Intelligent pre-printing plate correction method, system and equipment and medium

Abstract

The invention discloses an intelligent pre-press version correction method and system, computer equipment and a storage medium, and the method comprises the steps: carrying out the processing and analysis of a preset product standard sample and a product real object image based on machine vision and a convolutional neural network, and carrying out the detection through a double detection mode of first-layer verification and second-layer verification, the image is subjected to comparison verification, matching, feature extraction and anti-fake key point detection on the pixel level, the possible plate correction problem can be automatically detected and corrected, plate correction before printing becomes more standardized and processized, and the yield of anti-fake products can be effectively improved.

Description

Technical Field

[0001] This invention relates to the fields of machine vision and anti-counterfeiting technology, and in particular to an intelligent prepress proofing method, system, computer equipment, and storage medium based on machine vision and convolutional neural networks. Background Technology

[0002] In the anti-counterfeiting industry, the yield rate of anti-counterfeiting products is crucial. Given the characteristics of anti-counterfeiting products, even minor errors during production can lead to discrepancies between the product and the printed image, resulting in customer complaints and even returns. Therefore, careful proofreading of pre-production anti-counterfeiting products before printing is essential. Currently, traditional manual proofreading methods are not standardized or rigorous enough, often resulting in problems such as careless proofreading, deviations due to subjective human factors, omissions of key anti-counterfeiting points, color differences in printing, and incorrect code linking, leading to a low yield rate for anti-counterfeiting products. Summary of the Invention

[0003] To overcome the shortcomings of the prior art, embodiments of the present invention provide an intelligent prepress proofing method, system, computer equipment, and storage medium based on machine vision and convolutional neural networks.

[0004] A smart prepress proofing method is applied in a smart prepress proofing system consisting of at least three clients and at least one server.

[0005] In the first client, a preset code value link format and a preset anti-counterfeiting technology type are set, and the preset code value link format and the preset anti-counterfeiting technology type are uploaded to the server.

[0006] In the second client, the preset product standard sample is uploaded to the server;

[0007] On the server side, the first layer of verification is performed on the preset product standard sample according to the preset code value link format and the preset anti-counterfeiting technology type; if the verification fails, the first feedback result is pushed to the second client; if the verification succeeds, the server waits to receive the product image corresponding to the preset product standard sample sent by the third client.

[0008] In the third client, the physical image of the product corresponding to the preset product standard sample is uploaded to the server;

[0009] On the server side, the physical product image is compared with a preset product standard sample for a second layer of verification; if the verification fails, the second feedback result is pushed to the third client; if the verification succeeds, the production instruction is sent to the third client.

[0010] An intelligent prepress proofing system includes a first client, a second client, a third client, and a server; the server communicates with the first client, the second client, and the third client via wired or wireless means respectively.

[0011] The first client is used to set the preset code value link format and preset anti-counterfeiting technology type, and upload the preset code value link format and preset anti-counterfeiting technology type to the server.

[0012] The second client is used to upload preset product standard samples to the server.

[0013] The third client is used to upload the physical product image corresponding to the preset product standard sample to the server.

[0014] The server is used to perform the first-level verification of the preset product standard sample according to the preset code value link format and the preset anti-counterfeiting technology type; if the verification fails, the first feedback result is pushed to the second client; if the verification succeeds, it waits to receive the product image corresponding to the preset product standard sample sent by the third client.

[0015] It also includes a second-level verification method for comparing the actual product image with a preset product standard sample; if the verification fails, the second feedback result is pushed to the third client; if the verification succeeds, the production instruction is sent to the third client.

[0016] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-described intelligent prepress proofing method.

[0017] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described intelligent prepress proofing method.

[0018] The aforementioned intelligent pre-press proofing method, system, computer equipment, and storage medium, based on machine vision and convolutional neural networks, process and analyze preset product standard samples and product images. Through a dual detection method of first-layer verification and second-layer verification, the images are compared, matched, feature extracted, and anti-counterfeiting key point detected at the pixel level. It can automatically detect and correct possible proofing problems, making pre-press proofing more standardized and streamlined, and effectively improving the yield of anti-counterfeiting products. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart of an intelligent prepress proofing method in one embodiment of the present invention;

[0021] Figure 2 This is a framework diagram of an intelligent prepress proofing system according to an embodiment of the present invention;

[0022] Figure 3 This is a flowchart of an intelligent prepress proofing system according to an embodiment of the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] In one embodiment, such as Figure 1 As shown, an intelligent prepress proofing method is provided, which is applied in an intelligent prepress proofing system consisting of at least three clients and at least one server.

[0025] In the first client, a preset code value link format and a preset anti-counterfeiting technology type are set, and the preset code value link format and the preset anti-counterfeiting technology type are uploaded to the server.

[0026] In the second client, the preset product standard sample is uploaded to the server;

[0027] On the server side, the first layer of verification is performed on the preset product standard sample according to the preset code value link format and the preset anti-counterfeiting technology type; if the verification fails, the first feedback result is pushed to the second client; if the verification succeeds, the server waits to receive the product image corresponding to the preset product standard sample sent by the third client.

[0028] In the third client, the physical image of the product corresponding to the preset product standard sample is uploaded to the server;

[0029] On the server side, the physical product image is compared with a preset product standard sample for a second layer of verification; if the verification fails, the second feedback result is pushed to the third client; if the verification succeeds, the production instruction is sent to the third client.

[0030] Specifically, the first client is used by business personnel to set the code value link format and anti-counterfeiting technology type for the anti-counterfeiting product when placing an order. Examples include micro-hidden codes, QR codes, barcodes, serial numbers, and code-within-a-code technology. The code value link format is obtained through communication between the business personnel and the customer, while the types and number of anti-counterfeiting technologies used can be flexibly determined.

[0031] The second client is used by designers. Designers upload preset product standard samples to the server, whereby the preset product standard samples (image files) are the design drafts of the anti-counterfeiting products. Furthermore, the second client can receive the first feedback results pushed by the server in real time.

[0032] The third client is used by printing production staff. It's primarily for pre-press proofreading. After selecting a pre-printed product item, staff scan and upload an image of the sample (the actual product corresponding to the preset standard sample) using the camera on their handheld device. The third client can also receive real-time feedback from the server. Upon receiving correct proofreading results, printing production staff can proceed with production; if incorrect results are received, they communicate with the designers to correct the errors.

[0033] On the server side, upon receiving the product standard sample uploaded by the designer, the design draft is analyzed and verified based on the code link format and anti-counterfeiting technology type set by the business personnel when placing the order. The design draft verification process is as follows:

[0034] 1) Preprocess the design draft image, including but not limited to adjusting the image size, grayscale conversion, noise reduction and normalization, so that the design draft image can be analyzed at the pixel level.

[0035] 2) Decode any barcodes or QR codes that may be present in the design draft, and verify the decoded results against the code values ​​preset by the business personnel. Output the verification result. This output verification result is the first feedback result and is sent to the second client.

[0036] 3) Extract image features using a Convolutional Neural Network (CNN). Specifically, local features in the image are learned through convolution and pooling operations, and the image information is compressed into a set of feature vectors.

[0037] 4) Analyze the feature vectors obtained in step 3) and perform feature matching with the conditions preset by the business personnel, such as distance-based matching or similarity-based matching. If there are errors in the analysis results, the error identifier and the reason for the error are encapsulated into a second feedback result, and the second feedback result is sent to the operation interface of the third client designer.

[0038] Meanwhile, on the server side, after receiving the scanned image A (i.e., the actual product image) of the machine setup sample uploaded by the printing production personnel, it compares and verifies it with the product standard sample image B (i.e., the preset product standard sample) uploaded and verified by the designer. The specific process is as follows:

[0039] 1) Perform preprocessing on images A and B, including resizing, grayscale conversion, noise reduction, and normalization, so that the two images can be compared at the pixel level.

[0040] 2) Use a convolutional neural network to extract features from both, and use the extracted feature vectors for subsequent comparison operations.

[0041] 3) Perform matching analysis on the feature vectors of the two images extracted in 2), and perform similarity matching to conduct multiple key verifications on the anti-counterfeiting technology.

[0042] 4) The SURF algorithm is used to detect and describe feature points. Key points in two images are matched to evaluate similarity and calculate the Euclidean distance between their descriptors. If the minimum distance between the corresponding descriptors of the two images is less than a preset empirical threshold, they are considered to be matched; otherwise, the key point positions of the matching identification are marked by drawing lines or other methods.

[0043] 5) Loop through each corresponding anti-counterfeiting key point, mark the error type according to the feature matching result, and feed the error information back to the printing production personnel's operation interface along with the marking information.

[0044] It's important to note that convolution is one of the most common operations in CNNs. Convolution extracts features by sliding a convolutional kernel (also called a filter) across the image. The convolutional kernel is a small matrix that learns to extract different features, such as edges, corners, etc. At each location, the convolutional kernel is multiplied by a small region of the image and summed; this result becomes an element in a new feature map. In this way, the convolutional operation extracts a specific set of features from the image and transforms them into a new feature map.

[0045] Pooling operation: Pooling is an operation performed after convolution. Its purpose is to reduce the size of the feature map and avoid overfitting. There are generally two types of pooling operations: max pooling and average pooling. Max pooling selects the largest feature value in each region as the pooled value, while average pooling averages the feature values ​​within the region. This process can be seen as a dimensionality reduction operation because it reduces the size of the feature map, thereby reducing the amount of computation and the number of parameters. In practical applications, pooling layers are often added after convolutional layers to better extract and learn image features.

[0046] In one embodiment, such as Figure 2 As shown, an intelligent prepress proofing system is provided, including a first client, a second client, a third client, and a server; the server communicates with the first client, the second client, and the third client via wired or wireless means respectively.

[0047] The first client is used to set the preset code value link format and preset anti-counterfeiting technology type, and upload the preset code value link format and preset anti-counterfeiting technology type to the server.

[0048] The second client is used to upload preset product standard samples to the server.

[0049] The third client is used to upload the physical product image corresponding to the preset product standard sample to the server.

[0050] The server is used to perform the first-level verification of the preset product standard sample according to the preset code value link format and the preset anti-counterfeiting technology type; if the verification fails, the first feedback result is pushed to the second client; if the verification succeeds, it waits to receive the product image corresponding to the preset product standard sample sent by the third client.

[0051] It also includes a second-level verification method for comparing the actual product image with a preset product standard sample; if the verification fails, the second feedback result is pushed to the third client; if the verification succeeds, the production instruction is sent to the third client.

[0052] The overall workflow of the intelligent prepress proofing system is as follows: Figure 3 As shown, since its working method corresponds to the above-mentioned intelligent prepress proofing method, it will not be described again here to avoid repetition.

[0053] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the intelligent prepress proofing method described in the above embodiment. To avoid repetition, these steps will not be repeated here.

[0054] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it implements the intelligent prepress proofing method in the above method embodiment. To avoid repetition, it will not be described again here.

[0055] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention 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 the present invention, and should all be included within the protection scope of the present invention.

Claims

1. An intelligent prepress proofing method, characterized in that, The intelligent prepress proofing method is applied in an intelligent prepress proofing system consisting of at least three clients and at least one server. In the first client, a preset code value link format and a preset anti-counterfeiting technology type are set, and the preset code value link format and the preset anti-counterfeiting technology type are uploaded to the server. In the second client, a preset product standard sample is uploaded to the server; In the server, the preset product standard sample is verified in the first layer according to the preset code value link format and the preset anti-counterfeiting technology type; If the verification fails, the first feedback result will be pushed to the second client; If the verification is successful, wait to receive the product image corresponding to the preset product standard sample sent by the third client; In the third client, the physical image of the product corresponding to the preset product standard sample is uploaded to the server; In the server-side, the physical product image is compared with the preset product standard sample for a second layer of verification; If the verification fails, the second feedback result will be pushed to the third client; If the verification is successful, the production instruction will be sent to the third client.

2. The intelligent prepress proofing method as described in claim 1, characterized in that, The first layer of verification includes: The preset product standard sample is preprocessed to convert it into pixel-level image information. The barcode or QR code contained in the preset product standard sample is detected and decoded. The result of the decoding process is compared with the preset code value link format to obtain the verification result.

3. The intelligent prepress proofing method as described in claim 2, characterized in that, After obtaining the verification result, the following is also included: Image features are extracted from the preset product standard samples using a convolutional neural network. The image features are matched with the preset anti-counterfeiting technology type. If the matching result is incorrect, the error identifier and the reason for the error are encapsulated into the first feedback result.

4. The intelligent prepress proofing method as described in claim 1, characterized in that, The second layer of verification includes: Image preprocessing is performed on the physical product image and the preset product standard sample to convert the physical product image and the preset product standard sample into pixel-level image information; The convolutional neural network is used to extract features from the physical product image and the preset product standard sample to obtain their respective feature vectors; The feature vectors of the actual product image and the preset product standard sample are matched for similarity, and multiple verifications are performed on the anti-counterfeiting key points. The SURF algorithm is used to detect and describe feature points. Key points in two images are matched to evaluate similarity and calculate the Euclidean distance between their descriptors. If the minimum distance between the corresponding descriptors of the product image and the preset product standard sample is less than a preset empirical threshold, the match is considered successful. Otherwise, the key point positions of the match are marked. The system iteratively matches each corresponding anti-counterfeiting key point, marks the error type according to the feature matching result, and encapsulates the mark information together into the second feedback result.

5. The intelligent prepress proofing method as described in claim 3, characterized in that, The step of extracting image features from the preset product standard samples using a convolutional neural network includes: The local features in the preset product standard samples are learned through convolution and pooling operations, and the image information in the preset product standard samples is compressed into a set of feature vectors as the image features.

6. The intelligent prepress proofing method as described in claim 3, characterized in that, The image features are matched with the preset anti-counterfeiting technology type, including: The image features and the preset anti-counterfeiting technology type are matched using a preset algorithm based on distance matching or similarity matching.

7. An intelligent prepress proofing system, characterized in that, It includes a first client, a second client, a third client, and a server; the server communicates with the first client, the second client, and the third client via wired or wireless means, respectively. The first client is used to set a preset code value link format and a preset anti-counterfeiting technology type, and upload the preset code value link format and the preset anti-counterfeiting technology type to the server. The second client is used to upload a preset product standard sample to the server; The third client is used to upload the physical image of the product corresponding to the preset product standard sample to the server. The server is used to perform a first-level verification on the preset product standard sample according to the preset code value link format and the preset anti-counterfeiting technology type. If the verification fails, the first feedback result will be pushed to the second client; If the verification is successful, wait to receive the product image corresponding to the preset product standard sample sent by the third client; And a second-level verification method for comparing the physical image of the product with the preset product standard sample; If the verification fails, the second feedback result will be pushed to the third client; If the verification is successful, the production instruction will be sent to the third client.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the intelligent prepress proofing method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the intelligent prepress proofing method as described in any one of claims 1 to 6.

Images