back-end machines and image scanning data. Manually summarize and classify the defect detection results of defect calibration, and use this to generate detection reports and save them. For example, the database modules in all N back-end machines except the main back-end machine will send their own processed image scan data and the defect detection results of manual defect calibration based on the image scan data to the main back-end machine. The database management module in the terminal machine is summarized.

The inspection report includes at least one and a combination of: product type, inspection quantity, defect quantity, defect type, defect rate, and inspection excellence rate. Taking the sample to be tested as a power battery as an example, the product types can be divided into two types, such as cylindrical power batteries and square power batteries. The defect types are different under different product types. For example, the defect types of cylindrical batteries are different from the defect types of square batteries. The defect rate is equal to the ratio of defective samples to be inspected/total samples to be inspected. The detection excellence rate is equal to 1-defect rate.

Optionally, the inspection report can be summarized from the dimension of product type, counting the number of inspections, number of defects, defect types, defect rate, inspection excellence rate, etc. for each product type, and presented in the form of a data report. The inspection report can also be summarized from the dimension of defect type, counting the product type, inspection quantity, defect quantity, defect rate, inspection excellence rate, etc. for each defect type, and presented in the form of a data report. By reading the inspection report, you can understand the core information at a glance, such as product type, inspection quantity, defect number, defect type, defect rate, inspection excellence rate and other information.

It should be noted that when summarizing from different dimensions, although the parameter types involved are the same, the specific parameter data are different.

It is understandable that the back-end machine can also display the detection report to let the operator know the processing status of the back-end machine, and can also respond to the user's query request and display the consulted information, such as performing inspection on the consulted detection report. show. The review and display of inspection reports are of great significance for inspection big data analysis and traceability of product problems.

In order to alleviate the data accumulation problem to the greatest extent, the value of N is not less than the quotient of the time required to process an image scan data and the time required to image scan a sample to be detected to obtain the image scan data. For example, if it takes about 9 minutes to process an image scan data, it takes about 3 minutes to scan the image of a sample to be detected and obtain the image scan data (including about 1 minute for manual loading and about 2 minutes for scanning detection). ), then the value of N is not less than 9/3=3. If the quotient is not an integer, the value of N can be no less than the integer part of the obtained quotient rounded up. For example, if the quotient is 3. .X, then the integer obtained by rounding up is 5, and X represents the decimal part.

For a better understanding, the following examples are given. The time required for image processing is about 9 minutes. The time required for image scanning of the sample to be tested to obtain the image scan data is about 3 minutes. For example, manual loading is about 1 minute. According to the detection accuracy According to the requirements, set the scanning detection time to 2 minutes (the number of collected photos is 1000, the exposure time is 120ms/photo, and the number of merged photos is 1). In this way, the manual loading (about 1 minute) + scanning detection (about 2 minutes) time is about 3 minutes. If there is only one back-end machine, connect the software and hardware as shown in Figure 3, conduct repeated tests on n samples to be tested three times, use a stopwatch to time, and take the average as the final result. According to timing calculations, theoretically the average test time of a single sample to be tested satisfies the following formula:

Formula 1: Average test time T=9+3/n, n is the total number of samples.

Based on the above formula 1, it can be seen that the total time required to detect 6 samples to be tested is (9+3/6)×6=57 minutes. There is only one back-end machine in the image processing part. When the number of samples to be tested is 6, the average test time T tends to 9.5 minutes. Since the two steps of manual loading + scanning detection and the image processing step can be parallelized, in 6 samples to be tested, During the sample testing process, (57-9)/3 = 16 samples completed manual loading + scanning testing, but only 6 samples to be tested were actually processed, which means that the data of 10 samples to be tested has accumulated. . When the number of samples to be tested is large enough, the average test time T tends to 9 minutes, and this data accumulation phenomenon will become more serious.

In order to alleviate this phenomenon, in the embodiment of this application, two back-end machines are added to the image processing part. According to the requirements of detection accuracy, set the scanning time to 2 minutes (the number of collected photos is 1000, the exposure time is 120ms/photo, and the number of merged photos is 1), so that the manual loading (about 1 minute) + scanning detection (about 2 minutes) time It takes about 3 minutes to configure three backend machines on the backend. As shown in Figure 4, connect the software and hardware, test the 6 samples to be tested in sequence, and use a stopwatch to time. According to timing calculations, theoretically the average test time of a single sample to be tested satisfies the following formula:

Formula 2: Average test time T=3+9/n, n is the total number of samples.

Based on the above formula 2, it can be seen that the total time required to detect 6 samples is (3+9/6)×6=27 minutes. There are three back-end machines at the backend. When the number of samples to be tested is 6, the average testing time T tends to 4.5 minutes. Since the two steps of manual loading + scanning detection and the image processing step can be parallelized, when there are 6 samples to be tested, During the inspection process, (27-9)/3 = 6 samples completed manual loading + scanning inspection, and 6 samples to be inspected were actually processed, which means that there was no accumulation of image scanning data. When the number of samples to be tested is large enough, the average test time T tends to 3 minutes, and the problem of data accumulation is well solved.

It can be understood that different samples to be tested may have different corresponding times during loading, scanning and inspection, and image processing. Therefore, the above example of about 1 minute for loading and scanning, and scanning and inspection cannot be combined. The examples of about 2 minutes and image processing of about 9 minutes are understood to be limitations of this application.

Based on the same inventive concept, embodiments of the present application also provide an image processing method, as shown in Figure 4. The principle will be explained below with reference to Figure 4.

S1: The front-end machine obtains the image scan data of the sample to be detected, and distributes the image scan data of the sample to be detected to the target back-end machine among the N back-end machines connected to it.

The front-end machine can obtain the image scanning data of the sample to be detected from the CT equipment in real time, and distribute the image scanning data of the sample to be detected to the target back-end machine among the N back-end machines connected to it.

In an optional implementation, the front-end machine may distribute the image scanning data of the sample to be detected to the target back-end machine among the N back-end machines connected to it in a certain order (such as sequential distribution). For example, suppose The value of N is 3, and the three back-end machines are back-end machine 1, back-end machine 2, and back-end machine 3. Then the front-end machine can process the sample to be tested according to the order of back-end machine 1, back-end machine 2, and back-end machine. Machine 3, back-end machine 1, back-end machine 2, back-end machine 3... are allocated in the order. For example, the image scanning data of sample 1 to be detected is allocated to back-end machine 1, and the image scanning data of sample 2 to be detected is allocated. To back-end machine 2, the image scan data of sample 3 to be detected is assigned to back-end machine 3, the image scan data of sample 4 to be inspected is assigned to back-end machine 1, and the image scan data of sample 5 to be inspected is assigned to back-end machine 2 , the image scan data of sample 6 to be detected is assigned to back-end machine 3, and so on.

In yet another optional implementation, a monitoring module is deployed in each back-end machine, and the monitoring module is used to monitor the processing progress of the image scan data processed by the back-end machine. Then scan the image number of the sample to be detected The process of allocating the data to the target back-end machine among the N back-end machines connected to it may be: based on the feedback data of the monitoring module deployed on each back-end machine, determine the target back-end machine that is currently idle, and transfer the data to the target back-end machine that is currently idle. The image scan data of the inspection sample is distributed to the target back-end machine.

In addition, in addition to deploying monitoring modules in back-end machines, monitoring modules can also be deployed in CT equipment to monitor the progress of scanning and detection, and in front-end machines to monitor the progress of scanning and detection. Monitor the progress of task assignments. Through real-time monitoring of scanning detection, task allocation (i.e. data allocation), image processing and other processes, we can automatically allocate resources and issue tasks reasonably according to the processing progress of each back-end machine, effectively preventing the use of multiple back-end machines. The occurrence of repeated data processing and missing processing.

S2: The target backend machine performs image processing on the allocated image scan data.

When the target back-end machine performs image processing on the assigned image scan data, the process can be: the target back-end machine uses three-dimensional visualization software to import the image scan data for three-view display, and responds to the user's comments on the existing defect types in each view direction. Perform manual calibration operations to obtain corresponding defect detection results; the target back-end machine performs slicing processing in each view direction, converting the three-dimensional view into a two-dimensional view and saving it.

The target back-end machine is also used to generate an inspection report based on the obtained defect detection results of manual defect calibration based on the image scan data and the image scan data, and save them.

In one implementation, each back-end machine can generate a detection report based on the image scan data processed by itself and the defect detection results of manual defect calibration for the image scan data, and save it. In another embodiment, one of the main back-end machines may collect the image scanning data processed by all back-end machines and the defect detection results of manual defect calibration based on the image scanning data to generate a detection report and save it. Optionally, the N back-end machines include a main back-end machine, and the image processing method also includes: the main back-end machine performs defect detection results based on the image scanning data processed by all back-end machines and manually performs defect calibration on the image scanning data, Generate a detection report and save it.

The inspection report includes at least one and a combination of: product type, inspection quantity, defect quantity, defect type, defect rate, and inspection excellence rate. Taking the sample to be tested as a power battery as an example, the product types can be divided into two types, such as cylindrical power batteries and square power batteries. The defect types are different under different product types. For example, the defect types of cylindrical batteries are different from the defect types of square batteries. The defect rate is equal to the ratio of defective samples to be inspected/total samples to be inspected. The detection excellence rate is equal to 1- Defect rate.

The implementation principles and technical effects of the image processing method provided by the embodiments of the present application are the same as those of the aforementioned detection system embodiments. This is a brief description. For matters not mentioned in the method embodiments, please refer to the corresponding ones in the aforementioned detection system embodiments. content.

It should be noted that each embodiment in this specification is described in a progressive manner. Each embodiment focuses on its differences from other embodiments. The same and similar parts between the various embodiments are referred to each other. Can.

The above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application. should be covered by the protection scope of this application. Therefore, the protection scope of this application should be determined by the protection scope of the claims.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present application, but not to limit it; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: The technical solutions described in the foregoing embodiments can still be modified, or some or all of the technical features can be equivalently replaced; and these modifications or substitutions do not deviate from the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present application. The scope shall be covered by the claims and description of this application. In particular, as long as there is no structural conflict, the technical features mentioned in the various embodiments can be combined in any way. The application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

A detection system, including:

The front-end machine is used to distribute the image scan data of the sample to be tested;

N back-end machines, each of the N back-end machines is connected to the front-end machine, and each back-end machine is used to image the image scanning data allocated by the front-end machine. Processing, N is an integer greater than or equal to 2, and the value of N is not less than the quotient of the time required to process one image scan data and the time required to image scan one of the samples to be detected to obtain image scan data.
The detection system according to claim 1, wherein a monitoring module is deployed in each of the back-end machines, and the monitoring module is used to monitor the processing progress of the image scan data processed by the back-end machine;

The front-end machine is specifically configured to distribute the image scanning data of the sample to be detected to the target back-end machine among the N back-end machines based on the feedback data from the monitoring module deployed on each of the back-end machines. .
The detection system according to claim 1 or 2, wherein the N back-end machines include a main back-end machine;

The main back-end machine is used to generate a detection report based on the image scanning data processed by all back-end machines and the defect detection results of manual defect calibration for the image scanning data, and save it.
The detection system according to claim 3, wherein the detection report includes: any combination of product type, detection quantity, defect quantity, defect type, defect rate, and detection excellence rate.
The detection system according to any one of claims 1 to 4, wherein the detection system further includes: CT equipment connected to the front-end machine, the CT equipment is used to detect objects placed on its own workbench. The sample to be tested is image scanned to obtain the image scan data of the sample to be tested.
An image processing method, including:

The front-end machine obtains the image scan data of the sample to be detected, and distributes the image scan data of the sample to be detected to the target back-end machine among the N back-end machines connected to it, where N is an integer greater than or equal to 2, and N is The value is not less than the quotient of the time required to process one image scan data and the time required to obtain image scan data by image scanning one of the samples to be detected;

The target backend machine performs image processing on the allocated image scan data.
The image processing method according to claim 6, wherein a monitoring module is deployed in each back-end machine, and the monitoring module is used to monitor the processing progress of the image scanning data processed by the back-end machine; The image scan data of the sample is distributed to the target back-end machine among the N back-end machines connected to it, including:

Determine the target backend machine currently in an idle state based on the feedback data from the monitoring module deployed on each backend machine;

The image scan data of the sample to be detected is assigned to the target back-end machine.
The image processing method according to claim 6 or 7, wherein the target backend machine performs image processing on the allocated image scan data, including:

The target back-end machine uses three-dimensional visualization software to import the image scan data for three-view display, and responds to the user's manual calibration of existing defect types in each view direction to obtain corresponding defect detection results;

The target back-end machine performs slicing processing in each view direction, and converts the three-dimensional view into a two-dimensional view for storage.
The image processing method according to any one of claims 6-8, wherein the N back-end machines include a main back-end machine, and the method further includes:

The main back-end machine generates a detection report based on the image scan data processed by all back-end machines and the defect detection results of manual defect calibration based on the image scan data, and saves it.