Certificate quality inspection equipment and method

Abstract

The invention discloses certificate quality inspection equipment and a certificate quality inspection method. The equipment comprises a chip verification module, an image acquisition and detection unit 1, an image acquisition and detection unit 2 and an image acquisition and detection unit 3, the chip verification module is used for reading a certificate chip number, acquiring certificate making information from a database, connecting equipment with a key system, reading chip information, and comparing the chip information with the certificate making information to complete chip verification; the image acquisition and detection unit 1 acquires four images based on a color industrial camera and a visible light strip-shaped light source, reflects the four images through an optical plane mirror, generates a printing original image based on accreditation information, and sends the printing original image and the acquired images to a quality inspection engine for defect detection; the image acquisition and detection unit 2 is used for acquiring images based on a black and white industrial camera and an infrared strip-shaped light source, reflecting the images through an optical plane mirror, and sending a printed original image and the acquired images to a quality inspection engine for defect detection; and the image acquisition and detection unit 3 is used for acquiring front and back images of the certificate and sending the printed original image and the acquired image to a quality inspection engine for defect detection.

Description

Technical Field

[0001] The present invention relates to the technical field of document quality inspection, and more specifically, to a document quality inspection device and method. Background Art

[0002] Personalized production of security certificates is mainly completed by using personalized certificate-making equipment, including chip information writing, personal information printing and other processes. During the personalized production of certificates, defects may appear on the surface of the certificates due to various reasons. Currently, personalized certificates rely on manual inspection, which has problems such as large labor input, low detection efficiency, and easy fatigue. Therefore, it is necessary to develop automated certificate quality inspection equipment and systems to perform intelligent and automated detection of certificate quality, improve production efficiency, and reduce labor costs.

[0003] With the development of big data and artificial intelligence technology, the surface defect detection of documents based on machine vision technology has been continuously developed in combination with specific application practices, especially in the field of industrial quality inspection, where it is widely used, such as defect detection of circuit boards and defect detection of the appearance of consumer electronic products. For the quality inspection of high-security documents, such as identity cards and entry-exit documents, the introduction of machine vision technology for document quality inspection is gradually being considered. At present, there is a lack of document quality inspection equipment and methods. Summary of the invention

[0004] According to the present invention, a document quality inspection device and method are provided to solve the current problem of relying on manual inspection for personalized documents, which has the problems of large labor input, low inspection efficiency, easy fatigue, etc. Therefore, it is necessary to develop document automated quality inspection equipment and systems to perform intelligent and automated inspection of document quality, improve production efficiency, and reduce labor costs.

[0005] According to a first aspect of the present invention, there is provided a document quality inspection device, the device comprising a main control module, a quality inspection engine module, a display module, a card issuing module, a chip verification module, an image acquisition and detection unit 1, an image acquisition and detection unit 2, an image acquisition and detection unit 3, a steering module, and a sorting module;

[0006] The chip verification module is used to read the certificate chip number, obtain the certificate information from the database, connect the device to the key system, read the chip information, compare the chip information with the certificate information, and complete the chip verification;

[0007] The image acquisition and detection unit 1 is composed of a color industrial camera, a visible light strip light source, an optical plane mirror, and a light source controller. It collects four images based on the color industrial camera and the visible light strip light source, reflects them through the optical plane mirror, generates a printed original image based on the certification information, and sends the printed original image and the collected image to the quality inspection engine for defect detection;

[0008] The image acquisition and detection unit 2 is composed of a black and white industrial camera, an infrared strip light source, an optical plane mirror, and a light source controller. It acquires images based on the black and white industrial camera and the infrared strip light source, and reflects them through the optical plane mirror, and sends the printed original image and the acquired image to the quality inspection engine for defect detection;

[0009] The image acquisition and detection unit 3 is used to obtain the front and back images of the certificate, and send the printed original image and the acquired image to the quality inspection engine for defect detection.

[0010] Optionally, the main control module is used for whole machine control and deploying quality inspection control procedures:

[0011] The quality inspection engine module is deployed on the main control module or a separate server to detect surface defects of the certificate;

[0012] The display module is used to display the intelligent certificate quality detection system interface;

[0013] The card issuing module is used to send the certificate to be tested into the quality inspection equipment and has a certificate storage capacity.

[0014] Optionally, the diverting module is used to divert the documents to facilitate the sorting of the documents out of the card;

[0015] The sorting module is used to sort the certificates and output the cards. According to the requirements, a plurality of card output slots can be provided.

[0016] According to another aspect of the present invention, there is also provided a document quality inspection method, comprising:

[0017] The certificate is sent from the card issuing module to the quality inspection equipment to start the certificate quality inspection;

[0018] Based on the chip verification module, the chip number is read, the certificate information is obtained from the certificate database, and the print original image is generated to obtain the certificate information;

[0019] Reading chip information based on a reading and writing device, and comparing the chip information with the certification information;

[0020] If the chip information is consistent with the certification information, the image acquisition and detection unit 1 uses a multi-light source and an industrial camera to collect four images, and sends the four images to the quality inspection engine for defect detection;

[0021] When the document leaves the image acquisition and detection unit 1 and enters the acquisition and detection unit 2, pure K color detection of the document printing information is performed based on a black and white industrial camera and infrared strip light;

[0022] When the document leaves the image acquisition and detection unit 2 and enters the acquisition and detection unit 3, the document is scanned line by line on both sides based on the scanning device to obtain the front and back images of the document, and the front and back images are inspected for defects, including missing text on the front and back, front and back printing offset, two-color word defects, impurities in the non-printing area, and endorsement defects;

[0023] The front and back images of the certificate collected by the image acquisition and detection unit 3 are subjected to surface OCR verification and QR verification to achieve all-round multi-dimensional comparison and consistency verification of the card surface information, chip information, and certificate information to obtain the certificate quality inspection status.

[0024] Set the number of card output slots and classify the documents according to their quality inspection results.

[0025] Optionally, the image acquisition and detection unit 1 uses a multi-light source in conjunction with an industrial camera to acquire four images, and sends the four images to a quality inspection engine for defect detection, including:

[0026] The image acquisition and detection unit 1 uses multiple light sources in combination with an industrial camera to acquire images. The light source is turned on and off and switched by a program. Three groups of light strips are used for image acquisition. Four document images are acquired as required, and the printed original image and the acquired image are sent to the quality inspection engine module.

[0027] Among them, Image 1 is used to detect portrait defects and secondary image defects. Image 2 is used to detect film missing, film pull marks, adhesion defects, and film burn defects, as well as fish scale defects by adjusting gamma and aperture to reduce brightness. Image 3 and Image 4 are controlled by another two sets of strip lights for image acquisition, which are mainly used for film pull marks and fish scale defects.

[0028] Optionally, performing front and back text missing defect detection on the front and back images includes:

[0029] For the detection of missing text on the front and back of the certificate, a comparison is performed based on the original printed image and the captured image. The captured image is adjusted according to the size of the original printed image. The search starts near the set position according to the order of text arrangement. After the corresponding text is found, the text strokes are compared and the missing threshold is set.

[0030] Optionally, performing front-side and back-side printing deviation defect detection on the front-side and back-side images includes:

[0031] For the printing offset detection on the front and back of the certificate, including left and right offset, up and down offset and angle offset, the captured image is adjusted according to the size of the original printed image. For left and right offset and up and down offset, the text position is determined based on deep learning and compared with the standard position. The angle offset is combined with the text missing situation to determine the upper edge or lower edge of the text detection, and compared with the set threshold.

[0032] Optionally, performing two-color word missing defect detection on the front and back images includes:

[0033] For two-color character defects, including two-color character ghosting and two-color character color anomaly detection, the two-color characters are separated according to the printing reason, the number of color misaligned pixels is determined, and compared with the set threshold to determine whether it is two-color character ghosting. According to the color missing area, it is compared with the set threshold to determine whether it is a two-color character color anomaly.

[0034] Optionally, performing foreign matter defect detection on the non-printing area of ​​the front and back images includes:

[0035] For impurity detection in non-printing areas, the captured image is adjusted according to the size of the original printed image, the printing area is delineated, and the background interference in the non-printing area is filtered out through image processing to detect black spots and lines.

[0036] Optionally, set the number of card output slots and classify the documents according to the document quality inspection results, including:

[0037] The number of card output slots is set according to needs, which are divided into: completely normal, completely abnormal and blurred documents. Completely normal means that the document has been 100% normal after document quality inspection, completely abnormal means that the document has been 100% invalid after document quality inspection, and blurred documents are uncertain whether the document is invalid and require manual intervention and manual judgment.

[0038] Therefore, the certificate quality inspection equipment provided by the present invention integrates the verification of card surface information, chip information and certificate production information and surface defect detection, and has the functions of batch card issuance, chip verification, surface OCR verification, QR verification, certificate surface defect detection and classified card issuance, and can realize fast and accurate identification of the front and back information of the certificate, chip information reading, data comparison and consistency verification of card surface information, chip information and certificate production information, and intelligent detection of certificate surface defects, which greatly improves the quality inspection efficiency and reduces labor costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] A more complete understanding of exemplary embodiments of the present invention may be obtained by referring to the following drawings:

[0040] Figure 1 A schematic diagram of a document quality inspection device according to this embodiment;

[0041] Figure 2 It is a schematic diagram of a document quality inspection method described in this embodiment. DETAILED DESCRIPTION

[0042] Now, exemplary embodiments of the present invention are described with reference to the accompanying drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided to disclose the present invention in detail and completely and to fully convey the scope of the present invention to those skilled in the art. The terms used in the exemplary embodiments shown in the accompanying drawings are not intended to limit the present invention. In the accompanying drawings, the same units / elements are marked with the same reference numerals.

[0043] Unless otherwise specified, the terms (including technical terms) used herein have the commonly understood meanings to those skilled in the art. In addition, it is understood that the terms defined in commonly used dictionaries should be understood to have the same meanings as those in the context of the relevant fields, and should not be understood as idealized or overly formal meanings.

[0044] According to a first aspect of the present invention, there is provided a document quality inspection device, the device comprising a main control module, a quality inspection engine module, a display module, a card issuing module, a chip verification module, an image acquisition and detection unit 1, an image acquisition and detection unit 2, an image acquisition and detection unit 3, a steering module, and a sorting module;

[0045] The chip verification module is used to read the certificate chip number, obtain the certificate information from the database, connect the device to the key system, read the chip information, compare the chip information with the certificate information, and complete the chip verification;

[0046] The image acquisition and detection unit 1 is composed of a color industrial camera, a visible light strip light source, an optical plane mirror, and a light source controller. It collects four images based on the color industrial camera and the visible light strip light source, reflects them through the optical plane mirror, generates a printed original image based on the certification information, and sends the printed original image and the collected image to the quality inspection engine for defect detection;

[0047] The image acquisition and detection unit 2 is composed of a black and white industrial camera, an infrared strip light source, an optical plane mirror, and a light source controller. It acquires images based on the black and white industrial camera and the infrared strip light source, and reflects them through the optical plane mirror, and sends the printed original image and the acquired image to the quality inspection engine for defect detection;

[0048] The image acquisition and detection unit 3 is used to obtain the front and back images of the certificate, and send the printed original image and the acquired image to the quality inspection engine for defect detection.

[0049] Optionally, the main control module is used for whole machine control and deploying quality inspection control procedures:

[0050] The quality inspection engine module is deployed on the main control module or a separate server to detect surface defects of the certificate;

[0051] The display module is used to display the intelligent certificate quality detection system interface;

[0052] The card issuing module is used to send the certificate to be tested into the quality inspection equipment and has a certificate storage capacity.

[0053] Optionally, the diverting module is used to divert the documents to facilitate the sorting of the documents out of the card;

[0054] The sorting module is used to sort the certificates and output the cards. According to the requirements, a plurality of card output slots can be provided.

[0055] Specifically, refer to Figure 1 As shown, the main control module is responsible for the control of the entire machine. An industrial computer can be selected to deploy a quality inspection control program.

[0056] The quality inspection engine module is responsible for document surface defect detection. The quality inspection engine can be deployed on the main control module. One quality inspection device can deploy one set of quality inspection engines at the same time. The quality inspection engine can also be deployed on a separate server to enable multiple quality inspection devices to share one set of quality inspection engines. The quality inspection engine needs to be based on high-performance CPU and GPU for reasoning.

[0057] The display module is used to display the interface of the intelligent document quality inspection system.

[0058] The card issuing module is responsible for sending the certificate to be tested into the quality inspection equipment. The card issuing module has a certain certificate storage capacity.

[0059] The chip verification module is a mechanism with a chip reading function at the bottom, and can ensure that the edges of the document are fully exposed to facilitate image acquisition. The module can read the document chip number, and then obtain the certificate information from the database. The device is connected to the key system, which can read all the chip information, compare the chip information with the certificate information, and complete the chip verification.

[0060] A color industrial camera, a light source, and an optical plane mirror are installed above the chip verification module to form an image acquisition and detection unit 1. Considering the height of the equipment, the camera is horizontal and reflected by an optical plane mirror. Considering that image acquisition needs to highlight defects, a six-bar light source ring structure is used here to acquire four images, and the images are sent to the quality inspection engine for defect detection, which mainly detects portrait defects, secondary image defects, coating defects, fish scale patterns, etc.

[0061] For the Hong Kong and Macau Pass of the People's Republic of China, considering that some images and texts on the certificate are printed in pure K color instead of CMY three-color synthesis, it is necessary to collect images based on black and white industrial cameras and infrared strip light, and reflect them through optical plane mirrors, and send the collected images to the quality inspection engine for defect detection to ensure that the printer configuration is correct and the color is normal.

[0062] The image acquisition and detection unit 3 performs double-sided line-by-line scanning of the certificate based on a scanning device, obtains the front and back images of the certificate, and sends the printed original image and the acquired image to the quality inspection engine for defect detection, mainly for defects such as missing front text, two-color character defects, front printing offset, impurities in the non-printing area, and signature defects.

[0063] In addition, the images collected by the image acquisition and detection unit 3 are subjected to OCR verification, QR verification and three-party data comparison.

[0064] The steering module is mainly responsible for document steering, making it easier for documents to be sorted out and carded.

[0065] The sorting module is mainly responsible for the classification and card output of certificates. Multiple card output slots can be set according to needs.

[0066] According to another aspect of the present invention, a document quality inspection method 200 is also provided. Figure 2 As shown, the method 200 includes:

[0067] S201: The certificate is issued from the card issuing module, enters the quality inspection equipment, and starts the certificate quality inspection;

[0068] S202: Read the chip number based on the chip verification module, obtain the certificate information from the certificate database, generate a print original image, and obtain the certificate information;

[0069] S203: Reading chip information based on a reading and writing device, and comparing the chip information with the certification information;

[0070] S204: If the chip information is consistent with the certification information, the image acquisition and detection unit 1 uses a multi-light source and an industrial camera to acquire four images, and sends the four images to the quality inspection engine for defect detection;

[0071] S205: After the document leaves the image acquisition and detection unit 1, it enters the acquisition and detection unit 2, and performs pure K color detection of the document printing information based on a black and white industrial camera and infrared strip light;

[0072] S206: After the document leaves the image acquisition and detection unit 2, it enters the acquisition and detection unit 3, and the document is scanned line by line on both sides based on the scanning device to obtain the front and back images of the document, and the front and back images are inspected for defects, including front and back text missing, front and back printing offset, two-color word defects, impurities in the non-printing area, and endorsement defects;

[0073] S207: Perform surface OCR verification and QR verification on the front and back images of the certificate collected by the image collection and detection unit 3, realize all-round multi-dimensional comparison and consistency verification of the card surface information, chip information, and certificate information, and obtain the certificate quality inspection status;

[0074] S208: Set the number of card output slots and classify the documents according to the document quality inspection results.

[0075] Specifically, the certificate is sent out from the card issuing module, enters the quality inspection equipment, and starts the certificate quality inspection.

[0076] Based on the chip verification module, the chip number is read, the certificate information is obtained from the certificate database, and the printing original image is generated.

[0077] The chip information is read by the reading and writing device and compared with the certification information.

[0078] While the chip is being verified, image acquisition is performed based on the image acquisition detection unit 1. In order to highlight defects, multiple light sources are used in conjunction with industrial cameras for image acquisition, and the light source is turned on and off and switched by program. Three groups of strip lights are used here for image acquisition. Four certificate images are acquired as required, and the printed original image and acquired image are sent to the quality inspection engine for defect detection. Image 1 is used to detect portrait defects and secondary image defects. Image 2 is used to detect coating defects such as coating missing, coating stretch marks, bonding defects, and scalding by adjusting gamma and aperture to reduce brightness, as well as fish scale defect detection. Image 3 and Image 4 are controlled by another two groups of strip lights for image acquisition, which are mainly used for coating stretch marks and fish scale defect detection.

[0079] After leaving the image acquisition and detection unit 1, the document enters the acquisition and detection unit 2, where pure K color detection of the document printing information is performed based on a black and white industrial camera and infrared strip light to ensure that part of the document image and text information is printed in pure K color, not CMY three-color synthesis.

[0080] After leaving the image acquisition and detection unit 2, the document enters the acquisition and detection unit 3, where the document is scanned line by line on both sides based on the scanning device, and the front and back images of the document are obtained, and defect detection such as missing text on the front and back, front and back printing offset, two-color text defects, impurities in the non-printing area, and signature defects is performed.

[0081] The image acquisition and detection unit 3 collects the front and back images of the certificate, performs surface OCR verification and QR verification, and realizes all-round multi-dimensional comparison and consistency verification of the card surface information, chip information, and certificate information.

[0082] According to the document quality inspection, the documents are classified and the number of card slots can be set according to the needs. There are three categories: completely normal, completely abnormal and blurred documents. Completely normal means that the document quality inspection ensures that the document is 100% normal; completely abnormal means that the document quality inspection ensures that the document is 100% invalid; blurred documents are uncertain whether the document is invalid and require manual intervention and manual judgment.

[0083] The quality inspection engine module is responsible for the surface defect detection of the certificate. The quality inspection engine can be deployed on the main control module. One quality inspection device can deploy a set of quality inspection engines at the same time. The quality inspection engine can also be deployed on a separate server to achieve multiple quality inspection devices sharing one set of quality inspection engines. The quality inspection device communicates with the quality inspection engine based on the HTTP POST request method.

[0084] Here, image 1 captured by image acquisition unit 1 is used for portrait defect and secondary image defect detection, image 2 is used for coating defect detection such as coating missing, coating stretch marks, bonding defects, scalding, and fish scale defect detection, and image 3 and image 4 are used for coating stretch marks and fish scale detection.

[0085] The image 5 captured by the image capture unit 2 is used for detecting pure black of partial images and texts.

[0086] The front image 6 acquired by the image acquisition unit 3 is used to detect front text missing, two-color character defects, front printing deviation, and impurities in the non-printing area, and the back image 7 is used to detect back text missing and back printing deviation.

[0087] The details of the defect detection algorithm are as follows:

[0088] For the detection of missing text on the front and back of the certificate, a comparison is performed based on the original printed image and the captured image. The captured image is adjusted according to the size of the original printed image. The search starts near the set position according to the text arrangement order. After the corresponding text is found, the text strokes are compared. The missing threshold can be set here.

[0089] For the detection of printing offset on the front and back of the certificate, there are three detection items: left-right offset, up-down offset and angle offset. The captured image is adjusted according to the size of the original printed image. For left-right offset and up-down offset, the text position is determined based on deep learning and compared with the standard position. The angle offset is combined with the text missing situation to determine whether the text is detected at the top edge or the bottom edge, and compared with the set threshold.

[0090] For two-color character defects, it mainly includes two-color character ghosting and two-color character color anomaly detection. The two-color characters are separated according to the printing reason, the number of color misaligned pixels is determined, and compared with the set threshold to determine whether it is two-color character ghosting; according to the color missing area, it is compared with the set threshold to determine whether it is a two-color character color anomaly.

[0091] For impurity detection in non-printing areas, the captured image is adjusted according to the size of the original printed image, the printing area is delineated, and the background interference in the non-printing area is filtered out through image processing to detect black spots and lines.

[0092] Portrait defects mainly include missing portraits, ghosting, color spots, impurities, horizontal stripes, vertical stripes, diagonal stripes, portrait color cast, horizontal lines, etc. First, the captured image needs to be adjusted according to the size of the original printed image. For missing portraits, first align the original printed portrait with the captured image to determine the approximate range, and then perform inference calculations based on the deep convolutional neural network to determine the missing portrait. For defects such as ghosting, color spots, impurities, horizontal stripes, vertical stripes, diagonal stripes, portrait color cast, horizontal lines, etc., data is generated through algorithms and trained based on deep convolutional neural networks to complete module development.

[0093] The coating defects mainly include coating missing, coating pull marks, bonding defects, coating burns, etc. The defects are highlighted based on light sources and industrial cameras, data is generated through algorithms, and training is performed based on deep convolutional neural networks to complete module development.

[0094] For fish scale detection, defects are highlighted based on light sources and industrial cameras, data is generated through algorithms, and training is performed based on deep convolutional neural networks to complete module development.

[0095] The black edge detection of the certificate is based on the image processing algorithm.

[0096] For secondary image defect detection, the captured image is adjusted according to the size of the printed original image, and the position of the secondary image in the printed original image is determined. The secondary image of the original image is binarized, and the positions of the hair, facial features and other areas are determined for color detection and compared with the scanned image results. If the color of the scanned image area is abnormal, it is judged that the secondary image is missing.

[0097] Detect defects in the original printed image, determine the outline of the original printed image portrait, and detect impurities outside the area.

[0098] The intelligent document quality inspection system is an intelligent system used for document quality inspection and display. Its main functions include quality inspection monitoring, data query, parameter setting, and operation and maintenance management.

[0099] The quality inspection monitoring interface is used to display the real-time quality inspection screen, personal information side, front inspection picture, endorsement information side, back inspection picture, and mark defects on the inspection picture. For three-party data comparison, the certificate information and chip information are displayed, and abnormal data is marked in red. The certificate number and inspection results are displayed at the bottom of the interface.

[0100] Select a date from the calendar to view the test results during this period, the detection status of each defect type, data packet number, test time, machine number, printer ID, and printer serial number. If the test result is qualified, it will be indicated by a green check mark, and if it is unqualified, it will be indicated by a red cross.

[0101] There is an option of "View only defective documents", which only displays images of defective documents and error causes. If you check "View only defective documents", you can query the document images corresponding to the defects according to the defect type.

[0102] There is an ID number input box where you can enter the ID number and view the image of the specified ID number.

[0103] Click the "Details" button to query the current document image, including the original image printed on the front, the test result image, the original image printed on the back, and the test result image.

[0104] Parameter settings are divided into system settings and quality inspection item settings. System settings are used to set the number of days for data storage and the storage path for logs and images. Quality inspection item settings are used to set the quality inspection items on or off and configure the detection threshold.

[0105] Under normal circumstances, the quality inspection engine starts automatically when the computer is turned on. A service program monitors the engine and automatically restarts if an abnormality occurs.

[0106] Network connectivity test of the certificate-making program: ping the certificate-making program. If the connection fails, it indicates that the network connection has failed. If the connection can be made, it indicates that the network connection has succeeded.

[0107] Quality inspection algorithm network connectivity test: You can ping the quality inspection algorithm service. If it fails to connect, it indicates that the network connection failed; if it can connect, it indicates that the network connection is successful.

[0108] Quality inspection algorithm service test: You can test whether the interface is started. If the interface returns normal, it indicates that the interface is started; if the interface returns abnormal, it indicates that the interface is abnormal.

[0109] Optionally, the image acquisition and detection unit 1 uses a multi-light source in conjunction with an industrial camera to acquire four images, and sends the four images to a quality inspection engine for defect detection, including:

[0110] The image acquisition and detection unit 1 uses multiple light sources in combination with an industrial camera to acquire images. The light source is turned on and off and switched by a program. Three groups of light strips are used for image acquisition. Four document images are acquired as required, and the printed original image and the acquired image are sent to the quality inspection engine module.

[0111] Among them, Image 1 is used to detect portrait defects and secondary image defects. Image 2 is used to detect film missing, film pull marks, adhesion defects, and film burn defects, as well as fish scale defects by adjusting gamma and aperture to reduce brightness. Image 3 and Image 4 are controlled by another two sets of strip lights for image acquisition, which are mainly used for film pull marks and fish scale defects.

[0112] Optionally, performing front and back text missing defect detection on the front and back images includes:

[0113] For the detection of missing text on the front and back of the certificate, a comparison is performed based on the original printed image and the captured image. The captured image is adjusted according to the size of the original printed image. The search starts near the set position according to the order of text arrangement. After the corresponding text is found, the text strokes are compared and the missing threshold is set.

[0114] Optionally, performing front-side and back-side printing deviation defect detection on the front-side and back-side images includes:

[0115] For the printing offset detection on the front and back of the certificate, including left and right offset, up and down offset and angle offset, the captured image is adjusted according to the size of the original printed image. For left and right offset and up and down offset, the text position is determined based on deep learning and compared with the standard position. The angle offset is combined with the text missing situation to determine the upper edge or lower edge of the text detection, and compared with the set threshold.

[0116] Optionally, performing two-color word missing defect detection on the front and back images includes:

[0117] For two-color character defects, including two-color character ghosting and two-color character color anomaly detection, the two-color characters are separated according to the printing reason, the number of color misaligned pixels is determined, and compared with the set threshold to determine whether it is two-color character ghosting. According to the color missing area, it is compared with the set threshold to determine whether it is a two-color character color anomaly.

[0118] Optionally, performing foreign matter defect detection on the non-printing area of ​​the front and back images includes:

[0119] For impurity detection in non-printing areas, the captured image is adjusted according to the size of the original printed image, the printing area is delineated, and the background interference in the non-printing area is filtered out through image processing to detect black spots and lines.

[0120] Optionally, set the number of card output slots and classify the documents according to the document quality inspection results, including:

[0121] The number of card output slots is set according to needs, which are divided into: completely normal, completely abnormal and blurred documents. Completely normal means that the document has been 100% normal after document quality inspection, completely abnormal means that the document has been 100% invalid after document quality inspection, and blurred documents are uncertain whether the document is invalid and require manual intervention and manual judgment.

[0122] Thus, automatic full-volume quality inspection can be achieved. The equipment integrates card surface information, chip information, certificate production information verification and surface defect detection, and has the functions of batch card issuance, chip verification, surface OCR verification, QR verification, certificate surface defect detection and classified card output. It can realize fast and accurate identification of certificate front and back information, chip information reading, data comparison and consistency verification of card surface information, chip information and certificate production information, and intelligent detection of certificate surface defects, which greatly improves quality inspection efficiency and reduces labor costs.

[0123] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0124] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0125] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0126] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0127] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0128] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A document quality inspection device, characterized in that: The device includes a main control module, a quality inspection engine module, a display module, a card issuing module, a chip verification module, an image acquisition and detection unit 1, an image acquisition and detection unit 2, an image acquisition and detection unit 3, a steering module, and a sorting module; The chip verification module is used to read the certificate chip number, obtain the certificate information from the database, connect the device to the key system, read the chip information, compare the chip information with the certificate information, and complete the chip verification; The image acquisition and detection unit 1 is composed of a color industrial camera, a visible light strip light source, an optical plane mirror, and a light source controller. It collects four images based on the color industrial camera and the visible light strip light source, reflects them through the optical plane mirror, generates a printed original image based on the certification information, and sends the printed original image and the collected image to the quality inspection engine for defect detection; The image acquisition and detection unit 2 is composed of a black and white industrial camera, an infrared strip light source, an optical plane mirror, and a light source controller. It acquires images based on the black and white industrial camera and the infrared strip light source, and reflects them through the optical plane mirror, and sends the printed original image and the acquired image to the quality inspection engine for defect detection; The image acquisition and detection unit 3 is used to obtain the front and back images of the certificate, and send the printed original image and the acquired image to the quality inspection engine for defect detection.

2. The device according to claim 1, characterized in that The main control module is used to control the whole machine and deploy the quality inspection control program: The quality inspection engine module is deployed on the main control module or a separate server to detect surface defects of the certificate; The display module is used to display the intelligent certificate quality detection system interface; The card issuing module is used to send the certificate to be tested into the quality inspection equipment and has a certificate storage capacity.

3. The device according to claim 1, characterized in that The steering module is used to steering the documents to facilitate the sorting of the documents out of the card; The sorting module is used to sort the certificates and output the cards. According to the requirements, a plurality of card output slots can be provided.

4. A document quality inspection method, characterized in that: include: The certificate is sent from the card issuing module to the quality inspection equipment to start the certificate quality inspection; Based on the chip verification module, the chip number is read, the certificate information is obtained from the certificate database, and the print original image is generated to obtain the certificate information; Reading chip information based on a reading and writing device, and comparing the chip information with the certification information; If the chip information is consistent with the certification information, the image acquisition and detection unit 1 uses a multi-light source and an industrial camera to collect four images, and sends the four images to the quality inspection engine for defect detection; When the document leaves the image acquisition and detection unit 1 and enters the acquisition and detection unit 2, pure K color detection of the document printing information is performed based on a black and white industrial camera and infrared strip light; When the document leaves the image acquisition and detection unit 2 and enters the acquisition and detection unit 3, the document is scanned line by line on both sides based on the scanning device to obtain the front and back images of the document, and the front and back images are inspected for defects, including missing text on the front and back, front and back printing offset, two-color word defects, impurities in the non-printing area, and endorsement defects; The front and back images of the certificate collected by the image acquisition and detection unit 3 are subjected to surface OCR verification and QR verification to achieve all-round multi-dimensional comparison and consistency verification of the card surface information, chip information, and certificate information, and obtain the certificate quality inspection status; Set the number of card output slots and classify the documents according to their quality inspection results.

5. The method according to claim 1, characterized in that Based on the image acquisition detection unit 1, four images are collected by using multiple light sources in combination with an industrial camera, and the four images are sent to the quality inspection engine for defect detection, including: The image acquisition and detection unit 1 uses multiple light sources in combination with an industrial camera to acquire images. The light source is turned on and off and switched by a program. Three groups of light strips are used for image acquisition. Four document images are acquired as required, and the printed original image and the acquired image are sent to the quality inspection engine module. Among them, Image 1 is used to detect portrait defects and secondary image defects. Image 2 is used to detect film missing, film pull marks, adhesion defects, and film burn defects, as well as fish scale defects by adjusting gamma and aperture to reduce brightness. Image 3 and Image 4 are controlled by another two sets of strip lights for image acquisition, which are mainly used for film pull marks and fish scale defects.

6. The method according to claim 4, characterized in that Performing front and back text missing defect detection on the front and back images includes: For the detection of missing text on the front and back of the certificate, a comparison is performed based on the original printed image and the captured image. The captured image is adjusted according to the size of the original printed image. The search starts near the set position according to the order of text arrangement. After the corresponding text is found, the text strokes are compared and the missing threshold is set.

7. The method according to claim 6, characterized in that Performing front and back printing deviation defect detection on the front and back images, including: For the printing offset detection on the front and back of the certificate, including left and right offset, up and down offset and angle offset, the captured image is adjusted according to the size of the original printed image. For left and right offset and up and down offset, the text position is determined based on deep learning and compared with the standard position. The angle offset is combined with the text missing situation to determine the upper edge or lower edge of the text detection, and compared with the set threshold.

8. The method according to claim 7, characterized in that Performing two-color missing character defect detection on the front and back images includes: For two-color character defects, including two-color character ghosting and two-color character color anomaly detection, the two-color characters are separated according to the printing reason, the number of color misaligned pixels is determined, and compared with the set threshold to determine whether it is two-color character ghosting. According to the color missing area, it is compared with the set threshold to determine whether it is a two-color character color anomaly.

9. The method according to claim 4, characterized in that Performing foreign matter defect detection on the non-printing area of ​​the front and back images includes: For impurity detection in non-printing areas, the captured image is adjusted according to the size of the original printed image, the printing area is delineated, and the background interference in the non-printing area is filtered out through image processing to detect black spots and lines.

10. The method according to claim 4, characterized in that Set the number of card output slots and classify the documents according to their quality inspection results, including: The number of card output slots is set according to needs, which are divided into: completely normal, completely abnormal and blurred documents. Completely normal means that the document has been 100% normal after document quality inspection, completely abnormal means that the document has been 100% invalid after document quality inspection, and blurred documents are uncertain whether the document is invalid and require manual intervention and manual judgment.

Images