Method and system for monitoring packaging quality of electroluminescent device

Abstract

The invention relates to the technical field of electroluminescent device packaging, and discloses an electroluminescent device packaging quality monitoring method and system, and the method comprises the steps: obtaining a packaged electroluminescent device image, carrying out the graying processing of the electroluminescent device image, obtaining a device gray image, and carrying out the compression transmission of the device gray image; the background server decompresses the compressed data of the grayscale image of the device, and performs defect detection on the decompressed grayscale image of the device to obtain a quality defect parameter; extracting environmental interference characteristics of the device grayscale image, determining a quality interference parameter according to the environmental interference characteristics, and determining a packaging quality parameter according to the quality interference parameter and the quality defect parameter; and determining the packaging quality of the electroluminescent device according to the packaging quality parameter, and judging whether to carry out process regulation or not according to the packaging quality. According to the invention, the quality of the packaged electroluminescent device is monitored, so that the packaging quality can be effectively improved.

Description

Technical Field

[0001] This application relates to the field of electroluminescent device packaging technology, and more specifically, to a method and system for monitoring the packaging quality of electroluminescent devices. Background Technology

[0002] Electroluminescent devices (LEDs) are solid-state semiconductor light-emitting devices that directly convert electrical energy into light energy. As a modern new type of light source, LEDs have advantages such as long lifespan, high efficiency, energy saving, and environmental friendliness. Quality management is a crucial aspect of the manufacturing process of LEDs. With the rapid development of advanced packaging technologies and the introduction of various new materials and processes, the complexity of their manufacturing is constantly increasing, leading to a more diverse range of sources of quality problems.

[0003] Traditional quality management methods rely mainly on manual labor and experience-based rules; however, this approach is inefficient and no longer meets the needs of modern manufacturing. Summary of the Invention

[0004] This invention provides a method and system for monitoring the packaging quality of electroluminescent devices, to solve the problem of low efficiency in monitoring the packaging quality of electroluminescent devices in the prior art, including: The process involves acquiring an image of the packaged electroluminescent device, converting the image to grayscale to obtain a grayscale image, and then compressing and transmitting this image. The backend server decompresses the compressed data of the grayscale image and performs defect detection to obtain quality defect parameters. Environmental interference features of the grayscale image during the packaging process are extracted, and quality interference parameters are determined based on these features. Packaging quality parameters are then determined based on the quality interference and defect parameters. Finally, the packaging quality of the electroluminescent device is determined based on these quality parameters, and the need for process adjustments is assessed based on the overall packaging quality.

[0005] Further, the compression and transmission of the device grayscale image includes: obtaining a preset segmentation window; segmenting the device grayscale image according to the preset segmentation window to obtain several device grayscale image blocks; obtaining the grayscale value of each pixel in the device grayscale image block; clustering each pixel in the device grayscale image block according to the grayscale value to obtain k clusters; determining the distance between the corresponding cluster center of each cluster and the other cluster centers and the number of pixel types in each cluster according to the clustering results; determining the compression efficiency according to the distance between the corresponding cluster center of each cluster and the other cluster centers and the number of pixel types in each cluster; and selecting the compression... For inefficient device grayscale image blocks with compression efficiency less than a first preset threshold, a scaling parameter for a preset segmentation window is determined based on the compression efficiency of the inefficient device grayscale image block, and the preset segmentation window is scaled according to the scaling parameter. The inefficient device grayscale image block is segmented according to the scaled preset segmentation window to obtain several corrected device grayscale image blocks. The grayscale sequences of device grayscale image blocks with compression efficiency greater than or equal to the first preset threshold and the corrected device grayscale image blocks are counted. Huffman coding is performed on the grayscale sequences of each device grayscale image block to obtain compressed device grayscale image data, and the compressed device grayscale image data is transmitted to the backend server.

[0006] Further, determining the compression efficiency based on the distance between the corresponding cluster center of each cluster and the other cluster centers and the number of pixel types within each cluster includes: determining the compression efficiency according to a compression efficiency calculation formula, wherein the compression efficiency calculation formula is... ; in, For compression efficiency, For the first The distance between the cluster center and the other cluster centers corresponds to the cluster cluster. This represents the total number of clusters in the grayscale image block of the device. This represents the number of different pixel types in the grayscale image block of the device. For the first The number of pixel types in a cluster. This is the normalization function.

[0007] Further, the step of performing defect detection on the decompressed device grayscale image to obtain quality defect parameters includes: performing edge detection on the device grayscale image based on the Canny edge detection algorithm to obtain several edge extraction images of the device grayscale image; performing feature extraction on the several edge extraction images of the device grayscale image to extract the texture features of the several edge extraction images; determining the defect confidence of each edge extraction image based on the texture features of the edge extraction images; extracting the number of edge extraction images with defect confidence greater than a first preset threshold; and determining the quality defect parameters based on the number of edge extraction images with defect confidence greater than the first preset threshold.

[0008] Further, determining the defect confidence of each edge extraction image based on the texture features of the edge extraction image includes: acquiring the defect texture features of grayscale images of historical defective devices, establishing a training sample set based on the defect texture features of grayscale images of historical defective devices; establishing an initial defect evaluation model based on the training sample set and training the initial defect evaluation model to obtain a trained defect evaluation model; acquiring the texture features of the current edge extraction image, inputting the texture features of the current edge extraction image into the trained defect evaluation model to obtain the defect confidence of the corresponding edge extraction image.

[0009] Furthermore, the step of extracting environmental interference features of the device grayscale image during the packaging process and determining quality interference parameters based on these features includes: extracting electrostatic spark features and dust particle size features of the device grayscale image during the packaging process, and determining quality interference parameters based on these features.

[0010] Furthermore, determining the quality interference parameters based on the electrostatic spark characteristics and dust particle size characteristics of the device grayscale image includes: determining the quality interference parameters according to a quality interference parameter calculation formula, wherein the quality interference parameter calculation formula is as follows: , in, For quality interference parameters, Characterized by electrostatic sparks. Characteristics of dust particle size. To preset the electrostatic spark weight, Preset dust particle size weights.

[0011] Further, determining the packaging quality of the electroluminescent device based on the packaging quality parameters includes: obtaining preset standard packaging quality parameters; calculating the difference between the packaging quality parameters and the preset standard packaging quality parameters; determining whether the difference between the packaging quality parameters and the preset standard packaging quality parameters is greater than a second preset threshold; if the difference between the packaging quality parameters and the preset standard packaging quality parameters is greater than the second preset threshold, then the corresponding electroluminescent device is determined to have qualified packaging quality; if the difference between the packaging quality parameters and the preset standard packaging quality parameters is less than or equal to the second preset threshold, then determining whether the difference between the packaging quality parameters and the preset standard packaging quality parameters is greater than a third preset threshold; if the difference between the packaging quality parameters and the preset standard packaging quality parameters is greater than the third preset threshold, then the corresponding electroluminescent device is determined to have unqualified packaging quality.

[0012] Furthermore, the step of determining whether to perform process control based on packaging quality includes: obtaining the number of packaging failures of electroluminescent devices in a preset batch, calculating the ratio of the number of packaging failures of electroluminescent devices in a preset batch to the preset allowable number of failures, and obtaining the packaging failure rate; when the packaging failure rate is less than a fourth preset threshold, triggering a process control instruction to perform packaging process control.

[0013] To achieve the above objectives, the present invention also provides a packaging quality monitoring system for electroluminescent devices, comprising: The transmission module acquires images of the packaged electroluminescent device, performs grayscale processing on these images to obtain grayscale images, and then compresses and transmits the grayscale images. The detection module decompresses the compressed data of the grayscale images on the backend server, performs defect detection on the decompressed grayscale images, and obtains quality defect parameters. The environment module extracts environmental interference features from the grayscale images, determines quality interference parameters based on these features, and then determines packaging quality parameters based on the quality interference and quality defect parameters. The control module determines the packaging quality of the electroluminescent device based on the packaging quality parameters and determines whether process control is needed based on the packaging quality.

[0014] The beneficial effects of this invention are as follows: By applying the above technical solutions, this invention enhances the transmission efficiency of images by compressing and transmitting the obtained grayscale images of the devices, effectively improving the quality detection efficiency of electroluminescent devices. By combining the environmental interference characteristics of the electroluminescent devices during packaging and the quality defect parameters after packaging, the packaging quality of the electroluminescent devices can be accurately monitored and the process can be adjusted in a timely manner, ensuring the monitoring efficiency and packaging quality of the electroluminescent devices. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 The overall flowchart of a method for monitoring the packaging quality of an electroluminescent device according to an embodiment of the present invention is shown; Figure 2 A schematic diagram of a packaging quality monitoring system for electroluminescent devices proposed in an embodiment of the present invention is shown. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0018] This application provides a method for monitoring the packaging quality of electroluminescent devices, such as... Figure 1 As shown, it includes: S101: Obtain an image of the packaged electroluminescent device, perform grayscale processing on the electroluminescent device image to obtain a grayscale image of the device, and compress and transmit the grayscale image of the device. In this embodiment, after the electroluminescent device is packaged, an image of the packaged electroluminescent device is captured by a high-definition camera. The image is then converted to grayscale to facilitate subsequent quality monitoring. Since there are a large number of electroluminescent devices in a batch, the grayscale images of the devices after grayscale processing need to be compressed and transmitted to improve the efficiency of quality inspection.

[0019] In some embodiments of this application, the compression and transmission of the device grayscale image includes: obtaining a preset segmentation window; segmenting the device grayscale image according to the preset segmentation window to obtain several device grayscale image blocks; obtaining the grayscale value of each pixel in the device grayscale image block; clustering each pixel in the device grayscale image block according to the grayscale value to obtain k clusters; determining the distance between the corresponding cluster center of each cluster and the other cluster centers and the number of pixel types in each cluster according to the clustering results; and determining the compression efficiency according to the distance between the corresponding cluster center of each cluster and the other cluster centers and the number of pixel types in each cluster. Inefficient device grayscale image blocks with compression efficiency less than a first preset threshold are selected. A scaling parameter for a preset segmentation window is determined based on the compression efficiency of the inefficient device grayscale image blocks, and the preset segmentation window is scaled according to the scaling parameter. The inefficient device grayscale image blocks are segmented according to the scaled preset segmentation window to obtain several corrected device grayscale image blocks. The grayscale sequences of device grayscale image blocks with compression efficiency greater than or equal to the first preset threshold and the corrected device grayscale image blocks are statistically analyzed. Huffman coding is performed on the grayscale sequences of each device grayscale image block to obtain compressed device grayscale image data, and the compressed device grayscale image data is transmitted to the backend server.

[0020] In this embodiment, the size of the preset segmentation window is determined based on the size of the device grayscale image. The device grayscale image is divided into several grayscale image blocks by the preset segmentation window. Huffman coding is performed on each image block according to the compression efficiency of each image block. Huffman coding achieves compression by assigning shorter codes to data with higher frequencies and longer codes to data with lower frequencies. Therefore, when the frequencies of grayscale values ​​of pixels in an image block differ significantly, Huffman coding can achieve high compression efficiency. Since electroluminescent device images have local similarity, the grayscale values ​​in their local locations are quite similar, and the repetition rate of some grayscale values ​​is high, resulting in a large frequency difference for each grayscale value in local locations. Huffman coding can achieve high compression efficiency. Therefore, a scaling parameter is set by a preset compression efficiency-scaling parameter mapping table. The lower the compression efficiency, the smaller the corresponding scaling parameter, and the smaller the area of ​​the preset segmentation window after scaling. The image block is then segmented a second time by the scaled preset segmentation window, making the frequency difference of each grayscale value in each image block larger, thereby improving the compression efficiency.

[0021] In some embodiments of this application, determining the compression efficiency based on the distance between the corresponding cluster center of each cluster and the other cluster centers and the number of pixel types within each cluster includes: determining the compression efficiency according to a compression efficiency calculation formula, wherein the compression efficiency calculation formula is... ; in, For compression efficiency, For the first The distance between the cluster center and the other cluster centers corresponds to the cluster cluster. This represents the total number of clusters in the grayscale image block of the device. This represents the number of different pixel types in the grayscale image block of the device. For the first The number of pixel types in a cluster. This is the normalization function.

[0022] In this embodiment, the compression efficiency is determined by the distance between the corresponding cluster center of each cluster and the other cluster centers, and the number of pixel types within each cluster. The larger the value, the greater the grayscale difference in the image block, and the higher the compression efficiency. Several grayscale value ranges are preset; the type of a pixel is determined by the grayscale value range to which it belongs. The difference between the average number of pixel types in all pixels in a grayscale image block and the number of pixel types in a cluster is calculated. The greater the difference, the more disordered the pixel distribution in the cluster and the lower the compression efficiency. The compression efficiency of each image block can be accurately calculated using the compression efficiency calculation formula.

[0023] S102, the background server decompresses the compressed data of the device grayscale image, performs defect detection on the decompressed device grayscale image, and obtains quality defect parameters; In some embodiments of this application, the step of performing defect detection on the decompressed device grayscale image to obtain quality defect parameters includes: performing edge detection on the device grayscale image based on the Canny edge detection algorithm to obtain several edge extraction images of the device grayscale image; performing feature extraction on the several edge extraction images of the device grayscale image to extract the texture features of the several edge extraction images; determining the defect confidence of each edge extraction image based on the texture features of the edge extraction images; extracting the number of edge extraction images with defect confidence greater than a first preset threshold; and determining the quality defect parameters based on the number of edge extraction images with defect confidence greater than the first preset threshold.

[0024] In this embodiment, several edge extraction images of the device grayscale image are detected by the Canny edge detection algorithm. By performing feature extraction on each edge extraction image, the texture features of each edge extraction image are extracted, thereby obtaining the defect confidence of each edge extraction image and calculating the quality defect parameters.

[0025] In some embodiments of this application, determining the defect confidence of each edge extraction image based on the texture features of the edge extraction image includes: acquiring the defect texture features of grayscale images of historical defective devices, establishing a training sample set based on the defect texture features of the grayscale images of historical defective devices; establishing an initial defect evaluation model based on the training sample set and training the initial defect evaluation model to obtain a trained defect evaluation model; acquiring the texture features of the current edge extraction image, inputting the texture features of the current edge extraction image into the trained defect evaluation model to obtain the defect confidence of the corresponding edge extraction image.

[0026] In this embodiment, an initial defect assessment model is established based on a DNN neural network. The DNN neural network has an Encoder-Decoder structure. The initial defect assessment model is trained using the defect texture features of grayscale images of historical defective devices. The trained defect assessment model is then used to assess the defects in the current edge extraction image to be detected, evaluating the defect confidence that the edge extraction image has a defect. The input of the defect assessment model is the texture features of the edge extraction image, and the output is the center point of the bounding box, the regressed width and height of the bounding box, and the defect confidence corresponding to the bounding box. The loss function for training the initial defect assessment model is the mean squared error loss function, and its training process is existing technology and will not be described in detail here.

[0027] S103, extract the environmental interference features of the device grayscale image during the packaging process, determine the quality interference parameters based on the environmental interference features, and determine the packaging quality parameters based on the quality interference parameters and quality defect parameters. In some embodiments of this application, the step of extracting environmental interference features of the device grayscale image during the packaging process and determining quality interference parameters based on the environmental interference features includes: extracting electrostatic spark features and dust particle size features of the device grayscale image during the packaging process, and determining quality interference parameters based on the electrostatic spark features and dust particle size features of the device grayscale image.

[0028] In some embodiments of this application, determining the quality interference parameters based on the electrostatic spark characteristics and dust particle size characteristics of the device grayscale image includes: determining the quality interference parameters according to a quality interference parameter calculation formula, wherein the quality interference parameter calculation formula is as follows: , in, For quality interference parameters, Characterized by electrostatic sparks. Characteristics of dust particle size. To preset the electrostatic spark weight, Preset dust particle size weights.

[0029] In this embodiment, a high-definition camera captures an image of the encapsulation area of ​​the electroluminescent device during the encapsulation process, corresponding to a grayscale image of the device. Electrostatic spark features and dust particle size features are extracted from the captured encapsulation area image using brightness and morphological features. Quality interference parameters are calculated based on these features using preset electrostatic spark weights and preset dust particle size weights. The quality interference parameters and quality defect parameters are then normalized and multiplied to obtain the encapsulation quality parameters.

[0030] S104: Determine the packaging quality of the electroluminescent device based on the packaging quality parameters, and determine whether process control is required based on the packaging quality.

[0031] In some embodiments of this application, determining the packaging quality of the electroluminescent device based on packaging quality parameters includes: obtaining preset standard packaging quality parameters; calculating the difference between the packaging quality parameters and the preset standard packaging quality parameters; determining whether the difference between the packaging quality parameters and the preset standard packaging quality parameters is greater than a second preset threshold; if the difference between the packaging quality parameters and the preset standard packaging quality parameters is greater than the second preset threshold, then the corresponding electroluminescent device is determined to have qualified packaging quality; if the difference between the packaging quality parameters and the preset standard packaging quality parameters is less than or equal to the second preset threshold, then determining whether the difference between the packaging quality parameters and the preset standard packaging quality parameters is greater than a third preset threshold; if the difference between the packaging quality parameters and the preset standard packaging quality parameters is greater than the third preset threshold, then the corresponding electroluminescent device is determined to have unqualified packaging quality.

[0032] In this embodiment, a second preset threshold is set to be greater than a third preset threshold, and the packaging quality is monitored by the difference between the packaging quality parameters and the preset standard packaging quality parameters.

[0033] In some embodiments of this application, the step of determining whether to perform process control based on packaging quality includes: obtaining the number of packaging failures of electroluminescent devices in a preset batch, calculating the ratio of the number of packaging failures of electroluminescent devices in the preset batch to the preset allowable number of failures, and obtaining the packaging failure rate; when the packaging failure rate is less than a fourth preset threshold, triggering a process control instruction to perform packaging process control.

[0034] In this embodiment, the packaging failure rate is used to determine whether process control is needed, and the packaging process parameters are adjusted in a timely manner, which effectively ensures the packaging quality of the electroluminescent device.

[0035] Based on the same technological concept, such as Figure 2 As shown, the present invention also provides a packaging quality monitoring system for electroluminescent devices, comprising: The transmission module acquires images of the packaged electroluminescent device, performs grayscale processing on these images to obtain grayscale images, and then compresses and transmits the grayscale images. The detection module decompresses the compressed data of the grayscale images on the backend server, performs defect detection on the decompressed grayscale images, and obtains quality defect parameters. The environment module extracts environmental interference features from the grayscale images, determines quality interference parameters based on these features, and then determines packaging quality parameters based on the quality interference and quality defect parameters. The control module determines the packaging quality of the electroluminescent device based on the packaging quality parameters and determines whether process control is needed based on the packaging quality.

[0036] By applying the above technical solutions, this invention acquires an image of the packaged electroluminescent device, performs grayscale processing on the image to obtain a grayscale image of the device, and then compresses and transmits the grayscale image. The backend server decompresses the compressed data of the grayscale image, performs defect detection on the decompressed image to obtain quality defect parameters, extracts environmental interference features from the grayscale image, determines quality interference parameters based on these features, and determines packaging quality parameters based on the quality interference parameters and quality defect parameters. The packaging quality of the electroluminescent device is then determined based on these parameters, and process control is assessed based on the packaging quality. This invention effectively improves packaging quality by monitoring the quality of the packaged electroluminescent device.

[0037] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented in hardware or by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0038] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for monitoring the packaging quality of an electroluminescent device, characterized in that, The method includes: The image of the packaged electroluminescent device is obtained, and the image of the electroluminescent device is converted to grayscale to obtain a grayscale image of the device. The grayscale image of the device is then compressed and transmitted. The backend server decompresses the compressed data of the device grayscale image, performs defect detection on the decompressed device grayscale image, and obtains quality defect parameters. Extract environmental interference features of device grayscale images during the packaging process, determine quality interference parameters based on environmental interference features, and determine packaging quality parameters based on quality interference parameters and quality defect parameters. The packaging quality of the electroluminescent device is determined based on the packaging quality parameters, and the need for process control is determined based on the packaging quality.

2. The method for monitoring the packaging quality of electroluminescent devices according to claim 1, characterized in that, The process of compressing and transmitting the grayscale image of the device includes: Obtain a preset segmentation window, and segment the device grayscale image according to the preset segmentation window to obtain several device grayscale image blocks; Obtain the grayscale value of each pixel in the grayscale image block of the device, and cluster each pixel in the grayscale image block of the device according to the grayscale value to obtain k clusters; Based on the clustering results, determine the distance between the corresponding cluster center of each cluster and the other cluster centers, as well as the number of pixel types within each cluster. Then, determine the compression efficiency based on the distance between the corresponding cluster center of each cluster and the other cluster centers, as well as the number of pixel types within each cluster. Inefficient device grayscale image blocks with compression efficiency less than a first preset threshold are selected. The scaling parameters of the preset segmentation window are determined based on the compression efficiency of the inefficient device grayscale image blocks. The preset segmentation window is then scaled according to the scaling parameters. The grayscale image block of the inefficient device is segmented according to the scaled preset segmentation window to obtain several grayscale image blocks of the corrected device. The grayscale image blocks of the devices with a compression efficiency greater than or equal to a first preset threshold and the grayscale sequences of the corrected device grayscale image blocks are statistically analyzed. Huffman coding is performed on the grayscale sequences of each device grayscale image block to obtain the compressed data of the device grayscale image. The compressed data of the device grayscale image is then transmitted to the backend server.

3. The method for monitoring the packaging quality of electroluminescent devices according to claim 2, characterized in that, The step of determining the compression efficiency based on the distance between the corresponding cluster center of each cluster and the other cluster centers, and the number of pixel types within each cluster, includes: The compression efficiency is determined according to the compression efficiency calculation formula, which is as follows: ； in, For compression efficiency, For the first The distance between the cluster center and the other cluster centers corresponds to the cluster cluster. This represents the total number of clusters in the grayscale image block of the device. This represents the number of different pixel types in the grayscale image block of the device. For the first The number of pixel types in a cluster. This is the normalization function.

4. The method for monitoring the packaging quality of electroluminescent devices according to claim 1, characterized in that, The defect detection of the decompressed device grayscale image to obtain quality defect parameters includes: Edge detection is performed on the grayscale image of the device based on the Canny edge detection algorithm to obtain several edge extraction images of the grayscale image of the device; Feature extraction is performed on several edge extraction images of the device grayscale image to extract the texture features of the several edge extraction images; The defect confidence of each edge extraction image is determined based on the texture features of the edge extraction image. The number of edge extraction images with a defect confidence greater than a first preset threshold is extracted. The quality defect parameter is determined based on the number of edge extraction images with a defect confidence greater than the first preset threshold.

5. The method for monitoring the packaging quality of electroluminescent devices according to claim 4, characterized in that, The step of determining the defect confidence of each edge extraction image based on the texture features of the edge extraction image includes: Obtain the defect texture features of grayscale images of historical defective devices, and establish a training sample set based on the defect texture features of grayscale images of historical defective devices; An initial defect assessment model is established based on the training sample set, and the initial defect assessment model is trained to obtain a trained defect assessment model. Obtain the texture features of the current edge extraction image, input the texture features of the current edge extraction image into the trained defect assessment model, and obtain the defect confidence of the corresponding edge extraction image.

6. The method for monitoring the packaging quality of electroluminescent devices according to claim 1, characterized in that, The extraction of environmental interference features from the grayscale image of the device during the packaging process, and the determination of quality interference parameters based on these environmental interference features, include: Extract electrostatic spark features and dust particle size features from the grayscale image of the device during the packaging process, and determine the quality interference parameters based on the electrostatic spark features and dust particle size features of the grayscale image of the device.

7. The method for monitoring the packaging quality of electroluminescent devices according to claim 6, characterized in that, The determination of quality interference parameters based on the electrostatic spark characteristics and dust particle size characteristics of the device grayscale image includes: The quality interference parameters are determined according to the formula for calculating quality interference parameters, which is as follows: ， in, For quality interference parameters, Characterized by electrostatic sparks. Characteristics of dust particle size. To preset the electrostatic spark weight, Preset dust particle size weights.

8. The method for monitoring the packaging quality of electroluminescent devices according to claim 1, characterized in that, Determining the packaging quality of the electroluminescent device based on packaging quality parameters includes: Obtain the preset standard packaging quality parameters and calculate the difference between the packaging quality parameters and the preset standard packaging quality parameters; Determine whether the difference between the packaging quality parameters and the preset standard packaging quality parameters is greater than the second preset threshold. If the difference between the packaging quality parameters and the preset standard packaging quality parameters is greater than the second preset threshold, then the corresponding electroluminescent device packaging quality is deemed to be qualified. If the difference between the packaging quality parameter and the preset standard packaging quality parameter is less than or equal to the second preset threshold, then determine whether the difference between the packaging quality parameter and the preset standard packaging quality parameter is greater than the third preset threshold. If the difference between the packaging quality parameters and the preset standard packaging quality parameters is greater than the third preset threshold, the corresponding electroluminescent device packaging quality is determined to be unqualified.

9. The method for monitoring the packaging quality of electroluminescent devices according to claim 8, characterized in that, The step of determining whether to perform process control based on packaging quality includes: Obtain the number of packaging failures of electroluminescent devices within a preset batch, calculate the ratio of the number of packaging failures to the preset allowable number of failures, and obtain the packaging failure rate. When the packaging defect rate is less than the fourth preset threshold, a process control instruction is triggered to control the packaging process.

10. A packaging quality monitoring system for electroluminescent devices, characterized in that, include: The transmission module is used to acquire images of the packaged electroluminescent device, perform grayscale processing on the electroluminescent device image to obtain a grayscale image of the device, and compress and transmit the grayscale image of the device. The detection module is used by the backend server to decompress the compressed data of the device grayscale image, perform defect detection on the decompressed device grayscale image, and obtain quality defect parameters. The environment module is used to extract environmental interference features from the grayscale image of the device, determine the quality interference parameters based on the environmental interference features, and determine the packaging quality parameters based on the quality interference parameters and quality defect parameters. The control module is used to determine the packaging quality of the electroluminescent device based on the packaging quality parameters, and to determine whether process control is required based on the packaging quality.

Images