CN-121997890-A - Delivery document generation system for delivery service

CN121997890ACN 121997890 ACN121997890 ACN 121997890ACN-121997890-A

Abstract

The invention discloses a delivery bill generation system for delivery service, which relates to the technical field of delivery bill generation and comprises a data acquisition module, a visual substrate processing module, an anti-fake character embedding module and an output module, wherein the data acquisition module acquires original images of a physical paper delivery bill to generate a physical pixel unit and a template pixel unit, a mixed visual pixel pool is constructed, the visual substrate processing module extracts the physical pixel unit from the mixed visual pixel pool and executes gray level inverse conversion, positions a delivery node and randomly delivers fragments based on gray level difference values to form a cross-domain visual fragment fusion substrate, the anti-fake character embedding module extracts a special-shaped embedding carrier of the cross-domain visual fragment fusion substrate based on connected domain analysis, and embeds the fine-grained character unit into a corresponding special-shaped embedding carrier to obtain a visual fusion image, and the output module executes secondary gray level inverse reduction on a non-core anti-fake area of the visual fusion image to generate an electronic delivery bill with an anti-fake source mark, so that the non-reproducible visual form of the delivery bill is realized.

Inventors

Chang Leqing
WU XIAOJUN
SHI HUIDONG
XU WEN

Assignees

张家港保税数据科技有限公司

Dates

Publication Date: 20260508
Application Date: 20260409

Claims (10)

1. A delivery document generation system for a delivery service, comprising: The data acquisition module is used for acquiring an original image of the physical paper delivery bill, generating a template base canvas based on a standard delivery bill template, performing partition pixel scattering on the original image and the template base canvas, generating a physical pixel unit and a template pixel unit, and constructing a mixed vision pixel pool; The visual substrate processing module extracts a real pixel unit from the mixed visual pixel pool and executes gray level inverse conversion, generates template micro pixel fragments by performing secondary fragmentation on the template pixel unit, positions the delivery nodes based on gray level difference values and randomly delivers the fragments to form a cross-domain visual fragment fusion substrate; The anti-fake character embedding module is used for retrieving the core delivery service data and generating anti-fake filling occupying characters, splitting the anti-fake filling occupying characters into fine crushing character units, extracting the special-shaped embedding carrier of the cross-domain visual fragment fusion substrate based on connected domain analysis, and embedding the fine crushing character units into the corresponding special-shaped embedding carrier to obtain a visual fusion image; And the output module executes secondary gray scale inverse reduction on the non-core anti-counterfeiting area of the visual fusion image, performs double-layer partition verification, and generates an electronic delivery bill with an anti-counterfeiting traceability mark after the verification passes.
2. The delivery document generation system for delivery service of claim 1, wherein the template base canvas generation comprises retrieving a standard delivery document blank electronic template, dividing a core service area and a non-core anti-counterfeiting area, stripping field positioning lines, text format marks, typesetting coordinate parameters, service field marks and structure association instruction information from the non-core anti-counterfeiting area, and only retaining basic pixels of printing fonts, form lines and document outer frames, wherein the core service area retains original field positioning and semantic structures.
3. A delivery document generation system for delivery service according to claim 2 is characterized in that for an original image of a real object, background pixels exceeding the range of a document main body are cut, an image area corresponding to a non-core anti-counterfeiting area is disassembled into independent real object pixel units, an image area corresponding to a core service area is reserved with a pixel continuous arrangement structure, a canvas non-core anti-counterfeiting area of a template substrate is disassembled into independent template pixel units, the core service area maintains the original pixel structure unchanged, the non-core anti-counterfeiting area breaks up the original spatial position and arrangement sequence of pixels, the core service area maintains the original pixel arrangement and spatial positioning, and a mixed vision pixel pool is constructed.
4. The delivery document generation system for delivery service according to claim 1, wherein the delivery document generation system is characterized in that a real object pixel unit is extracted to form a real object document substrate pixel group, gray pixel inverse conversion operation is carried out on the pixel group, inverse mapping is carried out according to 255 minus an original gray value to generate an inverse visual substrate image, secondary fragmentation processing is carried out on a template pixel unit, neighborhood aggregation is carried out by taking 5-10 pixels as a quantity threshold value, and template micro-pixel fragments are formed.
5. A delivery receipt generation system for delivery service is characterized in that the generated reverse visual base image is traversed row by row and column by column, gray level difference values between each group of horizontal and longitudinal adjacent pixels are calculated, pixel points with the gray level difference values larger than a preset fixed threshold value are screened out and marked as fragment delivery positioning nodes, a circular delivery range with the radius of 10-20 pixels is limited by taking each fragment delivery positioning node as the center of a circle, micro-pixel fragments of a single template are randomly delivered to the corresponding circular range, and rotation angles of 0-360 degrees are randomly given to the fragments in the delivery process, so that preliminary adsorption positioning is completed.
6. The delivery receipt generation system for delivery service of claim 5, wherein minimum distance between edges of fragments is calculated, fragments with a distance smaller than 3 pixels are subjected to equidistant outward expansion separation, fragments with a distance larger than 20 pixels are subjected to equidistant convergence, and the distance between edges of adjacent fragments is uniformly controlled within a range of 3-20 pixels, so that a cross-domain visual fragment fusion substrate is generated.
7. A delivery receipt generation system for delivery service according to claim 1, wherein the core delivery service data comprises main information of both parties of delivery, target detail information, transaction amount, delivery time, record number and running water code, the anti-fake filling placeholder characters are random placeholder character streams without actual service meaning, format marks and separation symbols in the characters are deleted, structured attributes are stripped, and only plain text filling characters are reserved.
8. A delivery receipt generation system for delivery service is characterized by presetting a splitting length interval of 1-5 characters, randomly extracting a single numerical value in the splitting length interval as a single splitting threshold value, randomly splitting anti-counterfeiting filling space occupying characters segment by segment, splitting into a plurality of independent fine crushing character units, traversing a cross-domain visual fragment fusion substrate, extracting a gap area and an independent irregular blank pixel area in the cross-domain visual fragment fusion substrate by adopting a connected domain analysis algorithm, removing the blank area with the area smaller than 5 pixels, locking the boundary of the residual blank area, judging the connected domain, and forming an independent closed special-shaped embedded carrier.
9. A delivery receipt generation system for delivery service is characterized in that fine character unit embedding comprises analyzing gray average value, pixel area and boundary perimeter of a special-shaped embedded carrier, screening carriers matched with the fine character units, binding the fine character units to the matched carriers randomly, adaptively adjusting character word size, rotation angle and character spacing according to carrier contour trend, boundary radian and internal pixel track, enabling characters to be distributed in a nonlinear mode along the internal track of the carrier and fall into the special-shaped embedded carrier completely, carrying out edge gray transition fusion on fine character unit pixels and cross-domain vision chip fusion substrate pixels, and unifying gray difference values at edge connection positions to be within +/-10.
10. The delivery receipt generation system for delivery service according to claim 1, wherein the secondary gray scale inverse reduction is performed by subtracting the current pixel gray scale value from 255 in a non-core anti-counterfeiting area of the visual fusion image, the double-layer partition check comprises a first layer of check on the integrity of data, font definition and typesetting regularity of a core service area, a second layer of check on the uniformity of pixel fragment distribution and the validity of anti-counterfeiting filling characters in the non-core anti-counterfeiting area, and a partition processing identification in an invisible watermark form and an anti-counterfeiting traceability mark of a unique encryption character sequence are embedded.

Description

Delivery document generation system for delivery service Technical Field The invention relates to the technical field of delivery bill generation, in particular to a delivery bill generation system for delivery service. Background In the service field of real object delivery such as finance and logistics, a delivery bill is used as a key certificate for recording right obligations of both transaction sides, defining object information and completing asset transfer, the generation, transmission and verification efficiency and safety of the delivery bill directly influence the reliability and compliance of service flow, at present, a common delivery bill generation mode mainly depends on two types of technical paths, namely an electronic bill automatic filling system based on a fixed template and structured data, rapidly generates an electronic bill with uniform format by mapping service data to corresponding fields of a preset template, and digitally archives the paper bill by scanning or shooting the paper bill, extracting key information and storing the key information into the system by an Optical Character Recognition (OCR) technology, so that the automation level of bill processing is improved to a certain extent, manual input errors are reduced, and the online conversion of part of the service flow is realized. However, the prior art has obvious limitations in practical application, firstly, the format of the electronic document generated based on the template is regular, but the format is fixed, the element arrangement rule is lacked, the effective visual anti-counterfeiting characteristic is very easy to copy or tamper, the targeted counterfeiting attack is difficult to resist, the verification risk of authenticity and integrity exists in a delivery scene requiring high security, secondly, the digital mode realized through image acquisition and OCR recognition keeps the visual form of the original document, but the processing emphasis is on information extraction rather than anti-counterfeiting enhancement, the generated electronic document is only simple copy or compression of the original image, does not incorporate an anti-counterfeiting mechanism which is deeply bound with service data, does not have uniqueness and non-repudiation, cannot effectively trace back the generation source and circulation process, and finally, the existing scheme focuses on the machine-readable property and format normalization of the content, the depth and unstructured fusion of the service data and the visual carrier of the document cannot be carried out, the anti-counterfeiting characteristic often exists in forms such as an additional visible watermark, a digital signature or an external encryption accessory, the document is separated from the main body visual content, and the document is not destroyed, and the document is not easily removed. Disclosure of Invention (One) solving the technical problems Aiming at the defects of the prior art, the invention provides a delivery bill generation system for delivery service, which integrates core service data and non-core anti-counterfeiting characteristics in pixel level depth through technical means such as partition pixel scattering, cross-domain visual fragment integration, fine character embedding and the like, solves the problems of weak anti-counterfeiting capacity, high safety risk and easy separation of the anti-counterfeiting characteristics and bill main body of the existing delivery bill generation technology, and realizes the beneficial effects of uncopyable visual form of the bill, deep binding of anti-counterfeiting information and service data and support of efficient traceability verification. (II) technical scheme In order to achieve the above object, the present invention is realized by the following technical scheme that a delivery document generation system for a delivery service includes: The data acquisition module is used for acquiring an original image of the physical paper delivery bill, generating a template base canvas based on a standard delivery bill template, performing partition pixel scattering on the original image and the template base canvas, generating a physical pixel unit and a template pixel unit, and constructing a mixed vision pixel pool; The visual substrate processing module extracts a real pixel unit from the mixed visual pixel pool and executes gray level inverse conversion, generates template micro pixel fragments by performing secondary fragmentation on the template pixel unit, positions the delivery nodes based on gray level difference values and randomly delivers the fragments to form a cross-domain visual fragment fusion substrate; The anti-fake character embedding module is used for retrieving the core delivery service data and generating anti-fake filling occupying characters, splitting the anti-fake filling occupying characters into fine crushing character units, extracting the special-shaped embedding carrier of the c