Control method and device of automatic stamping device capable of being expanded in cascading mode

Abstract

The invention relates to the technical field of office automation machinery control, and discloses a control method and device of an automatic stamping device capable of being expanded in a cascading mode, and a document is controlled to be conveyed in the automatic stamping device along a preset path; when it is detected that a document is transmitted to a document content identification and verification area of the automatic stamping device, whether the document is matched with pre-stored standard information or not is judged, and if yes, the document is stamped according to each identified pre-marked stamping area in the document and a corresponding cascade stamp in the automatic stamping device; stamping control parameters of each seal are generated so as to control the seal to execute corresponding stamping control operation; and after the seal control operation is executed, multi-dimensional seal quality analysis is executed on the document to judge whether the document meets a preset seal quality condition or not, and if yes, the document is transmitted to a document collection device. Therefore, the control flexibility of the automatic stamping device can be improved while the stamping efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of office automation machinery control, and in particular to a control method and device of a cascadingly expandable automatic stamping device. BACKGROUND

[0002] As an important tool for improving office efficiency and standardizing the stamping process, the automatic stamping device has gradually replaced part of the traditional manual stamping operation.

[0003] Currently, the solutions on the market for processing multiple stamping requirements mainly fall into two categories: one is to use a single stamping device and rely on manual stamping replacement. This method takes several minutes to replace the stamp each time, and frequent disassembly can introduce positioning errors, which is low in operation efficiency and cannot adapt to batch and multi-category document processing scenarios. The second is to use an all-in-one device with a fixed number of built-in stamps. The number and type of stamps are fixed at the time of factory shipment, and when the business of an enterprise grows or the type of stamp changes, the original device cannot be upgraded to adapt, and the whole machine must be replaced, resulting in high investment cost and low equipment reuse rate.

[0004] Therefore, it is particularly important to propose a technical solution that improves stamping efficiency while improving the control flexibility of the automatic stamping device. SUMMARY

[0005] The present application provides a control method and device of a cascadingly expandable automatic stamping device, which can improve stamping efficiency while improving the control flexibility of the automatic stamping device.

[0006] To solve the above technical problems, the present application discloses a control method of a cascadingly expandable automatic stamping device, which comprises: controlling the document to be conveyed along a preset path in the automatic stamping device; When it is detected that the document is conveyed to the document content recognition and verification area of the automatic stamping device, it is determined whether the document matches the pre-stored standard information, and if so, the document is conveyed to the positioning and stamping area of the automatic stamping device; When it is detected that the document is conveyed to the positioning and stamping area, the stamping control parameters of each stamp are generated according to the identified pre-marked stamping area in the document and the cascading stamps in the automatic stamping device, to control the stamps to perform corresponding stamping control operations, the stamping control parameters including movement control parameters and stamping pressure control parameters; When the seal control operation is completed, the document is transmitted to a seal review area; multi-dimensional seal quality analysis is performed on the document to determine whether the document meets the preset seal quality condition, and if so, the document is transmitted to a document collection device.

[0007] As an optional implementation, in the first aspect of the present application, the determination of whether the document matches the pre-stored standard information comprises: Obtaining a complete image of the document; Performing a preprocessing operation on the complete image, the preprocessing operation comprising at least one of a color separation operation, a distortion correction operation, and a noise filtering operation; Inputting the preprocessed complete image into a preset text recognition engine to recognize and extract full-text key information of the document, the full-text key information comprising at least one of a title, a number, a body, and an icon of the document; Comparing the full-text key information with the pre-stored standard information to calculate a target similarity value between the full-text key information and the pre-stored standard information; Determining whether the target similarity value is greater than or equal to a preset similarity threshold, and when it is determined that the target similarity value is greater than or equal to the preset similarity threshold, it is determined that the document matches the pre-stored standard information.

[0008] As an optional implementation, in the first aspect of the present application, the generation of seal control parameters of each seal in the document according to the identified pre-marked seal area and the cascadeable seal in the corresponding automated seal device to control the seal to perform corresponding seal control operation comprises: For each pre-marked seal area in the document, extracting the region coordinates of the seal area according to a preset image segmentation and contour extraction algorithm; Obtaining the current position coordinates of the cascadeable seal in the corresponding automated seal device corresponding to the seal area; Generating seal control parameters of each seal according to the current position coordinates and the region coordinates to control the seal to perform corresponding seal control operation, the movement control parameters comprising horizontal movement control parameters and vertical movement control parameters.

[0009] As an optional implementation, in the first aspect of the present application, the multi-dimensional seal quality analysis performed on the document to determine whether the document meets the preset seal quality condition comprises: Obtaining a local image of an actual seal part of the document in the seal review area; Simultaneously perform text clarity analysis and text position compliance analysis on the local image; the text clarity analysis includes calculating the gradient features of the text edges in the local image to obtain a clarity value; the text position compliance analysis includes matching the text outline in the local image with the pre-stored standard text outline by feature points to calculate the position offset. When the clarity value meets the preset clarity standard and the position offset is less than or equal to the preset tolerance threshold, the document is determined to meet the preset stamping quality conditions.

[0010] As an optional implementation, in the first aspect of the present invention, comparing the full-text key information with pre-stored standard information to calculate the target similarity value between the full-text key information and the pre-stored standard information includes: Semantic similarity calculation is performed between the full-text key information and the corresponding standard text in the pre-stored standard information; the semantic similarity calculation includes extracting the semantic feature vectors of the full-text key information and the standard text respectively, and calculating the cosine similarity between the two semantic feature vectors as the target similarity value.

[0011] As an optional implementation, in the first aspect of the present invention, for each pre-marked stamped area in the document, the step of extracting the region coordinates of the stamped area according to a preset image segmentation and contour extraction algorithm includes: The image coordinates of pre-set reference marks on the document are identified using a visual sensor; The image coordinates of the reference mark are combined with pre-calibrated visual perception parameters to convert them into world coordinates in the coordinate system of the stamping device; Based on the world coordinates and the pre-defined offset of the stamping area relative to the reference mark, the regional coordinates of the stamping area are calculated.

[0012] As an optional implementation, in the first aspect of the present invention, the method further includes: Based on the processing status of the document content recognition and verification area and the ready status of the positioning and stamping area, the start and stop of the conveying device in the automated stamping device are dynamically controlled so that the document is paused when it arrives at the corresponding area until the processing of that area is completed. When a document transmission anomaly is detected by a sensor located in the transmission path of the automated stamping device, the diversion mechanism in the automated stamping device is controlled to guide the abnormal document to an independent collection path and trigger an alarm signal.

[0013] A second aspect of the present invention discloses a control device for an automated stamping device that can be cascaded and expanded, the device comprising: The delivery control module is used to control the delivery of documents along a preset path in the automated stamping device; The detection and judgment module is used to determine whether the document matches the pre-stored standard information when the document is detected to have been transmitted to the document content recognition and verification area of ​​the automated stamping device. The transmission control module is also used to transmit the document to the positioning and stamping area of ​​the automated stamping device if a match is found. The detection control module is used to generate stamping control parameters for each stamp based on each pre-marked stamping area in the document and the corresponding cascaded stamps in the automated stamping device when the document is detected to have been delivered to the positioning stamping area. The stamping control parameters include movement control parameters and stamping pressure control parameters. The transmission control module is also used to transmit the document to the stamping review area after the stamping control operation is completed. The detection and judgment module is also used to perform multi-dimensional seal quality analysis on the document to determine whether the document meets the preset stamping quality conditions. The transmission control module is further configured to transmit the document to the document collection device if the conditions are met.

[0014] As an optional implementation, in a second aspect of the present invention, the specific method by which the detection and judgment module determines whether the document matches pre-stored standard information includes: Obtain a complete image of the document; Perform preprocessing operations on the complete image, the preprocessing operations including at least one of color separation operation, distortion correction operation, and noise filtering operation; The preprocessed complete image is input into a preset text recognition engine to identify and extract the full text key information of the document, which includes at least one of the document's title, number, body text, and icon. The key information in the full text is compared with the pre-stored standard information to calculate the target similarity value between the key information in the full text and the pre-stored standard information. Determine whether the target similarity value is greater than or equal to a preset similarity threshold. If the target similarity value is determined to be greater than or equal to the preset similarity threshold, then the document is determined to match the pre-stored standard information.

[0015] As an optional implementation, in a second aspect of the present invention, the detection control module generates stamping control parameters for each stamp based on each pre-marked stamping area in the document and the corresponding cascaded stamps in the automated stamping device, in order to control the stamps to perform corresponding stamping control operations. The specific method includes: For each pre-marked stamped area in the document, the region coordinates of the stamped area are extracted according to a preset image segmentation and contour extraction algorithm; Obtain the current position coordinates of the cascaded stamps in the automated stamping device corresponding to the stamping area; Based on the current position coordinates and the area coordinates, stamping control parameters are generated for each stamp to control the stamp to perform the corresponding stamping control operation. The movement control parameters include horizontal movement control parameters and vertical movement control parameters.

[0016] As an optional implementation, in a second aspect of the present invention, the specific method by which the detection and judgment module performs multi-dimensional seal quality analysis on the document to determine whether the document meets preset stamping quality conditions includes: Obtain a partial image of the actual stamped area of ​​the document within the stamped review area; Simultaneously perform text clarity analysis and text position compliance analysis on the local image; the text clarity analysis includes calculating the gradient features of the text edges in the local image to obtain a clarity value; the text position compliance analysis includes matching the text outline in the local image with the pre-stored standard text outline by feature points to calculate the position offset. When the clarity value meets the preset clarity standard and the position offset is less than or equal to the preset tolerance threshold, the document is determined to meet the preset stamping quality conditions.

[0017] As an optional implementation, in a second aspect of the present invention, the specific method by which the detection and judgment module compares the full-text key information with pre-stored standard information to calculate the target similarity value between the full-text key information and the pre-stored standard information includes: Semantic similarity calculation is performed between the full-text key information and the corresponding standard text in the pre-stored standard information; the semantic similarity calculation includes extracting the semantic feature vectors of the full-text key information and the standard text respectively, and calculating the cosine similarity between the two semantic feature vectors as the target similarity value.

[0018] As an optional implementation, in the second aspect of the present invention, for each pre-marked stamped area in the document, the specific method by which the detection control module extracts the region coordinates of the stamped area according to a preset image segmentation and contour extraction algorithm includes: The image coordinates of pre-set reference marks on the document are identified using a visual sensor; The image coordinates of the reference mark are combined with pre-calibrated visual perception parameters to convert them into world coordinates in the coordinate system of the stamping device; Based on the world coordinates and the pre-defined offset of the stamping area relative to the reference mark, the regional coordinates of the stamping area are calculated.

[0019] As an optional implementation, in a second aspect of the present invention, the transmission control module is further configured to dynamically control the start and stop of the transmission device in the automated stamping device according to the processing status of the document content recognition and verification area and the ready status of the positioning stamping area, so that the document is paused when it arrives at the corresponding area until the processing of that area is completed. The transmission control module is also used to control the diversion mechanism in the automated stamping device to guide the abnormal document to an independent collection path and trigger an alarm signal when the document transmission abnormality is detected by the sensor set in the transmission path of the automated stamping device.

[0020] A third aspect of the present invention discloses a control device for another cascadeable and expandable automated stamping device, the device comprising: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the control method of the cascaded and expandable automated stamping device disclosed in the first aspect of the present invention.

[0021] The fourth aspect of the present invention discloses a computer storage medium storing computer instructions, which, when invoked, are used to execute the control method of the cascadeable and expandable automated stamping device disclosed in the first aspect of the present invention.

[0022] Compared with the prior art, the embodiments of the present invention have the following beneficial effects: In this embodiment of the invention, the document is controlled to be transported along a preset path in the automated stamping device. When the document is detected to have reached the document content recognition and verification area of ​​the automated stamping device, it is determined whether the document matches the pre-stored standard information. If they match, the document is transported to the positioning stamping area of ​​the automated stamping device. When the document is detected to have reached the positioning stamping area, stamping control parameters for each stamp are generated based on each stamping area pre-marked in the document and the corresponding cascaded stamps in the automated stamping device. These parameters control the stamps to perform corresponding stamping control operations. The stamping control parameters include movement control parameters and stamping pressure control parameters. After the stamping control operations are completed, the document is transported to the stamping review area. Multi-dimensional imprint quality analysis is performed on the document to determine whether it meets the preset stamping quality conditions. If it does, the document is transported to the document collection device. It is evident that implementing this invention, by constructing a complete automated closed-loop process of "transmission-content verification-positioning and stamping-quality review," can significantly improve the continuity and automation of stamping operations. This frees manual labor from repetitive and error-prone tasks such as document delivery, verification, alignment, and inspection, thereby reducing labor costs and the subjective risk of operational errors, achieving highly efficient and reliable fully automated stamping operations. By introducing "document content recognition and verification" as a pre-stamping decision-making step, the accuracy and security of stamping operations can be effectively improved, thus preventing stamping errors caused by document content errors (such as text errors or unexpected documents) from the source, thereby ensuring the legal validity and seriousness of the seal and achieving intelligent risk control. By adopting a programmable and adaptive positioning method that "generates control parameters based on the identified stamping area," the flexibility, accuracy, and consistency of the stamping position can be significantly improved. This helps to eliminate reliance on fixed mechanical limits or manual alignment, and thus adapts to the diverse needs of different document formats and stamping positions, achieving sub-millimeter-level high-precision positioning and stamping. By setting up a "multi-dimensional seal quality analysis" step after stamping, the ability to monitor and ensure the quality of the final output can be effectively improved. This helps to promptly identify and eliminate unqualified products such as blurry seals and misaligned seals, thereby ensuring that the seal quality of each output document meets the standards and achieving closed-loop quality control throughout the entire process. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1This is a schematic flowchart of a control method for an automated stamping device that can be cascaded and expanded according to an embodiment of the present invention. Figure 2 This is a schematic diagram of a control scenario for an automated stamping device that can be cascaded and expanded according to an embodiment of the present invention. Figure 3 This is a schematic diagram of the structure of a control device for a cascaded and expandable automated stamping device disclosed in an embodiment of the present invention; Figure 4 This is a schematic diagram of the control device of another cascaded and expandable automated stamping device disclosed in an embodiment of the present invention. Detailed Implementation

[0025] To enable those skilled in the art to better understand the present invention, 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 embodiments of the present invention, and not all embodiments. 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.

[0026] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or end that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or ends.

[0027] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0028] This invention discloses a control method and apparatus for an automated stamping device that can be cascaded and expanded. By constructing a complete automated process loop of "transmission-content verification-positioning stamping-quality review," the continuity and automation of stamping operations can be significantly improved. This frees manual labor from repetitive and error-prone tasks such as document delivery, verification, alignment, and inspection, thereby reducing labor costs and subjective risks of operational errors, and achieving highly efficient and reliable fully automated stamping operations. By introducing "document content recognition and verification" as a pre-stamping decision-making step, the accuracy and security of stamping operations can be effectively improved. This helps to prevent stamping errors caused by document content errors (such as text errors or unexpected documents) from the source, thus ensuring the legal validity and seriousness of the seal and achieving intelligent risk prevention and control. By employing a programmable, adaptive positioning method that generates control parameters based on the identified stamping area, the flexibility, accuracy, and consistency of the stamping position can be significantly improved. This helps to eliminate reliance on fixed mechanical limits or manual alignment, and thus better adapt to diverse needs for different document formats and stamping positions, achieving sub-millimeter-level high-precision stamping. By incorporating a "multi-dimensional imprint quality analysis" step after stamping, the ability to monitor and guarantee the final output quality can be effectively improved. This facilitates the timely detection and removal of substandard products such as blurred imprints and positional misalignments, ensuring that the stamping quality of each document meets standards and achieving closed-loop quality control throughout the entire process. These will be explained in detail below.

[0029] Example 1 Please see Figure 1 , Figure 1 This is a flowchart illustrating a control method for a cascaded and expandable automated stamping device disclosed in an embodiment of the present invention. Figure 1 The described control method for the cascaded and expandable automated stamping device can be applied to the automated stamping device itself, and also to intelligent devices associated with the automated stamping device. These intelligent devices include, but are not limited to, one or more of the following: switching devices, cloud devices, edge computing devices, relay devices, base station devices, urban management devices, and intelligent connected devices. This invention does not limit the scope of these applications. Figure 1 As shown, the control method for this cascaded and expandable automated stamping device may include the following operations: 101. Control the document to be transported along a preset path in the automated stamping device; In this embodiment of the invention, step 101 can optionally be implemented by synchronous belt drive, roller friction drive, or belt-line drive. The main control unit (such as a PLC, industrial PC, or embedded controller) precisely controls the conveying speed by adjusting the speed or pulse frequency of the drive motor. Key workstations along the path (such as the identification area, stamping area, and review area) are equipped with position detection sensors (such as through-beam photoelectric sensors, diffuse reflection sensors, or microswitches) to provide trigger signals to the main control unit indicating document arrival.

[0030] 102. When a document is detected arriving at the document content recognition and verification area of ​​the automated stamping device, determine whether the document matches the pre-stored standard information; In this embodiment of the invention, step 102 is optionally a key decision node. This "matching" refers not only to complete consistency of textual content but can also extend to format compliance checks (such as whether the reserved position for the seal is correct) and logical verification (such as whether the contract amount is consistent in capitalization and numerals). The matching logic can be calculated locally by the main control unit, or the information can be encrypted and sent to a cloud server for more complex verification before receiving the verification results.

[0031] In this embodiment of the invention, as an optional implementation, the above-mentioned determination of whether the document matches the pre-stored standard information includes: Get the complete image of the document; Perform preprocessing operations on the complete image, including at least one of color separation, distortion correction, and noise filtering. The preprocessed complete image is input into a preset text recognition engine to identify and extract the full text key information of the document. The full text key information includes at least one of the document's title, number, body text, and icons. The key information in the full text is compared with the pre-stored standard information to calculate the target similarity value between the key information in the full text and the pre-stored standard information. Determine whether the target similarity value is greater than or equal to the preset similarity threshold. If the target similarity value is greater than or equal to the preset similarity threshold, then the document is determined to match the pre-stored standard information.

[0032] In this embodiment of the invention, optionally, the aforementioned "color separation operation" can be designed to eliminate interference. Besides converting from RGB to HSV color space for red ink region segmentation, for blue ink marks or document background colors, threshold segmentation of specific channels can be performed in the Lab color space. This isolates noise at its source, improving the signal-to-noise ratio of subsequent OCR processing.

[0033] Alternatively, the aforementioned "distortion correction operation" aims to eliminate geometric deformation. Besides Thin Plate Spline Transform (TPS), affine transformation can also be used for uniform deformation. Its key function is to "straighten" non-planar document images, restoring their true size and proportions, which is crucial for subsequent accurate stamping. In implementation, document edges or pre-pasted positioning marker corners can be detected as reference points for the transformation.

[0034] Further, optionally, the "noise filtering operation" described above aims to optimize image quality. Median filtering effectively removes salt-and-pepper noise while preserving edges, while Gaussian filtering excels at smoothing Gaussian noise. The filtering algorithm and its kernel size can be adaptively selected based on the primary noise type of the document image (such as paper texture or ink spots), achieving a balance between noise reduction and maintaining text sharpness.

[0035] Alternatively, the "preset text recognition engine" mentioned above is not limited to Tesseract; it can also integrate engines such as PaddleOCR and Tr. The process of "recognizing and extracting...key information from the whole text" can be further configured as follows: first, perform layout analysis to identify different areas such as titles, body text, and tables; then, perform OCR on specific areas to improve recognition efficiency and accuracy.

[0036] Alternatively, this method achieves full-process content control through "preprocessing for anti-interference + pre-stamp verification," with the specific steps as follows: Phase 1: Pre-verification of content before stamping: (1) Image acquisition and preprocessing: (1.1) When the document to be stamped is sent to the recognition area by the conveying module, the infrared beam sensor triggers a signal, the PLC controls the ring light to turn on (illuminance 1000 lux), and the 2-megapixel camera captures a full-frame image of the document (resolution 1920×1080, frame rate 30fps). (1.2) The edge computing module performs preprocessing on the acquired images: Color separation: Convert the RGB image to HSV space, segment the red ink area according to the threshold (H: 0-10, S: 50-255, V: 50-255), and mark it as "interference area"; Distortion correction: The Thin Plate Spline Transform (TPS) algorithm is used, and four corner points of the document are selected as references to perform geometric correction on the deformed area of ​​the curved surface (stretch rate ≤15%). Noise filtering: Eliminate noise caused by paper wrinkles through median filtering (3×3 template) while preserving clear text edges.

[0037] (2) OCR comparison and result judgment: (2.1) Input the preprocessed image into the Tesseract 5.0 OCR text recognition model to extract key elements such as title, number, and content; (2.2) The edge computing module compares the extracted text with the "standard information database" (document information of each batch uploaded and entered by the user in advance through the OA system) pre-stored in the main control unit: If the similarity is ≥99%, send a "pass" signal to the PLC to allow the process to proceed to the next stage. If the similarity is less than 99%, a "failure" signal is sent, the PLC controls the conveyor module to import the document into the waste bin, and displays a "content mismatch" alarm on the touch screen.

[0038] As can be seen, implementing this optional embodiment, through the operation of "acquiring and preprocessing the complete image of the document," can significantly improve the accuracy and robustness of subsequent text recognition. This helps eliminate image quality interference caused by uneven lighting, paper wrinkles, and existing imprints, thus laying a solid foundation for high-precision text information extraction and enabling reliable image acquisition in complex real-world scenarios. By "identifying and extracting key information from the entire document text," the comprehensiveness and depth of content verification can be greatly improved. This facilitates the verification of all elements such as document titles, body text, and numbering, and is more effective than partial verification in preventing risks such as content tampering and misattribution. This, in turn, enhances the security level of the entire stamping process, enabling intelligent verification based on full-text understanding. By "comparing the similarity between the key information in the entire text and pre-stored standard information," the objectivity and accuracy of verification decisions can be effectively improved. This helps transform vague manual judgments into precise quantitative indicator judgments, thereby reducing misjudgments or omissions caused by subjective factors and achieving data-driven, repeatable, automated review.

[0039] 103. If a match is found, the document is sent to the positioning and stamping area of ​​the automated stamping device; 104. When a document is detected to have arrived at the positioning and stamping area, stamping control parameters for each stamp are generated based on each stamping area pre-marked in the document and the corresponding cascaded stamps in the automated stamping device. These parameters control the stamps to perform the corresponding stamping control operations. The stamping control parameters include movement control parameters and stamping pressure control parameters. In this embodiment of the invention, step 104 above optionally embodies the modularity and synergy of the system. The main control unit needs to maintain a "seal-position" mapping table. When it is recognized that a document needs to be stamped with multiple seals (such as "official seal" and "financial seal"), the main control unit will send instructions to the corresponding "cascaded seal" modules sequentially or in parallel (if there are multiple independent motion mechanisms). The target coordinates contained in the instructions are the final execution coordinates after coordinate transformation and offset compensation.

[0040] Alternatively, for the aforementioned cascading expansion method for seals (solving the problem of "fixed seals and difficult expansion"), this method achieves flexible addition and removal of multiple seals through a "modular + standardized" design logic. The specific steps are as follows: Basic unit installation: Weld L-shaped metal brackets to both ends of the bottom of the telescopic rod to fix the basic telescopic rod unit to the top of the stamping machine body (locked with expansion screws to ensure parallelism with the conveyor table); fix the basic stamp unit to the reference position of the telescopic rod of the equipment with M4 screws to ensure that the bottom of the unit is parallel to the conveyor table (parallelism error ≤0.1mm); connect the power cord (DC24V) and signal control line of the basic unit to the IO interface of the main control unit PLC.

[0041] Cascading of expansion units: (1) Install a manual knob locking assembly on the side of the expansion unit slider. The threaded pin matches the positioning hole of the guide rail groove for secondary reinforcement after the elastic buckle locks (to prevent displacement caused by high-speed stamping vibration); (2) Fix the expansion stamp unit to the reference position of the equipment telescopic rod with M4 screws to ensure that the bottom of the unit is parallel to the conveyor table (parallelism error ≤ 0.1mm); connect the power line (DC24V) and signal control line of the expansion unit to the IO interface of the main control unit PLC; (3) If multiple expansion units need to be added, repeat the above steps.

[0042] Independent drive component integration: (1) Electric push rod installation: A miniature electric push rod is fixed above the stamp holder of each telescopic rod unit (basic + extension) by a bracket. The push rod output shaft is aligned with the center of the top of the stamp holder (to ensure that the thrust is applied vertically to the stamp); (2) Hall sensor assembly: A Hall position sensor (accuracy ±0.01mm) is attached to the side of the electric push rod housing. The sensor probe is aligned with the magnetic ring on the push rod output shaft to detect the stamp lifting position in real time (feedback to the main control unit to avoid excessive pressing); (3) Internal wiring connection: The electric push rod power line (DC24V), control line (PWM signal) and Hall sensor signal line are connected to the flexible pin of the extension unit (by soldering through terminal blocks to ensure a secure connection).

[0043] Unit Identification and Configuration: Upon powering on the main control unit, the PLC automatically scans the identification codes of the cascaded units via the MODBUS protocol (each seal unit has a preset unique ID, corresponding to types such as official seal / financial seal). The scanning results are displayed on the "Seal Management" interface on the touch screen. The user selects the seals to be enabled on the interface (such as "official seal + financial seal") and clicks "Save Configuration". The PLC automatically allocates the drive signal channels for each unit without the need for additional mechanical position calibration.

[0044] In this embodiment of the invention, as another optional implementation, the above-mentioned method of generating stamping control parameters for each stamp based on each pre-marked stamping area in the identified document and the corresponding cascaded stamps in the automated stamping device, to control the stamps to perform corresponding stamping control operations, includes: For each pre-marked stamped area in the document, the region coordinates of the stamped area are extracted according to the preset image segmentation and contour extraction algorithm; Obtain the current position coordinates of the cascaded stamps in the automated stamping device corresponding to the stamping area; Based on the current position coordinates and area coordinates, stamping control parameters are generated for each stamp to control the stamp to perform the corresponding stamping control operation. The movement control parameters include horizontal movement control parameters and vertical movement control parameters.

[0045] In this embodiment of the invention, optionally, the aforementioned "preset image segmentation and contour extraction algorithm" is the basis for coordinate acquisition. Image segmentation can employ thresholding (such as Otsu's method), edge detection (such as the Canny operator), or region growing to separate the stamped area or reference mark from the background. Contour extraction commonly uses algorithms such as Suzuki85 to track contours via chaincode and may perform contour approximation to fit geometric shapes such as circles and rectangles, thereby accurately calculating the pixel coordinates of their center points or feature points.

[0046] Alternatively, the acquisition of the aforementioned "current position coordinates of the cascaded stamps" relies on the feedback system of the motion mechanism. For open-loop controlled stepper motors, the current position is calculated by accumulating the number of pulses issued by the main control unit; for closed-loop controlled servo motors or motors equipped with encoders, the actual position with higher precision is obtained by directly reading the encoder feedback value. Each cascaded stamp module should have its own independent coordinate origin (mechanical origin or reference point set through a homing operation).

[0047] Optionally, the aforementioned "generation of stamping control parameters" is the core of motion control. Motion control parameters (pulse count, direction, speed curve) are typically calculated and generated by the positioning module of the motion control card or PLC based on the difference between the target position and the current position, combined with preset parameters such as acceleration, deceleration, and maximum speed. Stamping pressure control parameters may be converted into settings for the energizing time of the cylinder solenoid valve, the current threshold of the electric actuator, or the output torque of the servo motor, ensuring appropriate pressure that is clear without damaging the paper.

[0048] Alternatively, this method achieves automatic and accurate stamping through "visual positioning - precision driving - vibration reduction execution," with the specific steps as follows: (1) Positioning Mark Recognition: After the document passes content verification, it is transported to the stamping area. The PLC controls the 1-megapixel 3D vision sensor to start and capture the circular positioning mark (5mm in diameter, with a black border) on the surface of the document: (1.1) The sensor identifies the center of the marked circle through the "threshold segmentation + contour extraction" algorithm, and calculates the X / Y axis coordinates of the center of the circle in the device coordinate system (accuracy ±0.05mm) by combining the previously calibrated camera intrinsic parameters (focal length, pixel size). (1.2) If the positioning mark is not identified (e.g., the mark is missing), the sensor sends a "positioning failure" signal, the PLC control document imports the waste box and alarms.

[0049] (2) Target position calculation: The PLC automatically calculates the target stamping coordinates of each activated stamp based on the user's preset "stamp-mark position".

[0050] (3) Precision drive and stamping execution: (3.1) The PLC sends pulse signals to the stepper motor driver of the X / Y axis linear guide rail. The 42 stepper motor (step angle 1.8°, 6400 microsteps) drives the stamp unit to move along the X axis to the target X coordinate, and then along the Y axis to the target Y coordinate, with a displacement accuracy of 0.01mm. During the movement, the motor encoder provides real-time position signals. The PLC compares the actual position with the target coordinates. If the deviation is greater than 0.05mm, fine-tuning is triggered to ensure that the final positioning deviation is less than or equal to 0.1mm. (3.2) After positioning is completed, the PLC controls the micro cylinder of the stamp unit to press down (stroke 10mm, pressure 0.3MPa), the stamp contacts the document to complete the stamping, and then the cylinder resets (response time ≤0.1s).

[0051] (4) Vibration reduction and stabilization: During the stamping process, the honeycomb aluminum vibration damping base (damping coefficient 0.3) absorbs the vibration generated by the motor operation and cylinder action, and controls the amplitude to below 0.02mm to avoid the printed text being tilted or blurred.

[0052] As can be seen, implementing this optional embodiment, through the refined processing logic of "extracting the coordinates of each stamping area," can significantly improve its adaptability to complex stamping needs (such as multiple stamps for a single document or irregular stamping positions). This facilitates independent positioning and precise operation of each target area, thereby meeting the increasingly diverse and personalized stamping needs in modern offices and achieving one-to-one precise stamping control. The feedback control mechanism of "generating control parameters based on the current position coordinates of the stamp and the target area coordinates" effectively improves the accuracy and dynamic performance of stamp motion positioning. This facilitates rapid, smooth, and precise movement of the stamp from any current position to the target position, thereby reducing overshoot and oscillation during the positioning process and achieving stable and efficient precision motion control.

[0053] 105. After the stamping control operation is completed, transfer the document to the stamping review area; 106. Perform multi-dimensional seal quality analysis on the document to determine whether the document meets the preset seal quality conditions; In this embodiment of the invention, optionally, the aforementioned "performing multi-dimensional print quality analysis on the document" is the final step in the quality closed-loop control. This analysis is "multi-dimensional," and in practice, in addition to clarity and position, it can be extended to print color saturation analysis, print integrity (whether there are defects), etc., forming a comprehensive quality scoring system.

[0054] In another optional implementation of this invention, the above-mentioned multi-dimensional seal quality analysis of the document to determine whether the document meets the preset stamping quality conditions includes: Obtain a partial image of the actual stamped area of ​​the document within the stamped review area; Simultaneously perform text sharpness analysis and text position compliance analysis on local images; text sharpness analysis includes calculating the gradient features of text edges in the local image to obtain sharpness values; text position compliance analysis includes matching the text outline in the local image with the feature points of a pre-stored standard text outline to calculate the position offset. When the clarity value meets the preset clarity standard and the position offset is less than or equal to the preset tolerance threshold, the document is determined to meet the preset stamping quality conditions.

[0055] In this embodiment of the invention, optionally, the essence of the above-mentioned "print sharpness analysis" is to measure the sharpness of edges in an image. Calculating gradient features is a common method; for example, convolving the print region with the Laplacian operator, the variance of the response is often used as the sharpness value. The higher the variance value, the sharper the image. This method is very sensitive to defects such as print blurring and ink bleeding.

[0056] Optionally, the key to the aforementioned "imprint position compliance analysis" is image registration. The acquired imprint image is matched with a standard template image using "feature point matching." Commonly used feature points include corner points (such as Harris corner points), edge feature points (such as FAST), or more stable SIFT, SURF, and ORB feature descriptors. The transformation matrix between the two images (such as rigid body transformation or affine transformation) is estimated using methods such as the Random Sample Consensus Algorithm (RANSAC), thereby accurately calculating the translation and rotation offset of the imprint center. The "clarity standard" and "tolerance threshold" are not fixed values, but rather configuration parameters that can be adaptively adjusted according to different seal types (higher requirements for official seals, slightly lower requirements for signature seals) and different document importance levels. These parameters are stored in the system configuration file and are called by the main control unit during initialization.

[0057] Alternatively, after the document is stamped and transported to the review area, a signal can be triggered by the diffuse reflection sensor, and the camera can take another partial image of the stamped area (focusing on the printed area, resolution 1280×720).

[0058] (1) The edge computing module performs the following on the printed image: Clarity test: Calculate the gradient value of the printed edge (gradient value ≥ 20 is acceptable, indicating that the lines are clear and not blurry); Location compliance check: The offset between the center of the print and the preset positioning mark in the document is compared using the ORB feature matching algorithm (a deviation of ≤0.5mm is considered compliant). (2) If both indicators are qualified, it is judged as "finished product" and transported to the finished product collection box; if either indicator is unqualified, it is judged as "defective product" and imported into the waste box. At the same time, the fault type (such as "blurred printing") is recorded in the PLC log.

[0059] As can be seen, implementing this optional embodiment, through multi-dimensional quality inspection of "simultaneous analysis of clarity and positional compliance of the imprint," can significantly improve the comprehensiveness and scientific nature of quality assessment. This facilitates a comprehensive evaluation of stamping quality from two key dimensions: the recognizability of the imprint itself (clarity) and its relative relationship with the document (position). Consequently, it leads to more comprehensive and accurate conclusions of pass / fail, enabling refined quality grading and control. The quantitative evaluation method of "calculating the edge gradient features of the imprint to obtain a clarity value" effectively improves the objectivity and consistency of clarity judgment, thus avoiding fluctuations and disputes caused by subjective human judgment. This, in turn, helps establish stable and reliable quality inspection standards, achieving standardized inspection of imprint imaging quality. By "matching feature points between the actual imprint outline and the standard outline to calculate the offset," the accuracy and automation level of positional deviation detection can be significantly improved. This allows for precise quantification of the minute differences between the actual and ideal positions of the imprint, providing accurate data support for process improvement and equipment calibration, and achieving precise positional measurement based on machine vision.

[0060] 107. If satisfied, the document will be sent to the document collection device.

[0061] This invention embodiment defines a four-level core process for automated stamping control. Refer to the embodiments disclosed in this invention. Figure 2 , Figure 2 This is a schematic diagram illustrating a control scenario for a cascaded and expandable automated stamping device disclosed in an embodiment of the present invention. The specific implementation is as follows: After the device is powered on and initialized, the main control unit (e.g., a programmable logic controller (PLC) or an industrial computer) starts the conveyor belt motor, causing the document to be conveyed along a straight path at a constant speed (e.g., 10 cm / s). When the document blocks the beam of the photoelectric sensor installed at the content recognition station, the PLC determines that the document has reached the recognition area and then triggers the industrial camera (high-definition camera) installed above the station to capture a complete image of the document. This image is sent to the processing unit for content comparison. If the comparison passes, the PLC controls the conveyor belt to continue running until the document reaches the positioning stamping area. At this time, another photoelectric sensor is triggered, and the PLC, based on the stamping area coordinates identified by the vision system, controls the corresponding stamp module to move to the target position and press down to complete the stamping. After stamping, the document is continued to be conveyed to the review area, where another vision system inspects the stamp quality. Qualified products are finally guided to the finished product collection box. The coordinated scheduling of the above-mentioned conveying, recognition, positioning, and review stations is uniformly controlled by the PLC to ensure a smooth process.

[0062] As can be seen, implementing the embodiments of the present invention, by constructing a complete automated closed-loop process of "transmission-content verification-positioning and stamping-quality review," can significantly improve the continuity and automation of stamping operations. This frees manual labor from repetitive and error-prone tasks such as document delivery, verification, alignment, and inspection, thereby reducing labor costs and subjective risks of operational errors, and achieving highly efficient and reliable fully automated stamping operations. By introducing "document content recognition and verification" as a pre-decision-making step for stamping, the accuracy and security of stamping operations can be effectively improved, thus preventing stamping errors caused by document content errors (such as text errors or unexpected documents) from the source, thereby ensuring the legal validity and seriousness of the seal and achieving intelligent risk prevention and control. By adopting a programmable and adaptive positioning method that "generates control parameters based on the identified stamping area," the flexibility, accuracy, and consistency of the stamping position can be significantly improved. This helps to eliminate reliance on fixed mechanical limits or manual alignment, and thus adapts to the diverse needs of different document formats and stamping positions, achieving sub-millimeter-level high-precision positioning and stamping. By setting up a "multi-dimensional seal quality analysis" step after stamping, the ability to monitor and ensure the quality of the final output can be effectively improved. This helps to promptly identify and eliminate unqualified products such as blurry seals and misaligned seals, thereby ensuring that the seal quality of each output document meets the standards and achieving closed-loop quality control throughout the entire process.

[0063] In this optional embodiment, as an optional implementation, the above-mentioned comparison of the full-text key information with pre-stored standard information to calculate the target similarity value between the full-text key information and the pre-stored standard information includes: Semantic similarity calculation is performed between the key information in the full text and the corresponding standard text in the pre-stored standard information. The semantic similarity calculation includes extracting the semantic feature vectors of the key information in the full text and the standard text respectively, and calculating the cosine similarity between the two semantic feature vectors as the target similarity value.

[0064] In this embodiment of the invention, optionally, the "semantic similarity calculation" described above aims to understand the deeper meaning of the text. First, "extracting semantic feature vectors" is a crucial step. Traditional methods may use a bag-of-words model combined with TF-IDF statistics, while modern deep learning methods use pre-trained language models (such as BERT and ERNIE). These models can convert a piece of text into a high-dimensional, dense vector (i.e., an embedding vector), whose position and orientation in space contain the semantic information of the text.

[0065] Alternatively, the above-mentioned "calculation of cosine similarity between two semantic feature vectors" can be further explored. Cosine similarity focuses on the direction of the vectors rather than their magnitude (length), thus it is insensitive to text length and better measures semantic consistency. The result is a scalar between -1 and 1; the closer the value is to 1, the more consistent the directions of the two texts in the semantic space, i.e., the more semantically similar they are. This method can effectively handle situations where there is a literal mismatch but the meaning is consistent, such as synonym substitution and word order adjustment.

[0066] It is evident that implementing this optional embodiment, by "performing semantic similarity calculation" rather than simple literal matching, can greatly improve the intelligence and fault tolerance of content comparison, thereby facilitating the understanding of the core meaning of the text, effectively handling differences in expression such as synonym replacement and word order adjustment, and thus helping to reduce unnecessary verification failures while ensuring the substantive consistency of the content, achieving a more intelligent and humanized document content verification.

[0067] In this optional embodiment, as another optional implementation, the above-described method of extracting the region coordinates of each pre-marked stamped region in the document according to a preset image segmentation and contour extraction algorithm includes: The image coordinates of pre-set reference marks on the document are identified using a visual sensor; The image coordinates of the reference mark are combined with pre-calibrated visual perception parameters to convert them into world coordinates in the coordinate system of the stamping device; The regional coordinates of the stamping area are calculated based on the world coordinates and the pre-defined offset of the stamping area relative to the reference mark.

[0068] In this embodiment of the invention, optionally, the above-mentioned "image coordinates of the reference marker identified by the visual sensor" is the first step. The reference marker can be designed as a special pattern (such as AprilTag, ArUco code) to facilitate stable and rapid detection and recognition. The recognition algorithm typically includes image binarization, contour finding, perspective transformation correction, and then decoding or fitting to calculate the image coordinates of the marker center with sub-pixel accuracy.

[0069] Optionally, the aforementioned "pre-calibrated visual perception parameters" form the mathematical model foundation for coordinate transformation. Camera calibration involves photographing a specific calibration board (such as a checkerboard) to solve for the camera's intrinsic parameters (such as focal length and distortion coefficients) and extrinsic parameters (such as rotation matrices and translation vectors) relative to the world coordinate system. This process establishes a precise mathematical mapping between image pixels and world coordinate points.

[0070] Alternatively, the "conversion to world coordinates" mentioned above involves coordinate back-projection using a camera model (such as a pinhole camera). Using the calibrated intrinsic and extrinsic parameters, the two-dimensional image coordinates are transformed into a three-dimensional world coordinate system by solving linear or nonlinear equations. Subsequently, by superimposing a "pre-defined offset of the stamping area relative to the reference mark" (a fixed translation in the X, Y, Z directions), the absolute target coordinates in the machine coordinate system that the stamping actuator needs to reach are finally obtained.

[0071] As can be seen, implementing this optional embodiment, by "converting the image coordinates of the reference marker to world coordinates," can significantly improve the accuracy and adaptability of the positioning system. This facilitates the precise correlation between visual information based on two-dimensional images and the three-dimensional physical execution space, thereby overcoming errors caused by camera perspective, lens distortion, etc., and achieving accurate mapping from image perception to physical operation. By "calculating the stamping area coordinates based on world coordinates and preset offsets," the flexibility and configurability of the positioning operation can be effectively improved. This allows for quick adaptation to the stamping position requirements of different document templates by simply modifying the offset parameters, thereby enhancing the device's rapid response capability to different business scenarios and achieving flexible stamping position settings.

[0072] In an optional embodiment, the method further includes: Based on the processing status of the document content recognition and verification area and the ready status of the positioning and stamping area, the start and stop of the conveying device in the automated stamping device are dynamically controlled so that the document is paused when it arrives at the corresponding area until the processing of that area is completed. When a sensor in the automated stamping device detects an abnormal document delivery, the device controls a diversion mechanism to guide the abnormal document to an independent collection path and triggers an alarm signal.

[0073] In this embodiment of the invention, optionally, the "dynamic control of the start and stop of the conveyor device" is key to optimizing the production line cycle time. The main control unit monitors the "processing status" (e.g., "recognizing", "recognizing completed") and "ready status" (e.g., "stamping mechanism idle", "stamping mechanism busy") of each workstation in real time. Based on this status information, the main control unit implements an "event-driven" conveyor control strategy. For example, the conveyor is only started to send the document to the stamping station when the recognition station is complete and the stamping station is idle; otherwise, the document waits at its current position. This dynamic pause ensures sufficient processing time for each process, avoiding congestion or document strain.

[0074] Optionally, the handling mechanism for "sensor-detected document transmission anomalies" ensures system robustness. Anomalies include paper jams, document skew, and multiple overlapping sheets. Timeout signals from sensors (such as photoelectric sensors and ultrasonic sensors) trigger the anomaly handling routine. The main control unit immediately stops the conveyor belt and controls the diversion mechanism (such as a servo-driven baffle or an electromagnet-controlled flip-up plate) to change the document path, directing it to an "independent collection path" (such as a waste bin). Simultaneously, it notifies the operator by triggering an alarm signal (such as a warning box popping up on the touchscreen or an audible and visual alarm sounding). The main control unit records an anomaly log (such as the time of occurrence and the location of the anomaly) to provide data support for maintenance.

[0075] In this embodiment of the invention, optionally, the method can dynamically coordinate the timing of each module through a PLC to achieve full automation of the "paper feeding-identification-positioning-stamping-diversion" process. The specific steps are as follows: (1) Initialization and parameter settings: User settings via touchscreen: Conveying speed (5-20cm / s, default 10cm / s); Enable the use of seal types (such as "official seal + financial seal"); After clicking "Start", the PLC sends initialization signals to each module and checks whether the sensors, motors and cylinders are normal (if abnormal, an alarm will be triggered).

[0076] (2) Automatic paper feeding and conveying: The manual staff stacks the documents to be stamped on the feeding table (up to 500 documents). The PLC controls the synchronous belt motor to start, and the conveyor wheels feed the documents one by one into the conveyor channel (with a 10cm gap between each document to avoid overlapping). An infrared sensor at the paper feed end detects the document spacing in real time. If the spacing is less than 5cm, the PLC reduces the motor speed (e.g., from 10cm / s to 5cm / s) to ensure that each sheet is fed independently.

[0077] (3) Multi-module timing coordination: When the document reaches the content recognition area, the PLC controls the conveyor motor to pause for 0.5 seconds (matching the OCR processing time), and resume conveying after the comparison result is returned; When the document reaches the positioning and stamping area, the PLC controls the conveyor motor to pause for 1.2 seconds (0.5 seconds for matching and positioning + 0.7 seconds for stamping), and the conveyor restarts after the stamping is completed; If a certain step of the process is delayed (e.g., OCR takes > 0.5s), the PLC will automatically extend the pause time to avoid document accumulation or omissions.

[0078] (4) Fault handling and traffic diversion: If the diffuse reflection sensor in the conveyor channel detects a paper jam (document dwell time > 3s), the PLC will immediately stop the machine, control the electromagnet of the waste bin baffle to be energized (switch the conveyor path), and at the same time the touch screen will display "Paper jam fault". After manual cleaning, click "Reset" to restart. If content recognition or positioning fails, the PLC controls the document to be imported into the waste box before it reaches the discharge end, while qualified documents enter the finished product collection box, thus achieving automatic diversion.

[0079] (5) Batch processing completed: When the feed station sensor detects no documents, the PLC continues to run until the remaining documents in the conveyor channel are processed, then automatically stops and displays "Processing Complete" and statistical data (total number of documents, number of qualified documents, number of defective documents) on the touch screen.

[0080] As can be seen, implementing this optional embodiment, by "dynamically controlling the start and stop of the conveyor device according to the processing status of each area," can significantly improve the work efficiency and coordination of the entire production line. This helps ensure sufficient processing time at each stage, avoids document congestion or equipment idleness, and thus optimizes production cycle time, increases the throughput per unit time, and achieves efficient and orderly collaborative operation between modules. By "automatically diverting and alarming when document transmission abnormalities are detected," the reliability and autonomy of the system can be greatly improved. This facilitates rapid response and isolation of problems such as paper jams and overlaps, preventing the escalation of faults or damage to documents. Consequently, it helps ensure that the qualified product process is not affected, and quickly notifies personnel to intervene, achieving intelligent fault diagnosis and handling, and ensuring the continuity and safety of production.

[0081] Example 2 Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a control device for a cascaded and expandable automated stamping device disclosed in an embodiment of the present invention. The control device for this cascaded and expandable automated stamping device can be applied to the automated stamping device itself, and can also be applied to intelligent devices associated with the automated stamping device. These intelligent devices include, but are not limited to, one or more of the following: switching devices, cloud devices, edge computing devices, relay devices, base station devices, urban management devices, and intelligent connected devices. The present invention does not limit the application to these devices. Figure 3 As shown, the control device for this cascaded and expandable automated stamping device may include: The transmission control module 201 is used to control the transmission of documents along a preset path in the automated stamping device; The detection and judgment module 202 is used to determine whether the document matches the pre-stored standard information when the document is detected to have been transmitted to the document content recognition and verification area of ​​the automated stamping device. The transmission control module 201 is also used to transmit the document to the positioning and stamping area of ​​the automated stamping device if a match is found. The detection control module 203 is used to generate stamping control parameters for each stamp based on each stamping area pre-marked in the document and the corresponding cascaded stamps in the automated stamping device when the document is detected to have been delivered to the positioning stamping area. The stamping control parameters include movement control parameters and stamping pressure control parameters. The transmission control module 201 is also used to transmit the document to the stamping review area after the stamping control operation is completed. The detection and judgment module 202 is also used to perform multi-dimensional seal quality analysis on the document to determine whether the document meets the preset stamping quality conditions. The transmission control module 201 is also used to transmit the document to the document collection device if the conditions are met.

[0082] It is evident that implementing this invention, by constructing a complete automated closed-loop process of "transmission-content verification-positioning and stamping-quality review," can significantly improve the continuity and automation of stamping operations. This frees manual labor from repetitive and error-prone tasks such as document delivery, verification, alignment, and inspection, thereby reducing labor costs and the subjective risk of operational errors, achieving highly efficient and reliable fully automated stamping operations. By introducing "document content recognition and verification" as a pre-stamping decision-making step, the accuracy and security of stamping operations can be effectively improved, thus preventing stamping errors caused by document content errors (such as text errors or unexpected documents) from the source, thereby ensuring the legal validity and seriousness of the seal and achieving intelligent risk control. By adopting a programmable and adaptive positioning method that "generates control parameters based on the identified stamping area," the flexibility, accuracy, and consistency of the stamping position can be significantly improved. This helps to eliminate reliance on fixed mechanical limits or manual alignment, and thus adapts to the diverse needs of different document formats and stamping positions, achieving sub-millimeter-level high-precision positioning and stamping. By setting up a "multi-dimensional seal quality analysis" step after stamping, the ability to monitor and ensure the quality of the final output can be effectively improved. This helps to promptly identify and eliminate unqualified products such as blurry seals and misaligned seals, thereby ensuring that the seal quality of each output document meets the standards and achieving closed-loop quality control throughout the entire process.

[0083] In this embodiment of the invention, as an optional implementation, the specific method by which the detection and judgment module 202 determines whether a document matches pre-stored standard information includes: Get the complete image of the document; Perform preprocessing operations on the complete image, including at least one of color separation, distortion correction, and noise filtering. The preprocessed complete image is input into a preset text recognition engine to identify and extract the full text key information of the document. The full text key information includes at least one of the document's title, number, body text, and icons. The key information in the full text is compared with the pre-stored standard information to calculate the target similarity value between the key information in the full text and the pre-stored standard information. Determine whether the target similarity value is greater than or equal to the preset similarity threshold. If the target similarity value is greater than or equal to the preset similarity threshold, then the document is determined to match the pre-stored standard information.

[0084] As can be seen, implementing this optional embodiment, through the operation of "acquiring and preprocessing the complete image of the document," can significantly improve the accuracy and robustness of subsequent text recognition. This helps eliminate image quality interference caused by uneven lighting, paper wrinkles, and existing imprints, thus laying a solid foundation for high-precision text information extraction and enabling reliable image acquisition in complex real-world scenarios. By "identifying and extracting key information from the entire document text," the comprehensiveness and depth of content verification can be greatly improved. This facilitates the verification of all elements such as document titles, body text, and numbering, and is more effective than partial verification in preventing risks such as content tampering and misattribution. This, in turn, enhances the security level of the entire stamping process, enabling intelligent verification based on full-text understanding. By "comparing the similarity between the key information in the entire text and pre-stored standard information," the objectivity and accuracy of verification decisions can be effectively improved. This helps transform vague manual judgments into precise quantitative indicator judgments, thereby reducing misjudgments or omissions caused by subjective factors and achieving data-driven, repeatable, automated review.

[0085] In this embodiment of the invention, as another optional implementation, the detection control module 203 generates stamping control parameters for each stamp based on each pre-marked stamping area in the identified document and the corresponding cascaded stamps in the automated stamping device, in order to control the stamps to perform the corresponding stamping control operations. The specific methods include: For each pre-marked stamped area in the document, the region coordinates of the stamped area are extracted according to the preset image segmentation and contour extraction algorithm; Obtain the current position coordinates of the cascaded stamps in the automated stamping device corresponding to the stamping area; Based on the current position coordinates and area coordinates, stamping control parameters are generated for each stamp to control the stamp to perform the corresponding stamping control operation. The movement control parameters include horizontal movement control parameters and vertical movement control parameters.

[0086] As can be seen, implementing this optional embodiment, through the refined processing logic of "extracting the coordinates of each stamping area," can significantly improve its adaptability to complex stamping needs (such as multiple stamps for a single document or irregular stamping positions). This facilitates independent positioning and precise operation of each target area, thereby meeting the increasingly diverse and personalized stamping needs in modern offices and achieving one-to-one precise stamping control. The feedback control mechanism of "generating control parameters based on the current position coordinates of the stamp and the target area coordinates" effectively improves the accuracy and dynamic performance of stamp motion positioning. This facilitates rapid, smooth, and precise movement of the stamp from any current position to the target position, thereby reducing overshoot and oscillation during the positioning process and achieving stable and efficient precision motion control.

[0087] In this embodiment of the invention, as another optional implementation, the detection and judgment module 202 performs multi-dimensional seal quality analysis on the document to determine whether the document meets the preset stamping quality conditions. The specific methods include: Obtain a partial image of the actual stamped area of ​​the document within the stamped review area; Simultaneously perform text sharpness analysis and text position compliance analysis on local images; text sharpness analysis includes calculating the gradient features of text edges in the local image to obtain sharpness values; text position compliance analysis includes matching the text outline in the local image with the feature points of a pre-stored standard text outline to calculate the position offset. When the clarity value meets the preset clarity standard and the position offset is less than or equal to the preset tolerance threshold, the document is determined to meet the preset stamping quality conditions.

[0088] As can be seen, implementing this optional embodiment, through multi-dimensional quality inspection of "simultaneous analysis of clarity and positional compliance of the imprint," can significantly improve the comprehensiveness and scientific nature of quality assessment. This facilitates a comprehensive evaluation of stamping quality from two key dimensions: the recognizability of the imprint itself (clarity) and its relative relationship with the document (position). Consequently, it leads to more comprehensive and accurate conclusions of pass / fail, enabling refined quality grading and control. The quantitative evaluation method of "calculating the edge gradient features of the imprint to obtain a clarity value" effectively improves the objectivity and consistency of clarity judgment, thus avoiding fluctuations and disputes caused by subjective human judgment. This, in turn, helps establish stable and reliable quality inspection standards, achieving standardized inspection of imprint imaging quality. By "matching feature points between the actual imprint outline and the standard outline to calculate the offset," the accuracy and automation level of positional deviation detection can be significantly improved. This allows for precise quantification of the minute differences between the actual and ideal positions of the imprint, providing accurate data support for process improvement and equipment calibration, and achieving precise positional measurement based on machine vision.

[0089] In this optional embodiment, as an optional implementation, the detection and judgment module 202 compares the key information of the full text with the pre-stored standard information to calculate the target similarity value between the key information of the full text and the pre-stored standard information in the following specific ways: Semantic similarity calculation is performed between the key information in the full text and the corresponding standard text in the pre-stored standard information. The semantic similarity calculation includes extracting the semantic feature vectors of the key information in the full text and the standard text respectively, and calculating the cosine similarity between the two semantic feature vectors as the target similarity value.

[0090] It is evident that implementing this optional embodiment, by "performing semantic similarity calculation" rather than simple literal matching, can greatly improve the intelligence and fault tolerance of content comparison, thereby facilitating the understanding of the core meaning of the text, effectively handling differences in expression such as synonym replacement and word order adjustment, and thus helping to reduce unnecessary verification failures while ensuring the substantive consistency of the content, achieving a more intelligent and humanized document content verification.

[0091] In this optional embodiment, as another optional implementation, the specific method by which the detection control module 203 extracts the region coordinates of each pre-marked stamped region in the document according to a preset image segmentation and contour extraction algorithm includes: The image coordinates of pre-set reference marks on the document are identified using a visual sensor; The image coordinates of the reference mark are combined with pre-calibrated visual perception parameters to convert them into world coordinates in the coordinate system of the stamping device; The regional coordinates of the stamping area are calculated based on the world coordinates and the pre-defined offset of the stamping area relative to the reference mark.

[0092] As can be seen, implementing this optional embodiment, by "converting the image coordinates of the reference marker to world coordinates," can significantly improve the accuracy and adaptability of the positioning system. This facilitates the precise correlation between visual information based on two-dimensional images and the three-dimensional physical execution space, thereby overcoming errors caused by camera perspective, lens distortion, etc., and achieving accurate mapping from image perception to physical operation. By "calculating the stamping area coordinates based on world coordinates and preset offsets," the flexibility and configurability of the positioning operation can be effectively improved. This allows for quick adaptation to the stamping position requirements of different document templates by simply modifying the offset parameters, thereby enhancing the device's rapid response capability to different business scenarios and achieving flexible stamping position settings.

[0093] In an optional embodiment, the transmission control module 201 is further configured to dynamically control the start and stop of the transmission device in the automated stamping device according to the processing status of the document content identification and verification area and the ready status of the positioning stamping area, so that the document is paused when it arrives at the corresponding area until the processing of that area is completed. The conveying control module 201 is also used to control the diversion mechanism in the automated stamping device to guide the abnormal document to an independent collection path and trigger an alarm signal when an abnormal document conveying is detected by a sensor set in the conveying path in the automated stamping device.

[0094] As can be seen, implementing this optional embodiment, by "dynamically controlling the start and stop of the conveyor device according to the processing status of each area," can significantly improve the work efficiency and coordination of the entire production line. This helps ensure sufficient processing time at each stage, avoids document congestion or equipment idleness, and thus optimizes production cycle time, increases the throughput per unit time, and achieves efficient and orderly collaborative operation between modules. By "automatically diverting and alarming when document transmission abnormalities are detected," the reliability and autonomy of the system can be greatly improved. This facilitates rapid response and isolation of problems such as paper jams and overlaps, preventing the escalation of faults or damage to documents. Consequently, it helps ensure that the qualified product process is not affected, and quickly notifies personnel to intervene, achieving intelligent fault diagnosis and handling, and ensuring the continuity and safety of production.

[0095] Example 3 Please see Figure 4 , Figure 4This is a schematic diagram of the control device for another cascaded and expandable automated stamping device disclosed in an embodiment of the present invention. The control device for this cascaded and expandable automated stamping device can be applied to the automated stamping device itself, and also to intelligent devices associated with the automated stamping device. These intelligent devices include, but are not limited to, one or more of the following: switching devices, cloud devices, edge computing devices, relay devices, base station devices, urban management devices, and intelligent connected devices. This embodiment of the present invention does not impose any limitations. Figure 4 As shown, the control device for this cascaded and expandable automated stamping device may include: Memory 301 that stores executable program code.

[0096] Processor 302 coupled to memory 301.

[0097] The processor 302 calls the executable program code stored in the memory 301 to execute the steps in the control method of the cascaded and expandable automated stamping device described in Embodiment 1 of the present invention.

[0098] Example 4 This invention discloses a computer storage medium storing computer instructions. When these computer instructions are invoked, they are used to execute the steps in the control method of the cascaded and expandable automated stamping device described in Embodiment 1 of this invention.

[0099] Example 5 This invention discloses a computer program product, which includes a non-transitory computer storage medium storing a computer program, and the computer program is operable to cause a computer to perform the steps in the control method of the cascaded and expandable automated stamping device described in Embodiment 1.

[0100] The device embodiments described above are merely illustrative. The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0101] Through the detailed description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0102] Finally, it should be noted that the control method and apparatus for an automated stamping device that can be cascaded and expanded according to the embodiments of the present invention are only preferred embodiments of the present invention and are only used to illustrate the technical solutions of the present invention, not 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 therein; and these 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.

Claims

1. A control method for an automated stamping device that can be cascaded and expanded, characterized in that, The method includes: Control the document to be transported along a preset path in the automated stamping device; When the document is detected to have been transmitted to the document content recognition and verification area of ​​the automated stamping device, it is determined whether the document matches the pre-stored standard information. If they match, the document is transmitted to the positioning stamping area of ​​the automated stamping device. When the document is detected to have been delivered to the positioning stamping area, stamping control parameters for each stamp are generated based on each stamping area pre-marked in the document and the corresponding cascaded stamps in the automated stamping device, so as to control the stamp to perform the corresponding stamping control operation. The stamping control parameters include movement control parameters and stamping pressure control parameters. After the stamping control operation is completed, the document is sent to the stamping review area; a multi-dimensional print quality analysis is performed on the document to determine whether the document meets the preset stamping quality conditions. If it does, the document is sent to the document collection device.

2. The control method for the cascaded and expandable automated stamping device according to claim 1, characterized in that, The determination of whether the document matches the pre-stored standard information includes: Obtain a complete image of the document; Perform preprocessing operations on the complete image, the preprocessing operations including at least one of color separation operation, distortion correction operation, and noise filtering operation; The preprocessed complete image is input into a preset text recognition engine to identify and extract the full text key information of the document, which includes at least one of the document's title, number, body text, and icon. The key information in the full text is compared with the pre-stored standard information to calculate the target similarity value between the key information in the full text and the pre-stored standard information. Determine whether the target similarity value is greater than or equal to a preset similarity threshold. If the target similarity value is determined to be greater than or equal to the preset similarity threshold, then the document is determined to match the pre-stored standard information.

3. The control method for the cascaded and expandable automated stamping device according to claim 1, characterized in that, The step of generating stamping control parameters for each stamp based on each pre-marked stamping area in the identified document and the corresponding cascaded stamps in the automated stamping device, to control the stamps to perform corresponding stamping control operations, includes: For each pre-marked stamped area in the document, the region coordinates of the stamped area are extracted according to a preset image segmentation and contour extraction algorithm; Obtain the current position coordinates of the cascaded stamps in the automated stamping device corresponding to the stamping area; Based on the current position coordinates and the area coordinates, stamping control parameters are generated for each stamp to control the stamp to perform the corresponding stamping control operation. The movement control parameters include horizontal movement control parameters and vertical movement control parameters.

4. The control method for the cascaded and expandable automated stamping device according to claim 1, characterized in that, The step of performing multi-dimensional seal quality analysis on the document to determine whether the document meets preset stamping quality conditions includes: Obtain a partial image of the actual stamped area of ​​the document within the stamped review area; Simultaneously perform text clarity analysis and text position compliance analysis on the local image; the text clarity analysis includes calculating the gradient features of the text edges in the local image to obtain a clarity value; the text position compliance analysis includes matching the text outline in the local image with the pre-stored standard text outline by feature points to calculate the position offset. When the clarity value meets the preset clarity standard and the position offset is less than or equal to the preset tolerance threshold, the document is determined to meet the preset stamping quality conditions.

5. The control method for the cascaded and expandable automated stamping device according to claim 2, characterized in that, The step of comparing the full-text key information with pre-stored standard information to calculate the target similarity value between the full-text key information and the pre-stored standard information includes: Semantic similarity calculation is performed between the full-text key information and the corresponding standard text in the pre-stored standard information; the semantic similarity calculation includes extracting the semantic feature vectors of the full-text key information and the standard text respectively, and calculating the cosine similarity between the two semantic feature vectors as the target similarity value.

6. The control method for the cascaded and expandable automated stamping device according to claim 3, characterized in that, For each pre-marked stamped area in the document, the step of extracting the region coordinates of the stamped area according to a preset image segmentation and contour extraction algorithm includes: The image coordinates of pre-set reference marks on the document are identified using a visual sensor; The image coordinates of the reference mark are combined with pre-calibrated visual perception parameters to convert them into world coordinates in the coordinate system of the stamping device; Based on the world coordinates and the pre-defined offset of the stamping area relative to the reference mark, the regional coordinates of the stamping area are calculated.

7. The control method for the cascaded and expandable automated stamping device according to any one of claims 1-6, characterized in that, The method further includes: Based on the processing status of the document content recognition and verification area and the ready status of the positioning and stamping area, the start and stop of the transmission device are dynamically controlled so that the document is paused when it arrives at the corresponding area until the processing of that area is completed. When a document delivery anomaly is detected by a sensor installed in the delivery path, the control diversion mechanism guides the abnormal document to an independent collection path and triggers an alarm signal.

8. A control device for an automated stamping device that can be cascaded and expanded, characterized in that, The device includes: The delivery control module is used to control the delivery of documents along a preset path in the automated stamping device; The detection and judgment module is used to determine whether the document matches the pre-stored standard information when the document is detected to have been transmitted to the document content recognition and verification area of ​​the automated stamping device. The transmission control module is also used to transmit the document to the positioning and stamping area of ​​the automated stamping device if a match is found. The detection control module is used to generate stamping control parameters for each stamp based on each pre-marked stamping area in the document and the corresponding cascaded stamps in the automated stamping device when the document is detected to have been delivered to the positioning stamping area. The stamping control parameters include movement control parameters and stamping pressure control parameters. The transmission control module is also used to transmit the document to the stamping review area after the stamping control operation is completed. The detection and judgment module is also used to perform multi-dimensional seal quality analysis on the document to determine whether the document meets the preset stamping quality conditions. The transmission control module is further configured to transmit the document to the document collection device if the conditions are met.

9. A control device for an automated stamping device that can be cascaded and expanded, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the control method of the cascadeable and expandable automated stamping device as described in any one of claims 1-7.

10. A computer storage medium, characterized in that, The computer storage medium stores computer instructions, which, when invoked, are used to execute the control method of the cascadeable and expandable automated stamping device as described in any one of claims 1-7.

Images