An intelligent wafer packaging identification method, system and device

Abstract

The present invention belongs to the technical field of packaging, and relates to an intelligent wafer packaging identification method, system and device. The method of the present invention preprocesses the label image by acquiring the label image; obtains the front view of the label according to the preprocessed label image to make the label information clearer and easier to read; obtains the label text according to the front view of the label, and automatically completes the extraction and recognition of the text information; selects the corresponding packaging program according to the label text and starts the packaging program, which is beneficial to realizing packaging automation, improving production efficiency, and solving the problem that it is easy to make mistakes in manually selecting the packaging program; obtains the packaging picture after packaging, and judges whether the packaging method meets the requirements through the packaging picture, and monitors and detects the packaging quality in real time to ensure that the packaging method meets the requirements. The present invention can accurately identify and read the label image of the wafer, accurately call the packaging program, accurately identify the packaging bag after packaging, and ensure the packaging quality of the wafer.

Description

Technical Field

[0001] The present invention belongs to the technical field of packaging, and relates to an intelligent wafer packaging identification method, system and device. Background Art

[0002] With the increasingly diverse market demands, the packaging requirements for wafer shipments have become richer and more complex. However, at present, in this link, the existing operation methods still mainly rely on human eye recognition to distinguish different packages.

[0003] At the operation level, staff need to rely on personal experience and visual judgment to distinguish various packaging methods, which undoubtedly increases the complexity of operation and the possibility of errors. Due to the lack of effective anti-fooling measures, such as standardized operation procedures, visual assistance tools or automated recognition systems, there is a risk of mixed packaging, affecting product quality.

[0004] At the inspection level, there are also problems that cannot be ignored. At present, the packaged products are not controlled by a system to prevent mistakes, which means that packaging errors are difficult to be discovered and corrected later. Once the products with packaging errors are delivered to customers, it will not only lead to customer dissatisfaction and complaints, but also may trigger a series of subsequent problems such as returns and compensations, bringing losses to the company.

[0005] In summary, the following problems exist in the existing packaging process for shipments:

[0006] Operation level: Rely on personnel to distinguish different packaging methods, without anti-fooling measures, and it is easy to have the problem of mixed packaging;

[0007] Inspection level: There is no system anti-fooling control after packaging, and packaging errors are not easy to be discovered. Summary of the Invention

[0008] The purpose of the present invention is to provide an intelligent wafer packaging identification method, system and device to solve the technical problems of mixed packaging and difficult discovery of packaging errors. The present invention can accurately identify and read the label image of the wafer, accurately call the packaging program, and accurately identify the packaged packaging bag, ensuring the packaging quality of the wafer.

[0009] To achieve the above purpose, the present invention adopts the following technical solutions:

[0010] In the first aspect, the present invention provides an intelligent wafer packaging identification method, including the following steps:

[0011] Obtain a label image and preprocess the label image to obtain a preprocessed label image;

[0012] Obtain a front view of the label according to the preprocessed label image;

[0013] Obtain the label text according to the front view of the label;

[0014] Select the corresponding packaging program according to the label text and start the packaging program;

[0015] Obtain the packaged packaging image and determine whether the packaging method meets the requirements through the packaging image.

[0016] Furthermore, the obtaining of the label image and the preprocessing of the label image are as follows:

[0017] Obtain the label image;

[0018] Convert the color space of the label image to the LAB color space and separate the L channel;

[0019] Perform contrast-limited adaptive histogram equalization on the L channel;

[0020] Merge the L channel that has completed contrast-limited adaptive histogram equalization with the AB channel;

[0021] Convert the merged LAB color space to the RGB color space to obtain an RGB color space image;

[0022] Apply the median filtering method to denoise the RGB color space image and eliminate the interference of the complex background.

[0023] Furthermore, the performing of contrast-limited adaptive histogram equalization on the L channel is as follows:

[0024] Use contrast-limited adaptive histogram equalization to divide the image of the L channel into blocks of n x n pixels;

[0025] Perform histogram equalization on each block separately with a contrast of clipLimit = 4;

[0026] Use the interpolation method to smoothly connect the blocks after histogram equalization.

[0027] Furthermore, the obtaining of the front view of the label according to the preprocessed label image is as follows:

[0028] Obtain the four vertex coordinates of the label according to the preprocessed label image;

[0029] Obtain the front view of the label according to the preprocessed label image and the four vertex coordinates of the label.

[0030] Furthermore, the obtaining of the four vertex coordinates of the label according to the preprocessed label image is as follows:

[0031] Use the Canny algorithm to detect all edge information of the preprocessed label image. Set the Sobel operator size to 7 to calculate the image gradient, obtain the edge image, and perform binarization processing on the edge image to obtain a binary image;

[0032] Use the contour extraction algorithm to identify and extract all closed contour regions from the binary image, calculate the number of pixels covered by each closed contour region, obtain the top six closed contour regions with the largest number of pixels and sort them in descending order. For the six sorted closed contour regions, use the polygon approximation method to reduce the number of vertices of the contour and obtain the approximated polygon;

[0033] Obtain the four vertex coordinates of the label according to the approximated polygon.

[0034] Further, the front view of the label is obtained according to the preprocessed label image and the four vertex coordinates of the label, specifically as follows:

[0035] Calculate the perspective transformation matrix based on the four vertex coordinates of the label;

[0036] Use the perspective transformation matrix to perform spatial mapping on the preprocessed label image to correct the distortion and tilt of the preprocessed label image, and obtain the front view of the label.

[0037] Further, the label text is obtained according to the front view of the label, specifically as follows:

[0038] Collect standard label images containing a preset font, annotate the preset font on the standard label images, and construct a label text dataset according to the annotated standard label images;

[0039] Use the DB algorithm to obtain the text region in the front view of the label to be detected;

[0040] Use the CRNN algorithm combined with the label text dataset to perform text recognition on the text region and obtain the recognition result;

[0041] Process the skewed text in the recognition result through a direction classifier to obtain the label text.

[0042] Further, the packaged packaging picture is obtained, and it is judged whether the packaging method meets the requirements through the packaging picture, specifically as follows:

[0043] Obtain the packaged packaging picture, and use the small area sampling method to obtain the pixels of the packaging picture;

[0044] Split the red, green, and blue wavelength bands of the pixels of the packaging picture according to the image channel dimension to form grayscale images of single wavelength bands;

[0045] The grayscale image of a single optical band is combined by arranging and combining the bands to obtain a combined grayscale image;

[0046] The combined grayscale image is subjected to image fusion to form a new packaging image;

[0047] Based on the pixel attributes, the pixels in the detection area of the new packaging image are judged in multiple dimensions, and all packaging pictures are traversed for scoring. According to the results of the multi-dimensional judgment and the scoring results, it is judged whether the packaging type meets the requirements.

[0048] In a second aspect, the present invention provides an intelligent wafer packaging identification system, including:

[0049] A preprocessing module: used to obtain a label image and preprocess the label image to obtain a preprocessed label image;

[0050] A front view acquisition module: used to obtain the front view of the label according to the preprocessed label image;

[0051] A label text acquisition module: used to obtain the label text according to the front view of the label;

[0052] A packaging program startup module: used to select a corresponding packaging program according to the label text and start the packaging program;

[0053] A packaging judgment module: used to obtain the packaged packaging picture and judge whether the packaging method meets the requirements through the packaging picture.

[0054] In a third aspect, the present invention provides an electronic device, including: a processor; a memory for storing computer program instructions; and when the computer program is executed, it realizes the steps of the intelligent wafer packaging identification method.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] 1. By obtaining the label image and preprocessing the label image, the method of the present invention reduces the noise and interference in the image, improves the accuracy of subsequent processing, and obtains a preprocessed label image; obtaining the front view of the label according to the preprocessed label image makes the label information clearer and easier to read; obtaining the label text according to the front view of the label automatically completes the extraction and recognition of text information; selecting a corresponding packaging program according to the label text and starting the packaging program is conducive to realizing packaging automation, improving production efficiency, and solving the problem that it is easy to make mistakes in manually selecting the packaging program; obtaining the packaged packaging picture and judging whether the packaging method meets the requirements through the packaging picture, and monitoring and detecting the packaging quality in real time to ensure that the packaging method meets the requirements. The present invention can accurately identify and read the label image of the wafer, accurately call the packaging program, and accurately identify the packaged packaging bag, ensuring the packaging quality of the wafer.

[0057] 2. The system of the present invention includes: a preprocessing module, a front view acquisition module, a label text acquisition module, a packaging program startup module, and a packaging judgment module. The preprocessing module is used to acquire a label image and preprocess the label image to obtain a preprocessed label image; the front view acquisition module is used to acquire the front view of the label according to the preprocessed label image; the label text acquisition module is used to acquire the label text according to the front view of the label; the packaging program startup module is used to select a corresponding packaging program according to the label text and start the packaging program; the packaging judgment module is used to acquire the packaged packaging picture and judge whether the packaging method meets the requirements through the packaging picture. Each module cooperates with each other, can accurately identify and read the label image of the wafer, accurately call the packaging program, can accurately identify the packaged packaging bag, and ensure the packaging quality of the wafer.

[0058] 3. The device of the present invention can also accurately identify and read the label image of the wafer, accurately call the packaging program, can accurately identify the packaged packaging bag, and ensure the packaging quality of the wafer. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 is the flowchart of the method of the present invention;

[0060] Figure 2 is the system module diagram of the present invention;

[0061] Figure 3 is the flowchart of the method of the embodiment of the present invention;

[0062] Figure 4 is the effect diagram of acquiring the label text according to the front view of the label in the embodiment of the present invention. Among them, a is the front view of the label, and b is the label text acquired according to the front view of the label;

[0063] Figure 5 is the RGB pixel value distribution diagram of the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0064] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0065] It should be noted that the terms "first", "second", etc. in the description of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence. It should be understood that the data used in this way can be interchanged under appropriate circumstances, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0066] The present invention will be further described in detail below with reference to the accompanying drawings:

[0067] See Figure 1 , the present invention discloses an intelligent wafer packaging identification method, which includes the following steps:

[0068] S1, obtain a label image and preprocess the label image to obtain a preprocessed label image, specifically as follows:

[0069] Obtain the label image;

[0070] Convert the color space of the label image to the LAB color space and separate the L channel;

[0071] Perform contrast-limited adaptive histogram equalization processing on the L channel, specifically as follows:

[0072] Use contrast-limited adaptive histogram equalization to divide the image of the L channel into blocks of n x n pixels. Preferably, n is taken as 16 here;

[0073] Perform histogram equalization on each block separately with a contrast of clipLimit = 4;

[0074] Use an interpolation method to smoothly connect the blocks after histogram equalization;

[0075] Merge the L channel that has completed contrast-limited adaptive histogram equalization processing with the AB channel;

[0076] Convert the merged LAB color space to the RGB color space to obtain an RGB color space image;

[0077] Apply the median filtering method to denoise the RGB color space image to eliminate the interference of the complex background and improve the accuracy of inner label text recognition. See Figure 5 for the RGB pixel value distribution diagram.

[0078] It should be noted that the LAB color space is a non-linear color space based on human visual perception.

[0079] L: Represents lightness, with a value range from 0 to 100, where 0 represents pure black and 100 represents pure white.

[0080] A: Represents the color component from green to red, with a value range from -128 to 127, where negative values tend towards green and positive values tend towards red.

[0081] B: Represents the color component from blue to yellow, with a value range from -128 to 127, where negative values tend towards blue and positive values tend towards yellow.

[0082] The RGB color space is a color model constructed based on the three primary colors of red, green, and blue. The RGB color space is based on these three primary colors. By adjusting the brightness and superposition ratio of these three colors, a rich and extensive range of colors can be produced. There are infinitely many different colors in nature, while the human eye can only distinguish a limited number of different colors. The RGB mode can represent more than sixteen million different colors. In the view of the human eye, it is very close to the colors of nature, so it is also called the natural color mode.

[0083] clipLimit is the threshold used to limit the degree of contrast enhancement in the contrast-limited adaptive histogram equalization algorithm.

[0084] S2. Obtain the front view of the label according to the preprocessed label image, specifically as follows:

[0085] Obtain the four vertex coordinates of the label according to the preprocessed label image, specifically as follows:

[0086] Use the Canny algorithm to detect all edge information of the preprocessed label image. Set the Sobel operator size to 7 to calculate the image gradient, obtain the edge image, and perform binary processing on the edge image to obtain the binary image;

[0087] Use the contour extraction algorithm to identify and extract all closed contour regions from the binary image. Calculate the number of pixels covered by each closed contour region. Obtain the top six closed contour regions with the largest number of pixels and sort them in descending order. For the six sorted closed contour regions, use the polygon approximation method to reduce the number of vertices of the contour and obtain the approximated polygon;

[0088] Obtain the four vertex coordinates of the label according to the approximated polygon;

[0089] Obtain the front view of the label according to the preprocessed label image and the four vertex coordinates of the label, specifically as follows:

[0090] Calculate the perspective transformation matrix based on the four vertex coordinates of the label;

[0091] Use the perspective transformation matrix to perform spatial mapping on the preprocessed label image to correct the distortion and tilt of the preprocessed label image, and obtain the front view of the label.

[0092] It should be noted that the Canny algorithm is a multi-level edge detection algorithm. The Sobel operator is a discrete differential operator for edge detection, which combines the first-order gradient calculation of image brightness to identify edges in the image.

[0093] S3. Obtain the label text according to the front view of the label, specifically as follows:

[0094] Collect standard label images containing the preset font, label the preset font on the standard label images, and construct a label text dataset according to the labeled standard label images;

[0095] Use the DB algorithm to obtain the text region in the front view of the label to be detected;

[0096] Use the CRNN algorithm to combine the label text dataset to perform text recognition on the text region, and obtain the recognition result;

[0097] Process the inclined text in the recognition result through a direction classifier to obtain the label text, see Figure 4 .

[0098] It should be noted that the DB algorithm refers to the Differentiable Binarization algorithm, which is a scene text detection algorithm based on segmentation. The purpose of this algorithm is to convert the probability map generated by the segmentation method into a bounding box and a text region, which will include a post-processing process of binarization.

[0099] The CRNN algorithm, full name Convolutional Recurrent Neural Network, is a deep learning model, especially suitable for dealing with sequence recognition problems based on images, especially scene text recognition.

[0100] S4. Select the corresponding packaging program according to the label text, and start the packaging program;

[0101] S5. Obtain the packaged packaging picture, and judge whether the packaging method meets the requirements through the packaging picture, specifically as follows:

[0102] Obtain the packaged packaging picture, and use the small area sampling method to obtain the pixels of the packaging picture;

[0103] Split the red, green, and blue bands of the pixels in the packaging image along the image channel dimension to form grayscale images of single-light bands;

[0104] Combine the grayscale images of single-light bands in a permutation and combination manner to obtain a combined grayscale image;

[0105] Perform image fusion on the combined grayscale image to form a new packaging image;

[0106] Make multi-dimensional judgments on the pixels in the detection area of the new packaging image based on pixel attributes, and at the same time traverse all packaging images for scoring. Determine whether the packaging type meets the requirements based on the multi-dimensional judgment results and scoring results.

[0107] See Figure 1 , in another feasible embodiment of the present invention, the following is adaptively modified according to the situation. The steps include:

[0108] Obtain a label image and preprocess the label image to reduce noise and interference in the image and improve the accuracy of subsequent processing, obtaining a preprocessed label image;

[0109] Obtain the front view of the label based on the preprocessed label image to eliminate image distortion and distortion caused by different shooting angles and make the label information clearer and easier to read;

[0110] Obtain the label text based on the front view of the label to automatically complete the extraction and recognition of text information, greatly improving the processing efficiency and accuracy;

[0111] Select the corresponding packaging program according to the label text and start the packaging program, which is conducive to realizing packaging automation, improving production efficiency, and solving the problem that it is easy to make mistakes when manually selecting the packaging program;

[0112] Obtain the packaged packaging image, and judge whether the packaging method meets the requirements through the packaging image, and perform real-time monitoring and detection of the packaging quality to ensure that the packaging method meets the requirements. Once a packaging problem is found, the problem cause can be traced and located immediately, and corresponding corrective measures can be taken to avoid the spread and impact of quality problems. The present invention can accurately identify and read the label image of the wafer, accurately call the packaging program, and accurately identify the packaged packaging bag to ensure the packaging quality of the wafer.

[0113] Embodiment 1:

[0114] See Figure 3 , this embodiment discloses an intelligent wafer packaging identification method. By combining an automatic packaging machine, the present invention can call different packaging methods to perform packaging operations and automatically check the packaging results to ensure the correctness of the packaging method.

[0115] The method of the present invention is as follows:

[0116] S1. Use the intelligent identification inner packaging label of the present invention to extract relevant information such as customer name, product model, batch number, etc.; the method of the present invention can capture the images of these inner packaging labels and use image recognition algorithms to accurately read and extract the label texts.

[0117] S2. Feed the label text back to the machine system, and the machine system will automatically call the packaging program corresponding to the customer to perform packaging;

[0118] S3. After packaging, identify the packaging method through images, including checking whether the appearance, size, sealing method, etc. of the packaging meet the expectations, ensuring the packaging quality, and promptly discovering and correcting any possible packaging errors;

[0119] S4. The system compares whether the packaging method meets the requirements and presents the results to the operators for subsequent operations, so that they can promptly understand the situation of the packaging operation. If any non-compliant packaging is found, the operators can immediately take measures to correct it, thus avoiding the outflow of defective products.

[0120] Embodiment 2:

[0121] See Figure 1 , this embodiment discloses an intelligent wafer packaging identification method, including the following steps:

[0122] S11. Image preprocessing:

[0123] Aiming at problems such as uneven illumination and weak contrast in the label image shooting, convert the image color space to LAB and separate the L channel. Use a block size of 16x16 pixels and a contrast amplification limit of clipLimit = 4 to perform contrast-limited adaptive histogram equalization processing on the L luminance image. Finally, merge the L, A, and B channels and convert them to the RGB color space to improve the local contrast of the image. At the same time, there will be noise in the label image, and median filtering is applied to eliminate the interference of the complex background to improve the accuracy of the inner label text recognition. See Figure 5 for the RGB pixel value distribution diagram.

[0124] S12. Label detection:

[0125] First, use Canny to detect all the edges of the label grayscale image. Aiming at the problem that there is partial exposure in the label image, resulting in some label contours unable to be detected, set the size of the Sobel operator to 7 to calculate the image gradient and accurately obtain the label contour;

[0126] Secondly, an outline extraction algorithm is used to obtain all the outlines of the binary image, calculate the number of pixels covered by each extracted outline (i.e., the outline area), and sort the top six outlines with the largest outline areas in descending order to locate the outline of the binary image with the largest area.

[0127] Based on the top six outlines with the largest areas, within the maximum distance of 2% of the original outline perimeter, the curve is approximated by straight line segments to reduce the number of vertices of the outline, and it is judged whether the approximated polygon has exactly four vertices, and finally the edge of the wafer label picture is obtained.

[0128] S13. Label space mapping:

[0129] The four corner point coordinates of the positioning label have been obtained in S12. Calculate its perspective transformation matrix to ensure that the label maintains the correct proportion and positional relationship after transformation. The perspective transformation matrix is used to perform space mapping on the preprocessed label image to correct the distortion and tilt of the label image, so as to obtain the front view of the shipping label.

[0130] S14. Label text detection and recognition:

[0131] For the preset font of the in-factory label, a label text data set is constructed and labeled. The label text data set is used for training in combination with the CRNN algorithm to facilitate the accurate recognition of text. The DB (Differentiable Binarization) algorithm is used for text detection. By setting double thresholds to control the label text detection, in the probability map output by DB, the pixel points with scores higher than the preset threshold 1 are regarded as text pixel points; when the average score of all text pixel points within the detection result border exceeds the preset threshold 2, it is considered as the text area.

[0132] The CRNN algorithm, that is, the convolutional recurrent neural network algorithm, is used to perform text recognition on the above text area. A direction classifier is introduced to process the tilted text in the recognition result. When the prediction result points to a 180-degree rotation and the score exceeds the preset threshold, an automatic image flipping operation is performed to improve the text recognition accuracy. See Figure 4 To obtain the effect picture of the label text according to the front view of the label.

[0133] S15. Label post-processing:

[0134] In the case where the wafer label cannot be recognized or is recognized inaccurately, manual recheck is performed. The label text recognition information is timely fed back to the database.

[0135] S2. Realization of packaging type detection;

[0136] Fine detection area: To avoid the influence of the background of the in-plant shipping packaging environment and cause misidentification of the packaging type, the present invention determines the area to be inspected for the packaging based on statistical probability for packaging type identification. The method of small-area sampling is used to extract the pixels of the packaging picture of the packaging to be inspected, and multiple small areas are judged distributively to avoid the influence of relatively concentrated sampling on the judgment result; on the one hand, it provides a reliable basis for the source of detection samples for the classification and detection task of the shipping packaging; on the other hand, it can also avoid algorithm redundancy, improve the detection speed in the production process, and efficiently boost the in-plant production and shipping efficiency.

[0137] The present invention makes multi-dimensional judgments on the pixels in the detection area, comprehensively considers the attributes of all pixels, combines the properties of the shipping packaging materials, and realizes the accurate judgment of the packaging material type based on the multiple correlations of the pixel values.

[0138] Since the original pictures are difficult to distinguish the packaging type classification, the classification accuracy is low, and they do not have the classification characteristics of packaging classification. To solve this problem, based on the color gamut of the pixels of the packaging pictures, the method of band combination is used to seek new rules and find a new method for packaging type images that are easy to classify. This method includes splitting the red light band (R), green light band (G), and blue light band (B) of the pixels of the packaging picture according to the image channel (C) dimension. After splitting, single-light band R, G, and B grayscale images are formed, and the grayscale images are used according to for band combination, and single-band combined grayscale images of R+G, G+B, and B+R are generated. Finally, the single-light band combined grayscale images are image-fused to form new packaging images of R+G, G+B, and B+R. The generation of the new packaging images can show obvious color distinctions, which is beneficial to the accurate classification of packaging images.

[0139] The information of the new packaging images is beneficial to the distinction between packaging bag 2 and packaging bag 1, and clarifies the classification boundary of the two types of packaging. See Figure 5 , which is the RGB pixel value distribution diagram of packaging bag 1 and packaging bag 2.

[0140] Make multi-dimensional judgments on the pixels in the detection area of the new packaging images obtained by band combination, comprehensively consider the attributes of all pixels, combine the properties of the shipping packaging materials, and realize the accurate judgment of the packaging material type based on the multiple correlations of the pixel values. At the same time, all packaging pictures are traversed for scoring, and the packaging type is judged according to the scoring results, effectively improving the accuracy of the outer packaging classification results

[0141] Among them, traversing all packaging pictures for scoring is specifically as follows: Judge all the outer packaging pictures in each Lot, and determine the final outer packaging type based on the overall judgment to avoid misjudgment caused by the judgment of a single picture.

[0142] The present invention can achieve intelligent classification of packaging pictures and accurate positioning of each detection item; it can achieve accurate detection of inner packaging labels and extract accurate text information; it can eliminate the background interference of FOSB and perform multi-level parallel target detection; it can determine the type of wafer packaging by using the multiple correlations of pixel values; it can combine multi-source heterogeneous databases and images to compare the detection results with customer requirements.

[0143] In the packaging process of the present invention, there is no need to rely on an online system to call the packaging program. The product requirements can be distinguished offline by this method, and the corresponding packaging program can be executed.

[0144] Increase intelligent packaging operations and inspections, avoid packaging mix-ups and inspection errors, and improve packaging quality.

[0145] The method of the present invention can be an intelligent anti-fooling packaging operation method, which is of great significance.

[0146] Based on the above method, the present invention discloses an intelligent wafer packaging identification system, see Figure 2 , including:

[0147] Preprocessing module: used to obtain the label image and preprocess the label image to obtain the preprocessed label image.

[0148] Front view acquisition module: used to obtain the front view of the label according to the preprocessed label image.

[0149] Label text acquisition module: used to obtain the label text according to the front view of the label.

[0150] Packaging program startup module: used to select the corresponding packaging program according to the label text and start the packaging program.

[0151] Packaging judgment module: used to obtain the packaged packaging picture and judge whether the packaging method meets the requirements through the packaging picture.

[0152] Each module of the system of the present invention cooperates with each other, can accurately identify and read the label image of the wafer, accurately call the packaging program, and can accurately identify the packaged packaging bag to ensure the packaging quality of the wafer.

[0153] An electronic device includes: a processor; a memory for storing computer program instructions; and when the computer program is executed, it realizes the steps of the intelligent wafer packaging identification method.

[0154] A storage medium stores computer program instructions. When the computer program instructions are loaded and run by a processor, the processor executes the intelligent wafer packaging identification method.

[0155] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0156] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, as well as the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0157] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that realize the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0158] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0159] The above content is only to illustrate the technical idea of the present invention and cannot be used to limit the protection scope of the present invention. Any modification made on the basis of the technical solution according to the technical idea proposed by the present invention falls within the protection scope of the present invention.

Claims

1. An intelligent wafer packaging identification method, characterized in that: The following steps are involved: Acquire a label image and preprocess the label image to obtain a preprocessed label image; Obtain a label front view according to the preprocessed label image; Get the label text according to the label front view; Select the corresponding packaging program according to the label text and start the packaging program; Get the packaging picture after packaging, and judge whether the packaging method meets the requirements through the packaging picture, as follows: Obtain the packaged image after packaging, and use a small area sampling method to obtain pixels of the package image; The red light band, green light band and blue light band of the pixels of the packaging image are split according to the image channel dimension to form a grayscale image of a single light band; The grayscale images of the single light bands are combined in a permutation and combination manner to obtain a combined grayscale image; Perform image fusion on the grayscale image of the combined image to form a new packaging image; The pixels in the detection area of ​​the new packaging image are judged in multiple dimensions according to the pixel attributes, and all packaging images are traversed for scoring. Whether the packaging type meets the requirements is determined based on the multi-dimensional judgment results and the scoring results.

2. The intelligent wafer packaging identification method according to claim 1, characterized in that: The label image is obtained and preprocessed as follows: Get label image; Convert the color space of the label image to LAB color space and separate the L channel; Perform contrast-limited adaptive histogram equalization on the L channel; Merge the L channel and the AB channel after the contrast-limited adaptive histogram equalization processing; Convert the merged LAB color space to RGB color space to obtain an RGB color space image; The median filtering method is applied to denoise the RGB color space image and eliminate the interference of complex background.

3. The intelligent wafer packaging identification method according to claim 2, characterized in that: The contrast-limited adaptive histogram equalization process is performed on the L channel, specifically as follows: The image of L channel is divided into blocks of nxn pixels using contrast limited adaptive histogram equalization; Use clipLimit=4 contrast to perform histogram equalization on each block individually; Interpolation methods are used to smoothly connect the histogram equalized blocks.

4. The intelligent wafer packaging identification method according to claim 1, characterized in that: The label front view is obtained according to the preprocessed label image, specifically as follows: Obtain the coordinates of the four vertices of the label according to the preprocessed label image; The front view of the label is obtained according to the preprocessed label image and the coordinates of the four vertices of the label.

5. The intelligent wafer packaging identification method according to claim 4, characterized in that: The coordinates of the four vertices of the label are obtained according to the preprocessed label image, as follows: The Canny algorithm is used to detect all edge information of the preprocessed label image, and the Sobel operator size is set to 7 to calculate the image gradient to obtain the edge image. The edge image is binarized to obtain a binary image. A contour extraction algorithm is used to identify and extract all closed contour areas from the binary image, the number of pixels covered by each closed contour area is calculated, the first six closed contour areas with the largest number of pixels are obtained and sorted in descending order, and the polygon approximation method is used to reduce the number of contour vertices for the sorted six closed contour areas to obtain the approximated polygon; Get the coordinates of the four vertices of the label based on the approximated polygon.

6. The intelligent wafer packaging identification method according to claim 4, characterized in that: The front view of the label is obtained according to the preprocessed label image and the coordinates of the four vertices of the label, as follows: Calculate the perspective transformation matrix based on the four vertex coordinates of the label; The preprocessed label image is spatially mapped using a perspective transformation matrix to correct the distortion and tilt of the preprocessed label image, thereby obtaining a front view of the label.

7. The intelligent wafer packaging identification method according to claim 1, characterized in that: The label text is obtained according to the label front view, as follows: Collect standard label images containing preset fonts, annotate the preset fonts on the standard label images, and construct a label text dataset based on the annotated standard label images; Use DB algorithm to obtain the text area in the front view of the label to be detected; The CRNN algorithm is used in combination with the labeled text data set to perform text recognition on the text area and obtain the recognition result; The tilted text in the recognition result is processed by the direction classifier to obtain the label text.

8. An intelligent wafer packaging identification system, characterized in that: include: Preprocessing module: used to obtain the label image and preprocess the label image to obtain the preprocessed label image; Front view acquisition module: used to acquire the label front view according to the preprocessed label image; Label text acquisition module: used to obtain label text according to the label front view; Packaging program startup module: used to select the corresponding packaging program according to the label text and start the packaging program; Packaging judgment module: used to obtain the packaging picture after packaging, and judge whether the packaging method meets the requirements through the packaging picture, as follows: Obtain the packaged image after packaging, and use a small area sampling method to obtain pixels of the package image; The red light band, green light band and blue light band of the pixels of the packaging image are split according to the image channel dimension to form a grayscale image of a single light band; The grayscale images of the single light bands are combined in a permutation and combination manner to obtain a combined grayscale image; Perform image fusion on the grayscale image of the combined image to form a new packaging image; The pixels in the detection area of ​​the new packaging image are judged in multiple dimensions according to the pixel attributes, and all packaging images are traversed for scoring. Whether the packaging type meets the requirements is determined based on the multi-dimensional judgment results and the scoring results.

9. An electronic device, comprising: Processor; memory, electronic device used to store computer program instructions; characterized in that it is used to implement the steps of the intelligent wafer packaging identification method as described in any one of claims 1-7 when executing the computer program.

Images