A quality inspection method and system capable of performing image recognition

Abstract

The present invention discloses a quality inspection method and system capable of performing image recognition, which relates to the technical field of equipment quality inspection. The method first obtains an equipment image of the equipment to be inspected, then recognizes the equipment image to obtain equipment acquisition information of the equipment to be inspected, and finally compares the equipment acquisition information with equipment standard information of the equipment to be inspected to obtain the quality inspection result of the equipment to be inspected. In this way, the equipment image acquisition, recognition and comparative analysis processes can be automatically performed to automatically complete the equipment quality inspection process without the participation of quality inspection engineers, and the quality inspection efficiency is high.

Description

Technical Field

[0001] The present invention relates to the technical field of equipment quality inspection, and in particular to a quality inspection method and system capable of performing image recognition. Background Art

[0002] Currently, when conducting quality inspection on equipment, quality inspection engineers are often required to go to the site for inspection, which has high labor costs. In addition, the amount of equipment that needs to be inspected is large, while the number of quality inspection engineers is limited, resulting in low quality inspection efficiency.

[0003] Based on this, there is an urgent need for a technology that can perform automatic quality inspection. Summary of the Invention

[0004] The purpose of the present invention is to provide a quality inspection method and system capable of performing image recognition, which can automatically complete the quality inspection process of equipment without the participation of quality inspection engineers and has high quality inspection efficiency.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A quality inspection method capable of performing image recognition, the quality inspection method comprising:

[0007] Obtain equipment pictures of the equipment to be inspected;

[0008] Identify the device image to obtain device collection information of the device to be inspected;

[0009] The device collected information is compared with the device standard information of the device to be quality inspected to obtain the quality inspection result of the device to be quality inspected.

[0010] In some embodiments, obtaining a device image of the device to be inspected specifically includes:

[0011] Take a picture of the equipment to be inspected to obtain a picture of the equipment to be inspected;

[0012] Alternatively, a video captured by a camera installed at the device to be inspected is obtained, and the video is processed to obtain a device image of the device to be inspected.

[0013] In some embodiments, the device picture includes a picture of a faulty component of the device to be inspected; the faulty component is a component whose failure probability is greater than a set value.

[0014] In some embodiments, when obtaining a device image of the device to be inspected, the quality inspection method further includes:

[0015] Read the device information of the device to be inspected; the device information includes the appearance parameters and instrument panel values ​​of the device to be inspected.

[0016] In some embodiments, before identifying the device image, the quality inspection method further includes:

[0017] A similarity calculation is performed on the device image and the standard image of the device to be inspected, and a device image with a similarity greater than a preset threshold is selected as a new device image.

[0018] In some embodiments, the device collection information includes device type, device parameters and dashboard values; the device standard information includes standard data of device type, device parameters and dashboard values.

[0019] A quality inspection system capable of performing image recognition, the quality inspection system comprising:

[0020] Image acquisition module, used to obtain device images of the equipment to be inspected;

[0021] An image recognition module is used to recognize the device image and obtain the device collection information of the device to be inspected;

[0022] The image analysis module is used to compare the device collection information with the device standard information of the device to be quality inspected to obtain the quality inspection result of the device to be quality inspected.

[0023] In some embodiments, the image acquisition module is further used to read device information of the device to be inspected; the device information includes appearance parameters and instrument panel values ​​of the device to be inspected.

[0024] In some embodiments, the image recognition module is further configured to calculate a similarity between the device image and a standard image of the device to be inspected, and select a device image with a similarity greater than a preset threshold as a new device image.

[0025] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0026] The present invention is used to provide a quality inspection method and system that can perform image recognition. First, a device image of the device to be inspected is obtained, and then the device image is recognized to obtain device collection information of the device to be inspected. Finally, the device collection information is compared with the device standard information of the device to be inspected to obtain the quality inspection result of the device to be inspected. In this way, the device image collection, image recognition and comparative analysis processes can be automatically performed to automatically complete the quality inspection process of the equipment without the participation of quality inspection engineers, and the quality inspection efficiency is high. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0028] Figure 1 A flow chart of the quality inspection method provided in Example 1 of the present invention;

[0029] Figure 2 This is a flowchart of the quality inspection method provided in Example 1 of the present invention;

[0030] Figure 3 This is a system block diagram of the quality inspection system provided in Example 2 of the present invention;

[0031] Figure 4 This is a workflow diagram of the quality inspection system provided in Example 2 of the present invention;

[0032] Figure 5 A schematic diagram of the workflow of a system administrator in the quality inspection system provided in Example 2 of the present invention;

[0033] Figure 6 This is a schematic diagram of the workflow of the person in charge of a work order in the quality inspection system provided in Example 2 of the present invention;

[0034] Figure 7 This is a schematic diagram of the workflow of a quality inspection administrator in the quality inspection system provided in Example 2 of the present invention. DETAILED DESCRIPTION

[0035] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0036] The purpose of the present invention is to provide a quality inspection method and system capable of performing image recognition, which can automatically complete the quality inspection process of equipment without the participation of quality inspection engineers and has high quality inspection efficiency.

[0037] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0038] Example 1:

[0039] This embodiment is used to provide a quality inspection method that can perform image recognition, such as Figure 1 and Figure 2 As shown, the quality inspection method includes:

[0040] S1: Obtain the device image of the device to be inspected;

[0041] The equipment to be inspected in this embodiment may be medical equipment and industrial equipment. Medical equipment includes CT machines and MRI machines, and industrial equipment includes ship engines, etc.

[0042] Specifically, S1 may include: taking a photo of the device to be inspected to obtain a device image of the device to be inspected. In this embodiment, the device to be inspected may be photographed by a human handheld mobile device, and the device image obtained by taking the photo may be uploaded to a server for subsequent analysis and processing.

[0043] Since the working environment of some equipment to be inspected may be relatively narrow, closed or dangerous, this embodiment can obtain equipment images by installing fixed equipment around the equipment to be inspected. There is no need for manual handheld mobile devices to take pictures to obtain equipment images, which can avoid manual participation in the quality inspection process of dangerous equipment to be inspected and reduce the possibility of personal injury. At the same time, installing fixed equipment for automatic inspection can greatly improve the efficiency of obtaining equipment images and improve the efficiency of discovering faulty equipment compared to the manual method of obtaining equipment images.

[0044] Specifically, installing a fixed device can be installing a fixed camera near the equipment to be inspected. At this time, S1 can include: obtaining a video captured by a camera installed at the equipment to be inspected, and processing the video. The processing can be video frame extraction to obtain a device picture of the equipment to be inspected, thereby reducing human participation in taking pictures and improving the efficiency of obtaining equipment pictures.

[0045] To facilitate subsequent image analysis, the device images in this embodiment include images of faulty components of the equipment under inspection. Faulty components are components with a failure probability greater than a set value, which can be 60%. This means that the device images in this embodiment include components prone to failure within the equipment under inspection, allowing for targeted quality inspections and improving inspection efficiency. When photographing the equipment under inspection, it is important to include images of components prone to failure. Examples of these components include the coil and indicator light of an MRI scanner, as well as the piston and instrument panel of an engine.

[0046] The device pictures of this embodiment may be multiple pictures obtained by taking pictures of the quality inspection equipment from different angles. The device collection information can be subsequently obtained by analyzing the device pictures to perform quality inspection on the quality inspection equipment.

[0047] In order to conduct a more comprehensive quality inspection of the equipment to be inspected, when obtaining a device image of the equipment to be inspected, the quality inspection method of this embodiment further includes: reading the device information of the equipment to be inspected in real time. The device information includes the appearance parameters and instrument panel values ​​of the equipment to be inspected. Subsequently, the appearance parameters can be used to analyze whether there is damage to the exterior of the equipment to be inspected. The instrument panel values ​​can be used to analyze whether there is leakage of liquid inside the equipment to be inspected. The instrument panel values ​​can also be used to analyze whether the equipment to be inspected requires maintenance, such as whether the temperature is too high or whether water or fuel needs to be added. Specifically, the appearance parameters of this embodiment include color depth and edge line curvature. The color depth can be used to determine whether the color depth distribution is uniform and whether there is a significant change in color depth. The edge line curvature can be used to determine whether the curvature has significantly changed compared to the curvature standard value. If the distribution is uneven, there is a significant change in color depth, or the curvature has significantly changed compared to the curvature standard value, then the exterior of the equipment to be inspected is damaged, thereby achieving the function of determining whether there is external damage.

[0048] S2: Identify the device image to obtain device collection information of the device to be inspected;

[0049] Prior to S2, the quality inspection method of this embodiment further includes classifying and screening the collected device images, deleting invalid device images, using valid device images as new device images, and executing the image recognition process of S2. Specifically, the classification and screening process may include: using an image recognition engine to perform similarity analysis, that is, calculating the similarity between the device image and the standard image of the device to be inspected, classifying and screening based on the similarity, and selecting device images with a similarity greater than a preset threshold as new device images. Classification and screening may also be performed based on both similarity and recognition status, selecting device images with a similarity greater than a preset threshold and with important device parts (important parts can be set based on experience) or instrument panel values ​​that can be identified as new device images.

[0050] It should be noted that this embodiment uses a device image obtained by photographing standard equipment (standard equipment refers to original, unused, intact equipment awaiting quality inspection) at a predetermined shooting angle as the standard image. The predetermined shooting angle can be user-defined. The classification, screening, and image recognition processes of this embodiment can both utilize existing image recognition technologies, such as neural network algorithm engines, and will not be further described here.

[0051] S3: Compare the device collection information with the device standard information of the device to be quality inspected to obtain the quality inspection result of the device to be quality inspected.

[0052] In this embodiment, the device collection information includes the device type, device parameters, and instrument panel values. Device parameters may include the device number, the device's appearance at various angles and locations (such as shape and size), and the shape, width, length, and color of key device components. Device standard information includes standard data for the device type, device parameters, and instrument panel values. Standard data for instrument panel values ​​generally refers to a safe range for instrument panel values.

[0053] By comparing the device collection information with the device standard information, it is possible to determine whether the device under quality inspection has any problems and output the corresponding quality inspection results. In this embodiment, the user can customize the qualified standards, such as the qualified range of each indicator, so that by comparing the device collection information with the device standard information, it can be determined whether the device under quality inspection is qualified.

[0054] This embodiment provides a quality inspection method capable of image recognition. The quality inspector uses a mobile device to take a photo of the equipment to be inspected and upload it, or uses a camera to capture a video of the equipment to be inspected, to obtain a device image of the equipment to be inspected. The intelligent image recognition technology is used to extract the equipment acquisition information, and the equipment acquisition information is compared and analyzed with the equipment standard information. After calculation, a valid quality inspection result is output, and a corresponding quality inspection report can be given, thereby automatically inspecting the equipment to be inspected and providing a decision-making basis for the quality inspection engineer. By remotely uploading equipment images for quality inspection, the number of times the quality inspection engineer has to go to the site is reduced, which greatly saves on travel expenses and reduces the cost of quality inspection. By using automated image recognition technology for quality inspection, a large number of quality inspection engineers are no longer required to conduct quality inspections, which reduces the company's labor costs. The quality inspection efficiency is much higher than that of manual quality inspection, which greatly improves the quality inspection efficiency.

[0055] After the automatic quality inspection of this embodiment, the quality inspection engineer can only conduct further on-site quality inspection on equipment that does not meet the standards, which effectively reduces the frequency of on-site quality inspections, improves work efficiency, and reduces labor costs.

[0056] Example 2:

[0057] This embodiment is used to provide a quality inspection system that can perform image recognition, such as Figure 3 As shown, the quality inspection system includes:

[0058] The picture acquisition module is used to obtain equipment pictures of the equipment to be inspected; preferably, the picture acquisition module is also used to read equipment information of the equipment to be inspected, and the equipment information includes appearance parameters and instrument panel values ​​of the equipment to be inspected.

[0059] The image acquisition module of this embodiment is responsible for collecting information of the equipment to be inspected, including a manual acquisition submodule and an active acquisition submodule. The manual acquisition submodule supports taking pictures through a mobile device and uploading them to the server to obtain device images of the equipment to be inspected; the active acquisition module supports real-time reading of device information of the equipment to be inspected from the monitoring device of the equipment to be inspected.

[0060] The image recognition module is used to identify the device image and obtain the device collection information of the device to be inspected; the image recognition module is also used to calculate the similarity between the device image and the standard image of the device to be inspected, and select the device image with a similarity greater than a preset threshold as the new device image.

[0061] The image recognition module of this embodiment is responsible for classifying and screening the collected device images and deleting invalid device images. Specifically, it calls a third-party image recognition service to calculate the similarity between the device image and the standard image and to determine whether the important parts of the device or the instrument panel values ​​can be identified, so as to determine invalid device images and further extract valid device collection information such as device parameters and device types from valid device images for subsequent comparative analysis.

[0062] The image analysis module is used to compare the device collected information with the device standard information of the device to be inspected to obtain the quality inspection results of the device to be inspected.

[0063] The image analysis module of this embodiment is responsible for analyzing the equipment collection information. By comparing the equipment standard information and the equipment collection information, it analyzes whether the status of the equipment to be inspected is normal, obtains the quality inspection results, and outputs the quality inspection report according to the report parameters set by the system.

[0064] The quality inspection system of this embodiment may also include a system setting module, which is responsible for entering standard images of equipment, standard data of equipment parameters, equipment types and dashboard values, and basic data such as qualification standards into the quality inspection system to provide a basis for judgment for comparative analysis. The system setting module can also perform image recognition training to improve the success rate of device image recognition.

[0065] like Figure 4 、 Figure 5 、 Figure 6 and Figure 7 As shown, the system administrator refers to the IT manager of the entire system, who is responsible for the overall configuration and management of the system. The work order manager is the engineer responsible for the actual quality inspection. The quality inspection manager is the administrator responsible for the overall quality inspection business and manages the work order managers. The working process of the quality inspection system of this embodiment may include:

[0066] (1) Image collection: Obtain device images including device appearance and dashboard information through mobile devices and fixed monitoring devices, and upload the device images to intelligent quality inspection;

[0067] (2) Standard image upload: Upload standard images of standard equipment.

[0068] (3) Standard collection: Input the standard data of standard equipment (i.e., equipment standard information), such as the normal value range of the instrument panel, the normal color of the equipment, the normal width and other appearance indicators.

[0069] (4) Rule setting: Set identification rules and classification rules, classify standard images and bind them with standard values, and set the report content corresponding to various values.

[0070] (5) Intelligent quality inspection: Based on standard images, the collected equipment images are analyzed to classify the images, and then image recognition is performed on the valid equipment images to extract the equipment collection information. The equipment collection information is further compared and analyzed with the equipment standard information to generate a quality inspection report.

[0071] The quality inspection system provided in this embodiment is implemented based on image recognition and comparison technology. It consists of four functional modules and ten submodules that work together to complete image recognition quality inspection. First, the standard equipment is analyzed, and quality inspection indicators are compiled and entered into the quality inspection system as a basis for analysis. Images of the equipment to be inspected are captured by taking photos. Image recognition is performed on the device images to obtain device acquisition information. This acquisition information represents the acquisition values ​​of the quality inspection indicators. This acquisition information is then compared with the standard equipment information to obtain differentiated data. The standard equipment information represents the standard values ​​of the quality inspection indicators. Quality inspection results are then calculated based on this differentiated data.

[0072] Each embodiment in this specification focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0073] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A quality inspection method capable of image recognition, characterized in that: The quality inspection method includes: Obtaining device images of the device to be inspected; the device images are multiple images obtained by taking photos of the device to be inspected from different angles; Read device information of the device to be inspected; the device information includes appearance parameters and instrument panel values ​​of the device to be inspected; analyze whether there is damage to the exterior of the device to be inspected based on the appearance parameters, the appearance parameters include color depth and edge line curvature; judge whether the color depth distribution is uniform and whether there is a color depth change based on the color depth; judge whether the curvature is different from the curvature standard value based on the edge line curvature; if the color depth distribution is uneven, there is a color depth change, or the curvature is different from the curvature standard value, then there is damage to the exterior of the device to be inspected; analyze whether there is leakage of liquid inside the device to be inspected and whether the device to be inspected needs maintenance based on the instrument panel values; Calculate the similarity between the device image and the standard image of the device to be inspected, select the device image with a similarity greater than a preset threshold as a new device image, identify the new device image, and obtain device acquisition information of the device to be inspected; the device acquisition information includes the device type, device parameters, and instrument panel values; the device parameters include the device number, the appearance of each angle and each part of the device, and the shape, width, length, and color of important parts of the device; The device collection information is compared with the device standard information of the device to be inspected to obtain the quality inspection result of the device to be inspected; the device standard information includes standard data of device type, device parameters and instrument panel values.

2. The quality inspection method according to claim 1, characterized in that: The obtaining of the device image of the device to be inspected specifically includes: Take a picture of the equipment to be inspected to obtain a picture of the equipment to be inspected; Alternatively, a video captured by a camera installed at the device to be inspected is obtained, and the video is processed to obtain a device image of the device to be inspected.

3. The quality inspection method according to claim 1, characterized in that: The equipment picture includes a picture of a faulty component of the equipment to be inspected; the faulty component is a component whose failure probability is greater than a set value.

4. A quality inspection system capable of image recognition, characterized in that: The quality inspection system includes: The image acquisition module is used to obtain device images of the device to be inspected; the device images are multiple images taken from different angles of the device to be inspected; The image acquisition module is also used to read the device information of the device to be inspected; the device information includes the appearance parameters and instrument panel values ​​of the device to be inspected; according to the appearance parameters, it is analyzed whether there is damage on the outside of the device to be inspected, the appearance parameters include color depth and edge line curvature, and according to the color depth, it is judged whether the distribution of color depth is uniform and whether there is any color depth change, and according to the edge line curvature, it is judged whether the curvature is changed compared with the curvature standard value. If the distribution of color depth is uneven, there is a color depth change, or the curvature is changed compared with the curvature standard value, then there is damage on the outside of the device to be inspected, and according to the instrument panel values, it is analyzed whether there is leakage of liquid inside the device to be inspected and whether the device to be inspected needs maintenance; An image recognition module is configured to calculate the similarity between the device image and a standard image of the device to be inspected, select a device image with a similarity greater than a preset threshold as a new device image, identify the new device image, and obtain device acquisition information of the device to be inspected; the device acquisition information includes the device type, device parameters, and instrument panel values; the device parameters include the device number, the appearance of each angle and each part of the device, and the shape, width, length, and color of important parts of the device; The image analysis module is used to compare the device collection information with the device standard information of the device to be inspected to obtain the quality inspection result of the device to be inspected; the device standard information includes standard data of device type, device parameters and instrument panel values.

Images