A blockchain-based supply chain end-to-end traceability and trusted data sharing system

Abstract

This invention discloses a blockchain-based supply chain end-to-end traceability and trusted data sharing system, relating to the field of trusted sharing technology. The system includes: a dark-field development system that generates a dark-field scattering development map based on the bridging neck region corresponding to the RFID tamper-evident seal; a polarization overlay system that generates a total intensity image based on the first and third registration polarization images of the bridging neck region, and obtains a polarization overlay map based on the total intensity image; a pseudo-closure extraction system that generates a fused image based on the dark-field scattering development map and the polarization overlay map, and iteratively updates the particle parameter vector according to the fused image to obtain a pseudo-closure contact response map; and a geometric evidence system that generates a normal encoding map and obtains a geometric evidence map based on the normal encoding map. This invention can enhance the credibility of traceability event sharing.

Description

Technical Field

[0001] This invention relates to the field of trusted data sharing technology, and in particular to a blockchain-based supply chain end-to-end traceability and trusted data sharing system. Background Technology

[0002] In supply chain traceability and trusted data sharing scenarios, RFID tamper-evident seals are commonly used for the management of pharmaceuticals, food, cold chain goods, precision components, valuables, and cross-border logistics parcels. These seals typically accompany goods through various stages, including production outbound, warehousing inbound, loading and unloading, vehicle transportation, distribution, and final acceptance. At each stage, reading devices retrieve seal identifiers, batch information, node information, and operational information, forming a traceable data link. To improve the reliability of traceability data, existing solutions typically write supply chain event data into a blockchain ledger, leveraging the immutability of the blockchain ledger to achieve data sharing among multiple entities. RFID tamper-evident seals often have a bridging neck area, a localized narrowing connection point in the conductive circuit. This area is prone to physical changes such as cracking, warping, or partial adhesion when the seal is peeled, bent, pressed, or transferred.

[0003] Existing supply chain traceability solutions typically use the readability of RFID seals as evidence of their normal status, and write the read seal identifiers and corresponding supply chain events into the blockchain ledger. The problem with this approach is that even if an internal fracture has occurred in the bridging neck area, temporary continuity may still be created due to metal edge curling, conductive layer rebound, or external pressure, allowing the seal to still be identified by the reading device. This results in an inconsistency between the physical failure state of the seal and its on-chain identity continuity.

[0004] While existing technologies can ensure that data written into the blockchain ledger is not easily tampered with, it is difficult to prove that the bridging neck area inside the RFID tamper-evident seal is still in a true and intact state when the data is written. It is impossible to transform the state of the seal being broken but partially reattached and conductive into verifiable and shareable image evidence, thus affecting the reliable sharing of data and the traceability of responsibility in supply chain traceability events. Summary of the Invention

[0005] The purpose of this invention is to address the shortcomings of existing technologies that make it difficult to prove that the bridging neck area inside the RFID tamper-evident seal remains in a true and intact state when data is written. Therefore, this invention proposes a blockchain-based supply chain end-to-end traceability and trusted data sharing system.

[0006] To address the problems existing in the prior art, the present invention adopts the following technical solution: A blockchain-based supply chain end-to-end traceability and trusted data sharing system includes: The dark field development system generates a dark field scattering development map based on the bridging neck area corresponding to the RFID tamper-evident seal. The polarization overlay system generates a total intensity image based on the first and third registered polarization images of the bridging neck region, and obtains a polarization overlay map based on the total intensity image; The pseudo-closed extraction system generates a fused image based on the dark field scattering development map and the polarization overlay map, and iteratively updates the particle parameter vector according to the fused image to obtain the pseudo-closed contact response map; A geometric evidence system generates a normal encoding map, and then obtains a geometric evidence map based on the normal encoding map. The on-chain sharing system obtains image digest barcode images based on geometric evidence graphs and pseudo-closed contact response graphs, and performs trusted data sharing of supply chain traceability events based on the image digest barcode images.

[0007] Compared with the prior art, the beneficial effects of the present invention are: 1. In this invention, by determining the bridging neck area of ​​the RFID tamper-evident seal and using multi-directional dark field image registration and fusion to generate a dark field scattering development map, the scattering changes of cracked edges, warped edges and local bonding positions can be highlighted, so that the abnormality of the conductive structure inside the seal no longer depends solely on whether the RFID can be read, thereby improving the targeting and visualization of the front-end physical state acquisition and providing the original image basis for subsequent trusted sharing.

[0008] 2. In this invention, a polarization overlay map is obtained by linear polarization acquisition, polarization difference calculation, linear polarization degree generation and channel mapping fusion. A scattering response map and a polarization response map are generated under the overlay state. A pseudo-closed contact response map is obtained by particle swarm optimization, thereby improving the image representation ability of the local re-contact state. This makes the attached conduction state after internal fracture, which is difficult to observe directly, form processable evidence and improve the clarity of the expression of abnormal areas.

[0009] 3. In this invention, a geometric evidence diagram is formed by combining a normal encoding diagram and a height encoding diagram. The geometric evidence diagram is then combined with a pseudo-closed contact response diagram to generate a trusted shared visual credential diagram. This is then combined with an image digest barcode and a blockchain ledger to complete the sharing. This allows supply chain nodes to simultaneously verify physical state evidence and on-chain event data, reducing the bias in responsibility tracing caused by relying solely on on-chain identity continuity and enhancing the credibility of tracing event sharing. Attached Figure Description

[0010] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1This is a functional block diagram of a blockchain-based supply chain end-to-end traceability and trusted data sharing system, provided as an embodiment of the present invention. Detailed Implementation

[0011] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0012] This embodiment provides a blockchain-based supply chain end-to-end traceability and trusted data sharing system. (See also...) Figure 1 Specifically, including: The dark field development system generates a dark field scattering development map based on the bridging neck area corresponding to the RFID tamper-evident seal. In an embodiment of the present invention, a dark field scattering imaging map is generated based on the bridging neck region corresponding to the RFID tamper-evident seal, including: Identify the bridging neck area corresponding to the RFID tamper-evident seal; The bridging neck area refers to a locally narrowed connection area in the conductive bridging structure inside an RFID tamper-evident seal. This area is usually formed at the narrow band connection position between conductive lines. Its lateral dimension is smaller than that of the adjacent conductive area, and the thickness of the conductive material is relatively thin. When the seal is lifted, bent, pressed, pulled, or subjected to localized rebound, this area is more prone to localized deformation, edge lifting, conductive layer cracking, and overlapping cracks. Therefore, the bridging neck area corresponds to the location inside the RFID tamper-evident seal where structural changes are most likely to occur.

[0013] The process involves acquiring an original seal image containing the conductive bridging structure on the front of the RFID tamper-evident seal, converting the original seal image to grayscale to obtain a grayscale image of the seal, extracting edges from the grayscale image of the seal to obtain the edge image of the conductive lines, determining the extension direction and narrowing connection position of the conductive lines based on the edge image of the conductive lines, identifying the connection position between two conductive lines whose lateral width is smaller than that of the adjacent conductive lines as a candidate region for bridging necks, locating the circumscribed rectangle of the candidate region for bridging necks to obtain the region boundary coordinates, and extracting an image block containing the candidate region for bridging necks and its surrounding conductive lines from the original seal image based on the region boundary coordinates, and identifying this image block as the bridging neck region.

[0014] It should be noted that an RFID tamper-evident seal refers to a seal structure that simultaneously possesses radio frequency identification (RFID) and physical tamper-evident functions. This seal includes a seal substrate, a conductive antenna structure, an RFID chip, and a bridging conductive area for connecting the conductive lines. The conductive antenna structure receives and transmits RFID signals, the RFID chip stores seal identification information and responds to RFID communication requests from external readers, and the bridging conductive area maintains the electrical connection of the conductive antenna structure. When the RFID tamper-evident seal is intact, the conductive lines remain continuously conductive, and the RFID chip can interact with external readers. When the RFID tamper-evident seal is lifted, removed, stretched, bent, or moved, the bridging conductive area is prone to cracking, breaking, warping, or partial overlap, resulting in a change in conductivity. Therefore, an RFID tamper-evident seal corresponds to a physical carrier used to characterize the circulation and unsealing status of an item.

[0015] Acquire the first, second, third, and fourth dark-field images of the bridging neck region; The first dark field image, the second dark field image, the third dark field image, and the fourth dark field image refer to images of the bridging neck region acquired under different dark field illumination directions. Different dark field illumination directions correspond to low-angle incident light rays in different directions. When low-angle incident light rays propagate on the surface of the bridging neck region, flat areas generate less reflected light, while cracked edges, warped edges, local protrusions, and the edges of the conductive layer are more likely to generate scattered light. Therefore, the differences in brightness distribution in different dark field images correspond to the differences in scattering states in different directions on the surface of the bridging neck region.

[0016] The RFID tamper-evident seal is fixed on the imaging platform, with the bridging neck area positioned at the center of the camera's field of view and the camera's optical axis perpendicular to the seal surface. Ambient light is turned off, and low-angle dark-field illumination is used to illuminate the bridging neck area from the left, right, top, and bottom sides. Under each illumination direction, the camera focal length, aperture, exposure time, gain, seal position, and bridging neck area position remain unchanged while images are acquired. The image of the bridging neck area acquired under the left dark-field illumination is taken as the first dark-field image, the image of the bridging neck area acquired under the right dark-field illumination is taken as the second dark-field image, the image of the bridging neck area acquired under the top dark-field illumination is taken as the third dark-field image, and the image of the bridging neck area acquired under the bottom dark-field illumination is taken as the fourth dark-field image.

[0017] Image registration is performed on the first dark-field image, the second dark-field image, the third dark-field image, and the fourth dark-field image to obtain the first registered dark-field image, the second registered dark-field image, the third registered dark-field image, and the fourth registered dark-field image; The first registered dark field image, the second registered dark field image, the third registered dark field image, and the fourth registered dark field image refer to the dark field images after spatial coordinate unification. The same pixel position in each registered dark field image corresponds to the same physical position in the bridging neck region, so that the crack position, warping position, and conductive layer edge position in different dark field images form a spatial correspondence.

[0018] Using the first dark-field image as the reference image, and the second, third, and fourth dark-field images as images to be registered, grayscale normalization is performed on both the reference image and each image to be registered to obtain a reference normalized image and an image to be registered normalized image. The conductive line edge points, crack scattering points, and region corner points of the bridging neck region are extracted from the reference normalized image, and the corresponding conductive line edge points, crack scattering points, and region corner points are extracted from the image to be registered normalized image. Translation, rotation, and scale changes are calculated based on the point coordinates in the reference and image to be registered normalized images. An image space transformation relationship is generated based on the translation, rotation, and scale changes. Each image to be registered is mapped to the pixel coordinate system of the first dark-field image according to the corresponding image space transformation relationship. Bilinear interpolation is performed on pixel values ​​that do not fall at integer pixel positions after mapping to obtain the second, third, and fourth registered dark-field images. The first dark-field image is used as the first registered dark-field image.

[0019] Based on the first registered dark-field image, the second registered dark-field image, the third registered dark-field image, and the fourth registered dark-field image, fused pixel values ​​are generated; The fused pixel value refers to the pixel data formed after the brightness information of multiple registered dark field images at the same pixel position is fused. This fusion operation is used to integrate the scattering response intensity under different dark field illumination directions, so that the scattering information generated by local cracks, edge curling and overlapping positions in the bridging neck area under different illumination directions is concentrated at the same pixel position.

[0020] Using the pixel coordinate system of the first registered dark field image as the fusion coordinate system, the pixel brightness value of each pixel position in the fusion coordinate system is read from the first, second, third, and fourth registered dark field images. The four pixel brightness values ​​at the same pixel position are sorted from smallest to largest. The smallest pixel brightness value in the sorting result is deleted, and the remaining three pixel brightness values ​​are retained. The arithmetic mean of the retained three pixel brightness values ​​is calculated, and the arithmetic mean is used as the fusion pixel value of that pixel position. A fusion pixel matrix is ​​formed according to the fusion pixel values ​​corresponding to all pixel positions in the fusion coordinate system.

[0021] A dark field scattering development map is generated based on the fused pixel values.

[0022] Dark field scattering development map refers to the image of the bridging neck region generated based on the fused pixel values. The bright areas in this image correspond to the locations in the bridging neck region with strong scattering response, while the dark areas correspond to locations with relatively flat surfaces. The dark field scattering development map can reflect the spatial distribution of crack edges, local warping, conductive layer undulations, and crack overlap areas in the bridging neck region.

[0023] Using the pixel coordinate system of the first registered dark field image as the coordinate system of the developed image, the fused pixel value of each pixel position in the fused pixel matrix is ​​written into the gray-level channel of the same pixel position in the developed image coordinate system to obtain an initial dark field scattering map. The initial dark field scattering map is then subjected to gray-level normalization processing to distribute the gray-level values ​​of each pixel position in the initial dark field scattering map to the range of image gray-level values. The initial dark field scattering map after gray-level normalization is then subjected to conductive line region boundary constraint processing. The pixel positions outside the bridging neck region are set as background gray-level values, while the pixel positions inside the bridging neck region retain the corresponding normalized gray-level values, resulting in a dark field scattering developed map that only contains scattering information of the bridging neck region.

[0024] The polarization overlay system generates a total intensity image based on the first and third registered polarization images of the bridging neck region, and obtains a polarization overlay map based on the total intensity image; In an embodiment of the present invention, a total intensity image is generated based on a first registered polarization image and a third registered polarization image of the bridging neck region, including: Linear polarization images were acquired in the bridging neck region to obtain the first polarization image, the second polarization image, the third polarization image, and the fourth polarization image; Linearly polarized images refer to the images of the bridging neck region obtained after setting a linear polarizer in the camera's optical path. The pixel brightness in this image is formed by the reflection, scattering, and absorption of light with a specific polarization direction on the surface of the bridging neck region. The first polarization image, the second polarization image, the third polarization image, and the fourth polarization image refer to the images of the bridging neck region acquired under different linear polarization directions. Each polarization image is used to express the imaging differences of the metal layer, cracked edge region, pressed region, and undulating region on the surface of the bridging neck region under different polarization directions.

[0025] An RFID tamper-evident seal is fixed on an imaging platform, with the bridging neck area positioned at the center of the camera's field of view and the camera's optical axis perpendicular to the seal surface. A rotatable linear polarizer is placed in front of the camera lens. While keeping the camera position, focal length, aperture, exposure time, gain, seal position, and illumination position constant, the linear polarizer is adjusted to the first, second, third, and fourth polarization directions, respectively, to acquire images of the bridging neck area. The image of the bridging neck area acquired under the first polarization direction is used as the first polarized image, the image of the bridging neck area acquired under the second polarization direction is used as the second polarized image, the image of the bridging neck area acquired under the third polarization direction is used as the third polarized image, and the image of the bridging neck area acquired under the fourth polarization direction is used as the fourth polarized image.

[0026] Image registration is performed on the first polarization image, the second polarization image, the third polarization image, and the fourth polarization image to obtain the first registered polarization image, the second registered polarization image, the third registered polarization image, and the fourth registered polarization image; The first, second, third, and fourth registered polarization images refer to polarization images after spatial coordinate unification. The same pixel positions in each registered polarization image correspond to the same entity positions in the bridging neck region.

[0027] Using the first polarization image as the reference polarization image, and the second, third, and fourth polarization images as polarization images to be registered, grayscale normalization is performed on both the reference polarization image and each polarization image to be registered. The conductive line edge points, bridging neck region boundary points, and metal surface texture points of the bridging neck region are extracted from the reference polarization image, and the corresponding conductive line edge points, bridging neck region boundary points, and metal surface texture points are extracted from the polarization images to be registered. Translation, rotation, and scale changes are calculated based on the point coordinates in the reference and polarization images to be registered. An image space transformation relationship is generated based on the translation, rotation, and scale changes. Each polarization image to be registered is mapped to the pixel coordinate system of the first polarization image according to the corresponding image space transformation relationship. Bilinear interpolation is performed on pixel values ​​that do not fall at integer pixel positions after mapping to obtain the second, third, and fourth registered polarization images. The first polarization image is used as the first registered polarization image.

[0028] The total intensity image is generated by adding the pixel values ​​of the first and third registered polarization images.

[0029] The total intensity image refers to the image obtained by adding the pixel values ​​at the same pixel position of the first and third registered polarization images. This image is used to express the overall reflection and scattering intensity of the bridging neck region under the corresponding polarization direction combination.

[0030] Using the pixel coordinate system of the first registered polarization image as the coordinate system of the total intensity image, the first pixel value of each pixel position in the total intensity image coordinate system in the first registered polarization image is read, and the third pixel value of the same pixel position in the third registered polarization image is read. The first pixel value and the third pixel value are added to obtain the total intensity pixel value of the corresponding pixel position. The total intensity pixel values ​​of all pixel positions are written into the blank image matrix according to the corresponding pixel coordinates to form the total intensity image. If the total intensity pixel value exceeds the image grayscale storage range, the total intensity pixel value is stored in the high bit depth image data format according to the original calculation result, so that the total intensity image retains the superposition information of reflection intensity at the same physical position of the first registered polarization image and the third registered polarization image.

[0031] In an embodiment of the present invention, obtaining a polarization overlay map based on the total intensity image includes: A first polarization difference image is generated based on the subtraction of pixel values ​​between the first and third registered polarization images. The first polarization difference image refers to the image formed by performing pixel value difference calculation on the same pixel position of the first registered polarization image and the second registered polarization image. The pixel values ​​in this image correspond to the changes in reflection intensity and scattering intensity of the bridging neck region under illumination conditions of two different linear polarization directions. Since the conductive layer surface, cracked edge region, local pressing region and conductive layer undulation region in the bridging neck region have different reflective capabilities for light with different polarization directions, the brightness change in the first polarization difference image can reflect the degree of response of different positions in the bridging neck region to the change in polarization direction.

[0032] Using the pixel coordinate system of the first registered polarization image as the coordinate system of the first polarization difference image, the first pixel value of each pixel position in the first polarization difference image coordinate system in the first registered polarization image is read, and the third pixel value of the same pixel position in the third registered polarization image is read. The first pixel value is subtracted from the third pixel value to obtain the first polarization difference pixel value of the corresponding pixel position. The first polarization difference pixel values ​​of all pixel positions are written into the blank image matrix according to the corresponding pixel coordinates to form the first polarization difference image. When the first polarization difference pixel value is negative, the negative data is retained and stored in a signed image data format, so that the first polarization difference image retains the reflection difference direction and reflection difference amplitude of the bridging neck region in the first polarization direction and the third polarization direction.

[0033] A second polarization difference image is generated based on the pixel values ​​of the second and fourth registered polarization images. The second polarization difference image refers to the image formed by performing pixel value difference calculation on the same pixel position of the second registered polarization image and the fourth registered polarization image. The pixel values ​​in this image correspond to the reflection change state formed by the bridging neck region under another set of linear polarization directions. Since the conductive material edges, local raised areas and pressing contact areas in the bridging neck region will change the polarization maintenance state of the incident light, the pixel distribution in the second polarization difference image can reflect the local surface direction changes and local contact state changes in the bridging neck region.

[0034] Using the pixel coordinate system of the second registered polarization image as the coordinate system of the second polarization difference image, the second pixel value of each pixel position in the second polarization difference image coordinate system is read in the second registered polarization image. The fourth pixel value of the same pixel position in the fourth registered polarization image is read. The second pixel value is subtracted from the fourth pixel value to obtain the second polarization difference pixel value of the corresponding pixel position. The second polarization difference pixel values ​​of all pixel positions are written into the blank image matrix according to the corresponding pixel coordinates to form the second polarization difference image. When the second polarization difference pixel value is negative, the negative data is retained and stored in a signed image data format, so that the second polarization difference image retains the reflection difference direction and reflection difference amplitude of the bridging neck region in the second polarization direction and the fourth polarization direction.

[0035] A linear polarization degree image is generated based on the first polarization difference image, the second polarization difference image, and the total intensity image. The linear polarization degree image refers to the image calculated based on the first polarization difference image, the second polarization difference image, and the total intensity image. Each pixel value in this image corresponds to the polarization direction retention capability of the bridging neck region at the corresponding pixel position. The bright areas in the linear polarization degree image indicate that the corresponding position has a strong directional reflection capability for a specific polarization direction, while the dark areas in the linear polarization degree image indicate that the corresponding position has a small reflection difference for different polarization directions. The conductive laminated areas, local overlapping areas, and relatively flat areas in the bridging neck region usually form a high polarization retention state, while the cracked areas, undulating areas, and edge curled areas usually form a low polarization retention state.

[0036] Using the pixel coordinate system of the total intensity image as the coordinate system of the linear polarization degree image, the first polarization difference pixel value of each pixel position in the linear polarization degree image coordinate system is read in the first polarization difference image, the second polarization difference pixel value of the same pixel position in the second polarization difference image is read, and the total intensity pixel value of the same pixel position in the total intensity image is read. The first polarization difference pixel value is squared to obtain the first squared value, and the second polarization difference pixel value is squared to obtain the second squared value. The first squared value and the second squared value are added to obtain the sum of squares. The square root of the sum of squares is performed to obtain the polarization difference amplitude. The polarization difference amplitude is divided by the total intensity pixel value to obtain the linear polarization degree pixel value of the corresponding pixel position. The linear polarization degree pixel values ​​of all pixel positions are written into the blank image matrix according to the corresponding pixel coordinates to form the linear polarization degree image.

[0037] Channel mapping fusion is performed on the linear polarization degree image and the total intensity image to obtain a polarization overlay map.

[0038] Channel mapping fusion refers to writing polarization change data from a linear polarization degree image and overall brightness data from a total intensity image into different color channels or different grayscale channels of the same image according to the same pixel coordinate relationship. This makes the polarization change information and overall reflection information in the bridging neck region correspond in spatial position, so that the pressing area, cracked edge area, local conductive overlap area, and surface undulation area of ​​the conductive layer in the bridging neck region form a combined expression that simultaneously contains brightness differences and polarization differences in the fused image. Polarization overlay map refers to the image obtained by channel mapping fusion of the linear polarization degree image and the total intensity image. The pixel values ​​of this image simultaneously contain the overall reflection intensity information and polarization direction change information of the bridging neck region. The brightness changes and color distribution in the polarization overlay map are used to express the comprehensive reflection state formed by the pressing position, cracked edge position, local overlap position, conductive layer edge position, and surface undulation position in the bridging neck region under polarized illumination conditions.

[0039] Using the pixel coordinate system of the linear polarization degree image as the coordinate system of the polarization overlay map, the linear polarization degree image is subjected to grayscale normalization to obtain a linear polarization degree normalized image, and the total intensity image is subjected to grayscale normalization to obtain a total intensity normalized image. The pixel value of each pixel position in the polarization overlay map coordinate system in the linear polarization degree normalized image is written into the first image channel of the polarization overlay map, the pixel value of the same pixel position in the total intensity normalized image is written into the second image channel of the polarization overlay map, and the average result of the pixel values ​​of the same pixel position in the linear polarization degree normalized image and the total intensity normalized image is written into the third image channel of the polarization overlay map. The first image channel, the second image channel, and the third image channel together constitute the polarization overlay map.

[0040] The pseudo-closed extraction system generates a fused image based on the dark field scattering development map and the polarization overlay map, and iteratively updates the particle parameter vector according to the fused image to obtain the pseudo-closed contact response map; In an embodiment of the present invention, a fused image is generated based on a dark field scattering development map and a polarization overlay map, including: Dark-field azimuth complementary acquisition and scattering development fusion processing were performed on the bridging neck region under the holding state to obtain the dark-field scattering development map under the holding state. The holding state refers to the state after applying continuous contact pressure to the bridging neck region. Under this state, the cracked edge position, local warped edge position, and conductive layer overlap position in the bridging neck region will change the degree of contact and surface morphology due to the force. The holding state dark field scattering development image refers to the image obtained by fusing dark field azimuth complementary acquisition and scattering development under the holding state. The brightness distribution in this image corresponds to the scattering light distribution state of the bridging neck region under pressure conditions.

[0041] An RFID tamper-evident seal is fixed to the imaging platform, with the bridging neck area positioned at the center of the camera's field of view. A pressing device is used to apply contact pressure to the bridging neck area while maintaining the seal position, camera position, camera focal length, camera aperture, camera exposure time, and camera gain constant. Low-angle dark-field illumination is used to illuminate the bridging neck area from the left, right, top, and bottom sides, respectively, and four pressed dark-field images are acquired under the pressing condition. Image registration is performed on these four images to obtain the first... The first, second, third, and fourth registered dark-field images are registered and pressed together. The pixel coordinate system of the first registered dark-field image is used as the pressing state fusion coordinate system. The pixel brightness value of each pixel position in the pressing state fusion coordinate system is read from the first, second, third, and fourth registered dark-field images. The pixel brightness values ​​of the same pixel position are fused to obtain the pressing state fused pixel value of the corresponding pixel position. The pressing state fused pixel values ​​of all pixel positions are written into the image matrix to obtain the pressing state dark-field scattering development map.

[0042] A scattering response map is generated based on the absolute value of the difference between the same pixel coordinates of the dark field scattering development map and the dark field scattering development map under pressure. The scattering response map refers to an image generated based on the brightness difference at the same pixel position between the dark field scattering development map and the dark field scattering development map under pressure. The pixel values ​​in this image are used to express the degree of change in the scattering state of the bridging neck region before and after pressure. The bright areas in the scattering response map correspond to the positions where the scattering changes are more obvious, and the dark areas correspond to the positions where the scattering changes are weaker. The polarization compression map under pressure refers to an image obtained by polarization compression imaging processing under pressure. The pixel distribution in this image corresponds to the polarization reflection state and compression state of the bridging neck region under pressure conditions.

[0043] Using the pixel coordinate system of the dark field scattering development map as the coordinate system of the scattering response map, the pressed state dark field scattering development map is registered to the pixel coordinate system of the dark field scattering development map. The pressed state pixel value of each pixel position in the scattering response map coordinate system is read in the pressed state dark field scattering development map, and the natural attachment state pixel value of the same pixel position in the dark field scattering development map is read. The pressed state pixel value is subtracted from the natural attachment state pixel value, and the absolute value of the subtraction result is taken to obtain the scattering response pixel value of the corresponding pixel position. The scattering response pixel values ​​of all pixel positions are written into the blank image matrix according to the corresponding pixel coordinates to form the scattering response map.

[0044] Polarization pressure imaging was performed on the bridging neck region under the pressing state to obtain the polarization pressure image under the pressing state. An RFID tamper-evident seal is fixed to the imaging platform, with the bridging neck area positioned at the center of the camera's field of view. A pressing device is used to apply contact pressure to the bridging neck area while maintaining the seal position, camera position, camera focal length, camera aperture, camera exposure time, and camera gain constant. Low-angle dark-field illumination is used to illuminate the bridging neck area from the left, right, top, and bottom sides, respectively, and four pressed dark-field images are acquired under the pressing condition. Image registration is performed on these four images to obtain the first... The first, second, third, and fourth registered dark-field images are registered and pressed together. The pixel coordinate system of the first registered dark-field image is used as the pressing state fusion coordinate system. The pixel brightness value of each pixel position in the pressing state fusion coordinate system is read from the first, second, third, and fourth registered dark-field images. The pixel brightness values ​​of the same pixel position are fused to obtain the pressing state fused pixel value of the corresponding pixel position. The pressing state fused pixel values ​​of all pixel positions are written into the image matrix to obtain the pressing state dark-field scattering development map.

[0045] A polarization response map is generated based on the absolute value of the difference between the same pixel coordinates of the polarization pressure map and the polarization pressure map under pressure conditions. A polarization response map is an image generated based on the brightness difference between a polarization pressure map and a polarization pressure map at the same pixel position. The pixel values ​​in this image are used to express the degree of change in the polarization reflection state of the bridging neck region before and after pressure. The brightness change in the polarization response map corresponds to the changes in the contact state of the conductive layer, the surface flatness, and the local overlap state in the bridging neck region.

[0046] Using the pixel coordinate system of the dark field scattering development map as the coordinate system of the scattering response map, the pressed state dark field scattering development map is registered to the pixel coordinate system of the dark field scattering development map. The pressed state pixel value of each pixel position in the scattering response map coordinate system is read in the pressed state dark field scattering development map, and the natural attachment state pixel value of the same pixel position in the dark field scattering development map is read. The pressed state pixel value is subtracted from the natural attachment state pixel value, and the absolute value of the subtraction result is taken to obtain the scattering response pixel value of the corresponding pixel position. The scattering response pixel values ​​of all pixel positions are written into the blank image matrix according to the corresponding pixel coordinates to form the scattering response map.

[0047] Establish the particle parameter vector for the particle swarm optimization algorithm; wherein the particle parameter vector includes the first fusion weight, the second fusion weight, and the filter control parameters; Particle swarm optimization (PSO) is a data optimization method based on a group cooperative search approach. This method searches for the target result by observing the position and velocity changes of multiple particles in the parameter space. The particle parameter vector is a set of data used to describe the current search state of a single particle. The parameters in the particle parameter vector together correspond to the parameter combination state in a fusion process. The filter control parameter refers to the data used to control the filtering process of the fused image. This parameter is used to determine the degree of local smoothing, edge preservation, and pixel change suppression in the fused image.

[0048] The parameters used to generate the fused image are used as particle position data. A single particle parameter vector is formed by combining the first fusion weight, the second fusion weight, and the filtering control parameters. The first fusion weight controls the proportion of pixel values ​​from the scattering response map participating in the fused image, the second fusion weight controls the proportion of pixel values ​​from the polarization response map participating in the fused image, and the filtering control parameters control the neighborhood range and filtering intensity of the fused image in deterministic filtering. Multiple particle parameter vectors are combined into a particle swarm, and each particle parameter vector is configured with corresponding velocity data, current position data, historical position data, and image evaluation value data, so that each particle parameter vector corresponds to a fusion process of the scattering response map and the polarization response map.

[0049] Based on the first fusion weight and the second fusion weight, the scattering response map and the polarization response map are fused to obtain a fused image.

[0050] Fusion processing refers to the processing method of combining the scattering response map and the polarization response map at the same pixel position according to the first fusion weight and the second fusion weight, so that the scattering change information in the scattering response map and the polarization change information in the polarization response map form a corresponding relationship in the same image; the fused image refers to the image obtained after fusion processing. The pixel values ​​in the image correspond to the comprehensive change state of the bridging neck region before and after pressure. The brightness distribution in the fused image simultaneously reflects the crack edge change, pressing change, local overlap change and conductive layer surface change in the bridging neck region.

[0051] Using the pixel coordinate system of the scattering response map as the coordinate system of the fused image, the polarization response map is registered to the pixel coordinate system of the scattering response map. The scattering response pixel value of each pixel position in the fused image coordinate system is read in the scattering response map, and the polarization response pixel value of the same pixel position in the polarization response map is read. The scattering response pixel value is multiplied by the first fusion weight to obtain the first weighted pixel value, and the polarization response pixel value is multiplied by the second fusion weight to obtain the second weighted pixel value. The first weighted pixel value and the second weighted pixel value are added to obtain the fused pixel value of the corresponding pixel position. The fused pixel values ​​of all pixel positions are written into the blank image matrix according to the corresponding pixel coordinates to form the fused image.

[0052] In an embodiment of the present invention, the particle parameter vector is iteratively updated based on the fused image to obtain a pseudo-closed contact response map, including: Based on the filtering control parameters, deterministic filtering is performed on the fused image to obtain candidate contact response maps; Deterministic filtering refers to a processing method that performs neighborhood operations on pixel data in a fused image based on fixed pixel neighborhood relationships and fixed pixel calculation rules. In this process, the output pixel value corresponding to each pixel position in the fused image is calculated from the pixel data in the neighborhood around that pixel position according to fixed operation relationships. Deterministic filtering is used to reduce random brightness fluctuations, isolated bright spots, and local noise changes in the fused image, so that the continuous contact change area and continuous scattering change area in the bridging neck region form a more stable brightness distribution in the image. The candidate contact response map refers to the image formed after the fused image has undergone deterministic filtering. The pixel values ​​in this image are used to express the degree of local contact change in the bridging neck region between the natural attachment state and the pressing state. The bright areas in the candidate contact response map correspond to the locations in the bridging neck region where contact changes, edge cracking changes, or conductive layer overlap changes are more obvious due to pressure. The dark areas in the candidate contact response map correspond to the locations in the bridging neck region where the changes before and after pressure are weaker. The continuous bright band areas in the candidate contact response map correspond to the local edge cracking and re-pressing areas or the local contact areas of the conductive layer in the bridging neck region.

[0053] Using the pixel coordinate system of the fused image as the coordinate system of the candidate contact response map, the neighborhood range and pixel weight distribution of the pixels participating in the neighborhood operation are determined according to the filtering control parameters. The fused pixel value of each pixel position and its neighboring pixel positions is read in the fused image. The fused pixel value of each neighboring pixel position is multiplied by the corresponding pixel weight to obtain multiple neighborhood weighted pixel values. The multiple neighborhood weighted pixel values ​​are summed to obtain the filtered output pixel value of the pixel position. The filtered output pixel values ​​of all pixel positions are written into the blank image matrix according to the corresponding pixel coordinates to form the candidate contact response map.

[0054] Calculate the image evaluation value of the candidate contact response map; Image evaluation value refers to the data calculated based on the pixel brightness distribution, local brightness gradient change, and region continuity distribution in the candidate contact response map. This data is used to express the degree of separation between the contact change area and the non-contact change area in the candidate contact response map. The larger the image evaluation value, the more concentrated the contact change area in the candidate contact response map, the more continuous the edges, and the more obvious the brightness transition between the local contact area and the background area.

[0055] Using the pixel coordinate system of the candidate contact response map as the evaluation coordinate system, the candidate response pixel value at each pixel position in the candidate contact response map is read. The local gray-level gradient value is calculated based on the difference between the candidate response pixel values ​​between adjacent pixel positions. The local gray-level variance value is calculated based on the difference between the candidate response pixel value and the overall average pixel value of the candidate contact response map. The local gray-level gradient values ​​corresponding to all pixel positions are averaged to obtain the local gray-level gradient mean. The local gray-level variance values ​​corresponding to all pixel positions are averaged to obtain the local gray-level variance mean. The local gray-level gradient mean and the local gray-level variance mean are weighted and summed to obtain the image evaluation value of the candidate contact response map.

[0056] Based on the image evaluation values ​​of the candidate contact response maps, the particle parameter vector is iteratively updated to obtain the pseudo-closed contact response map.

[0057] The pseudo-closed contact response map refers to the candidate contact response map corresponding to the end of the particle parameter vector iteration update. The pixel brightness distribution in this image is used to express the cracked edge re-attachment area, the local conductive overlap area, and the local closed area formed after pressure in the bridging neck region. The continuous bright area in the pseudo-closed contact response map corresponds to the position in the bridging neck region where there is local re-contact or local conductive overlap. The discontinuous brightness area in the pseudo-closed contact response map corresponds to the position in the bridging neck region where only local pressure occurs but no stable contact is formed.

[0058] The image evaluation value of the candidate contact response map corresponding to each particle parameter vector is written into the evaluation record of that particle parameter vector. The image evaluation value is compared with the historical evaluation record of that particle parameter vector in the completed iterations, and the corresponding historical best particle parameter vector is saved. The historical evaluation records of each particle parameter vector are compared and the global best particle parameter vector is saved. The velocity data and position data of each particle parameter vector are updated according to the current position data, velocity data, historical best particle parameter vector, and global best particle parameter vector. The updated particle parameter vector is reused for the fusion processing of the scattering response map and polarization response map, deterministic filtering processing, and image evaluation value calculation. The particle parameter vector update process is continuously completed in the above manner. When the update process is completed, the candidate contact response map corresponding to the global best particle parameter vector is determined as the pseudo-closed contact response map.

[0059] A geometric evidence system generates a normal encoding map, and then obtains a geometric evidence map based on the normal encoding map. In an embodiment of the present invention, generating a normal encoding map includes: Under natural attachment conditions, multi-directional illumination images were acquired in the bridging neck region to obtain a group of multi-directional illumination images. Natural adhesion state refers to the state in which the RFID tamper-evident seal maintains its original adhesive state when not subjected to external pressure, stretching, or peeling. In this state, the conductive layer in the bridging neck area maintains a natural contact relationship with the seal substrate. Multi-directional illumination image acquisition refers to the processing method of projecting light onto the bridging neck area from different illumination directions and acquiring images separately while keeping the camera position and the bridging neck area position unchanged. Different illumination directions correspond to the brightness changes in different directions on the surface of the bridging neck area. Multi-directional illumination image group refers to the collection of images of the bridging neck area acquired under multiple different illumination directions. Each image in the multi-directional illumination image group corresponds to the reflected brightness distribution formed by the bridging neck area under a specific light incident direction.

[0060] The RFID tamper-evident seal is fixed on the imaging platform, with the bridging neck area located at the center of the camera's field of view and the camera's optical axis perpendicular to the seal surface. The camera position, focal length, aperture, exposure time, gain, and seal position remain unchanged. Multiple independently illuminating light sources are set around the bridging neck area, with each light source corresponding to a light incident direction. Each light source is illuminated sequentially while the others are turned off. An image of the bridging neck area is captured when each light source is illuminated. The images of the bridging neck area captured under different light incident directions are numbered and stored according to the position of the light source to form a multi-directional illumination image group.

[0061] Photometric stereo processing is performed on the multi-directional illumination image group to obtain the surface normal vector; Photometric stereoscopic resolution refers to the processing method of calculating the spatial orientation data of the surface of the bridging neck region based on the brightness variation relationship in the multi-directional illumination image group. The surface orientation state corresponding to each pixel position in the bridging neck region is determined by the brightness variation under different illumination directions. The surface normal vector refers to the spatial orientation data of the surface of the bridging neck region at the corresponding pixel position. The data in the surface normal vector is used to represent the tilt direction and tilt degree of the surface of the bridging neck region in space.

[0062] The illumination direction vectors corresponding to the first, second, third, and fourth directional illumination images are obtained. The illumination direction vectors are determined by the spatial position of the illumination source relative to the bridging neck region. At each pixel position of the multi-directional illumination image group, the pixel brightness value of the first, second, third, and fourth directional illumination images is read. A linear brightness equation system is established between each pixel brightness value and the corresponding illumination direction vector. The linear brightness equation system is solved by least squares to obtain the surface direction vector corresponding to the pixel position. The surface direction vector is length normalized to obtain the surface normal vector of the pixel position. The surface normal vectors corresponding to all pixel positions are stored according to the corresponding pixel coordinates to obtain the surface normal vector of the bridging neck region.

[0063] Image encoding is performed on the surface normal vectors to obtain the normal encoded map.

[0064] Image encoding refers to the processing method of converting the spatial direction data in the surface normal vector into image pixel data, so that different surface directions in the bridging neck region can be expressed in different image brightness or different color distribution forms; normal encoding map refers to the image formed by the surface normal vector after image encoding, and the pixel brightness and color changes in the image correspond to the changes in surface tilt direction, crack direction, and surface undulation of the conductive layer in the bridging neck region.

[0065] Using the pixel coordinate system of the multi-directional illumination image group as the coordinate system of the normal encoding map, the surface normal vector corresponding to each pixel position is read. The surface normal vector includes a first direction component, a second direction component, and a third direction component. The value range of the first direction component, the second direction component, and the third direction component is normalized to obtain the first encoding component, the second encoding component, and the third encoding component. The first encoding component is written into the first image channel of the normal encoding map, the second encoding component is written into the second image channel of the normal encoding map, and the third encoding component is written into the third image channel of the normal encoding map. The three encoding components corresponding to all pixel positions are written into the blank image matrix according to the corresponding pixel coordinates to form a normal encoding map used to express the surface orientation distribution of the bridging neck region.

[0066] In an embodiment of the present invention, obtaining a geometric evidence map based on a normal encoding map includes: Under natural attachment conditions, stripe projection images are acquired on the bridging neck region to obtain a stripe projection image sequence. Stripe projection image acquisition refers to the processing method of projecting structured light with a distribution of bright and dark stripes onto the surface of a bridging neck region and acquiring stripe deformation images through a camera. Cracks, warps, pressing positions, and undulations of the conductive layer in the bridging neck region will change the position and shape of the projected stripes in the image. Stripe projection image sequence refers to multiple stripe images of the bridging neck region acquired under different stripe displacement states. Multiple stripe images together record the spatial deformation of the bridging neck region surface caused by the projected stripes.

[0067] The RFID tamper-evident seal is placed on a flat imaging platform and its edges are secured with clamps, ensuring the bridging neck area is centered in the camera's field of view. The camera's optical axis is adjusted to be perpendicular to the seal surface. The stripe projection device is fixed to the side of the camera, ensuring its projection area covers the bridging neck area. Before data acquisition, a blank background image is captured and recorded on the imaging platform. While maintaining the camera position, focal length, aperture, exposure time, gain, seal position, stripe projection device position, and ambient light level, the stripe projection device projects alternating bright and dark stripes onto the bridging neck area. In this design, each projected stripe pattern undergoes spatial displacement relative to the previous projected stripe pattern along the stripe arrangement direction. The displaced stripes still cover the bridging neck region and its adjacent conductive line region. After each stable projection of the stripe pattern, an image of the bridging neck region is acquired by a camera. The background image is subtracted from each acquired bridging neck region image to remove the influence of the imaging platform's background brightness. The background-removed images are arranged according to the stripe pattern projection order to obtain a stripe projection image sequence. Each image in the stripe projection image sequence corresponds to the surface brightness distribution of the same bridging neck region under a stripe spatial displacement state.

[0068] Phase demodulation is performed on the stripe projection image sequence to obtain a phase map; Phase demodulation refers to the processing method of calculating the phase data of the fringe corresponding to the pixel position based on the brightness change relationship of the same pixel position in the fringe projection image sequence. This phase data is used to express the positional offset state of the bridging neck region surface to the projected fringe. The phase map is an image formed by arranging the phase data of each pixel position according to the pixel coordinates. The pixel values ​​in this image correspond to the phase changes of the bridging neck region surface at different positions to the projected fringe.

[0069] Using the pixel coordinate system of the first fringe projection image in the fringe projection image sequence as the phase map coordinate system, image registration is performed on the remaining fringe projection images in the fringe projection image sequence, so that the boundary of the bridging neck region, the edge of the conductive line, and the stripe coverage area in each fringe projection image are mapped to the phase map coordinate system. In the phase map coordinate system, the brightness value of each pixel position in each image of the fringe projection image sequence is read. Multiple brightness values ​​corresponding to the pixel position are arranged into a brightness change sequence according to the projection order of the fringe pattern. The wrapping phase value of the pixel position is calculated according to the periodic relationship of the alternation of brightness and darkness in the brightness change sequence. The wrapping phase values ​​of adjacent pixel positions are continuously expanded to form a continuous distribution of phase values ​​in the bridging neck region. The expanded phase values ​​are used as the phase pixel values ​​of the corresponding pixel positions. The phase pixel values ​​of all pixel positions are written into the blank image matrix according to the phase map coordinate system to obtain the phase map. The pixel values ​​in the phase map are used to represent the degree of spatial offset caused by the surface of the bridging neck region to the projected stripes.

[0070] A linear transformation is performed on the phase map to obtain the height map; Linear transformation refers to the process of converting phase data in a phase map into height data according to a calibration ratio, which is determined by the spatial relationship between the stripe projection device, the camera, and the imaging platform. The height map refers to the image obtained by linear transformation of the phase map, in which the pixel values ​​correspond to the relative heights of different positions on the surface of the bridging neck region.

[0071] The calibration scale coefficients obtained after calibrating the stripe projection device, camera, and imaging platform are acquired. The pixel coordinate system of the phase map is used as the height map coordinate system. The phase pixel value of each pixel position in the height map coordinate system is read in the phase map. The phase pixel value is multiplied by the calibration scale coefficient to obtain the height pixel value of the corresponding pixel position. The height pixel values ​​of all pixel positions are written into the blank image matrix according to the corresponding pixel coordinates to obtain the height map. Each pixel value in the height map corresponds to the height data of the surface of the bridging neck region relative to the reference plane of the imaging platform.

[0072] Image encoding is performed on the height map to obtain a height-coded map; A height-coded image is an image formed by converting the height data in a height map into the image's brightness or color channels. This image is used to represent the location distribution of surface steps, depressions, bulges, and pressed planes in the bridging neck region.

[0073] Using the pixel coordinate system of the height map as the coordinate system of the height encoding map, the height pixel value of each pixel position in the height map is read. The height pixel value is normalized according to the maximum and minimum values ​​of all height pixel values ​​in the height map to obtain the height encoded pixel value of the corresponding pixel position. The height encoded pixel values ​​of all pixel positions are written into the blank image matrix according to the corresponding pixel coordinates to form a single-channel height encoded map, or the height encoded pixel values ​​are mapped to multiple image channels to form a multi-channel height encoded map.

[0074] The normal encoding map and the height encoding map are spliced ​​together in a fixed format to obtain the geometric evidence map.

[0075] Fixed-format stitching refers to the process of combining multiple images into one image according to a predetermined image layout; geometric evidence image refers to an image formed by stitching together a normal encoding image and a height encoding image, which simultaneously expresses the directional and height changes of the surface of the bridging neck region.

[0076] Using the pixel coordinate correspondence of the bridging neck region in the normal encoding map and the height encoding map as the basis for stitching, the image size of the normal encoding map and the height encoding map is unified to make them have the same image height and the same display ratio of the bridging neck region. A blank stitched image is created, and the normal encoding map is written into the first image area of ​​the blank stitched image, and the height encoding map is written into the second image area of ​​the blank stitched image. A separating pixel band is set between the first image area and the second image area. The image name data corresponding to the normal encoding map and the image name data corresponding to the height encoding map are written into the edge marker positions of the corresponding image areas. The stitched image is used as a geometric evidence map, so that the geometric evidence map simultaneously contains the surface orientation distribution information and surface height distribution information of the bridging neck region.

[0077] The on-chain sharing system obtains image digest barcode images based on geometric evidence graphs and pseudo-closed contact response graphs, and performs trusted data sharing of supply chain traceability events based on the image digest barcode images.

[0078] In an embodiment of the present invention, an image digest barcode image is obtained based on a geometric evidence map and a pseudo-closed contact response map, including: By splicing the pseudo-closed contact response map and the geometric evidence map in a fixed format, the first credible shared visual credential map is obtained. The first trusted shared visual credential image refers to an image file formed by splicing pseudo-closed contact response images and geometric evidence images in a fixed format. This image file contains contact change image information of the bridging neck area before and after pressing, as well as image information of the direction and height changes of the surface of the bridging neck area. The pseudo-closed contact response image is used to express the positional distribution of edge pressing, local overlap, and contact state changes in the bridging neck area. The geometric evidence image is used to express the positional distribution of uplifts, depressions, steps, and surface tilts in the bridging neck area. The first trusted shared visual credential image stores the above two types of image content through the same image carrier, so that the contact response data and geometric shape data corresponding to the same RFID tamper-evident seal form a verifiable combined record in space.

[0079] Using the pixel coordinate system of the pseudo-closed contact response map and the pixel coordinate system of the geometric evidence map as the image source coordinate system, the pseudo-closed contact response map and the geometric evidence map are subjected to size unification processing to make the bridging neck area in the pseudo-closed contact response map and the geometric evidence map have the same display ratio. A blank voucher image is created, and the pseudo-closed contact response map is written into the first image area of ​​the blank voucher image, and the geometric evidence map is written into the second image area of ​​the blank voucher image. A separating pixel band is set between the first image area and the second image area. The image name data corresponding to the pseudo-closed contact response map and the image name data corresponding to the geometric evidence map are written into the edge marker position of the corresponding image area to obtain the first trusted shared visual voucher image.

[0080] The image digest is obtained by performing digest calculation on the first trusted shared visual credential image. The digest calculation process refers to the hash operation performed on the image data of the first trusted shared visual credential image. This image data includes the pixel arrangement order, pixel values ​​at each pixel position, and image size data of the first trusted shared visual credential image. The digest calculation process converts the above image data into a fixed-length data string according to a determined hash operation rule, so that when any pixel value, pixel position, or image content in the first trusted shared visual credential image changes, the corresponding data string changes synchronously. The image digest refers to the fixed-length data string obtained by the digest calculation process. This data string corresponds to the image content of the first trusted shared visual credential image and is used to represent the image data state of the first trusted shared visual credential image when it was generated. The image digest itself does not contain the image content of the bridging neck area that can be directly viewed, but serves as verification data for whether the first trusted shared visual credential image retains its original image data state.

[0081] Read the image size data, pixel channel data, and pixel value data of each pixel position of the first trusted shared visual credential image. Convert the image size data, pixel channel data, and pixel value data into continuous byte data according to the pixel arrangement order in the first trusted shared visual credential image. Perform a hash operation on the continuous byte data to obtain a fixed-length digest string. Determine the fixed-length digest string as the image digest. The image digest corresponds to the image data state of the first trusted shared visual credential image during digest calculation and processing.

[0082] The image digest is encoded using a two-dimensional barcode to obtain the image digest barcode image.

[0083] Two-dimensional barcode encoding refers to the process of converting character data in an image digest into a two-dimensional pixel array composed of light and dark modules according to the encoding rules of two-dimensional barcodes. This process converts the character order, character content, and verification information in the image digest into a barcode graphic that can be recognized by an image reading device. An image digest barcode image is an image obtained by encoding an image digest using two-dimensional barcodes. This image is composed of light and dark pixel modules arranged according to the rules of two-dimensional barcodes. The arrangement of the light and dark pixel modules corresponds to the character data in the image digest. The image digest barcode image can be embedded into subsequent images as a visual summary carrier of the first trusted shared visual credential image.

[0084] The character data in the image digest is converted into a binary data stream in character order. Error correction encoding is performed on the binary data stream to obtain an encoded data stream with error correction data. The encoded data stream with error correction data is written into the data area of ​​the two-dimensional barcode matrix according to the data filling rules of the two-dimensional barcode. The positioning pattern, alignment pattern and format information of the two-dimensional barcode are written into the corresponding area of ​​the two-dimensional barcode matrix. Bright and dark pixel modules are generated according to the data values ​​of each matrix unit in the two-dimensional barcode matrix. All bright and dark pixel modules are arranged into an image according to the row and column positions of the two-dimensional barcode matrix to obtain the image digest barcode image.

[0085] In embodiments of the present invention, reliable data sharing of supply chain traceability events based on image digest barcode images includes: The image digest barcode image is embedded into the embedding region of the first trusted shared visual credential image to obtain the second trusted shared visual credential image; The embedded region refers to the image area reserved in the first trusted shared visual credential image for writing the image digest barcode image, and this region corresponds to the fixed pixel position in the first trusted shared visual credential image; the second trusted shared visual credential image refers to the image formed after embedding the image digest barcode image into the first trusted shared visual credential image, and this image simultaneously contains contact change information, geometric morphology information and corresponding image digest barcode information of the bridging neck region.

[0086] Read the image size data and image channel data of the first trusted shared visual credential image. Determine the starting pixel position and ending pixel position of the embedding according to the reserved embedding area coordinates in the first trusted shared visual credential image. Adjust the size of the image digest barcode image so that the width and height of the image digest barcode image are the same as the width and height of the embedding area. Write the adjusted image digest barcode image pixel by pixel into the embedding area of ​​the first trusted shared visual credential image. Keep the pixel data of the first trusted shared visual credential image outside the embedding area unchanged to obtain the second trusted shared visual credential image.

[0087] Identify the supply chain traceability events corresponding to RFID tamper-evident seals; A supply chain traceability event refers to a logistics flow, warehousing operation, transportation operation, handover operation, or acceptance operation corresponding to an RFID tamper-evident seal. This event corresponds to a business record of the RFID tamper-evident seal during the supply chain flow process.

[0088] Read the RFID tamper-evident seal identifier, obtain the batch identifier, goods identifier, flow node identifier, operation entity identifier, and business operation type data associated with the seal identifier, combine the seal identifier, batch identifier, goods identifier, flow node identifier, operation entity identifier, and business operation type data to form basic event data, perform event number generation processing on the basic event data, and obtain the supply chain traceability event corresponding to the RFID tamper-evident seal.

[0089] Write the identifiers of supply chain traceability events into the blockchain ledger; The identifier of a supply chain traceability event refers to the data content used to uniquely identify a supply chain traceability event. This data content includes the time data, location data, operation object data, or flow node data corresponding to the supply chain traceability event.

[0090] Obtain the event number, seal identifier, batch identifier, goods identifier, flow node identifier, operating entity identifier, and business operation type data corresponding to the supply chain traceability event. Generate event-on-chain data by matching the event number, seal identifier, batch identifier, goods identifier, flow node identifier, operating entity identifier, and business operation type data with the field names and values. Perform hash calculation on the event-on-chain data to obtain the event summary. Write the event-on-chain data and event summary into the blockchain transaction data. Submit the blockchain transaction data to the blockchain ledger to form a blockchain record.

[0091] Based on the second trusted shared visual credential diagram and the blockchain ledger, data is shared in a trusted manner for supply chain traceability events.

[0092] A blockchain ledger refers to a data record structure formed by connecting multiple blocks in chronological order. Each block stores data records corresponding to supply chain traceability events, and subsequent blocks are connected to previous blocks through hash association. Trusted data sharing refers to a data sharing method that establishes a correspondence between a second trusted shared visual credential image and the supply chain traceability event data in the blockchain ledger. This enables different supply chain nodes to read the traceability data corresponding to the same RFID tamper-evident seal based on the image content in the second trusted shared visual credential image and the event records in the blockchain ledger.

[0093] The system acquires and reads the image digest barcode image from the second trusted shared visual credential image. It then reads the blockchain record corresponding to the supply chain traceability event from the blockchain ledger, obtains the event-on-chain data and event digest from the blockchain record, and determines the seal identifier, batch identifier, goods identifier, circulation node identifier, operating entity identifier, and business operation type data corresponding to the supply chain traceability event based on the event-on-chain data. The second trusted shared visual credential image, image digest, event-on-chain data, and blockchain record are provided as shared data to the supply chain participating nodes, enabling them to read the traceability event data corresponding to the same RFID tamper-evident seal based on the data recorded in the blockchain ledger.

[0094] It should be noted that during the entire supply chain traceability process, after a genuine breakage of the bridging neck conductive structure inside the RFID tamper-evident seal, temporary conductivity may be created due to metal edge curling, localized rebound of the conductive layer, or external pressure. This allows the seal to still be identified by the RFID reader despite the internal breakage, resulting in an inconsistency between the physical failure state of the seal and the continuous status of the identity on the blockchain. Relying solely on the supply chain traceability event records in the blockchain ledger cannot prove the authenticity and reliability of the physical state of the seal at the time of data writing. However, the second trusted shared visual credential image simultaneously retains the scattering change image, surface geometry image, and image summary barcode information of the bridging neck region. This allows for a visual record of the seal's internal breakage but partial re-attachment and conductivity, and the corresponding supply chain traceability event data and summary data are stored in the blockchain ledger. This enables supply chain participating nodes to simultaneously read the physical state evidence of the seal and the on-chain event records, thereby achieving trusted data sharing of supply chain traceability events.

[0095] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A blockchain-based supply chain whole-process traceability and data trustable sharing system, characterized in that, include: The dark field development system generates a dark field scattering development map based on the bridging neck area corresponding to the RFID tamper-evident seal. The polarization overlay system generates a total intensity image based on the first and third registered polarization images of the bridging neck region, and obtains a polarization overlay map based on the total intensity image; The pseudo-closed extraction system generates a fused image based on the dark field scattering development map and the polarization overlay map, and iteratively updates the particle parameter vector according to the fused image to obtain the pseudo-closed contact response map; A geometric evidence system generates a normal encoding map, and then obtains a geometric evidence map based on the normal encoding map. The on-chain sharing system obtains image digest barcode images based on geometric evidence graphs and pseudo-closed contact response graphs, and performs trusted data sharing of supply chain traceability events based on the image digest barcode images. 2.The blockchain-based supply chain whole-process traceability and data trustable sharing system according to claim 1, wherein, Based on the bridging neck region corresponding to the RFID tamper-evident seal, a dark field scattering imaging map is generated, including: Identify the bridging neck area corresponding to the RFID tamper-evident seal; Acquire the first, second, third, and fourth dark-field images of the bridging neck region; Image registration is performed on the first dark-field image, the second dark-field image, the third dark-field image, and the fourth dark-field image to obtain the first registered dark-field image, the second registered dark-field image, the third registered dark-field image, and the fourth registered dark-field image; Based on the first registered dark-field image, the second registered dark-field image, the third registered dark-field image, and the fourth registered dark-field image, fused pixel values ​​are generated; A dark field scattering development map is generated based on the fused pixel values.

3. The blockchain-based supply chain end-to-end traceability and trusted data sharing system according to claim 2, characterized in that, A total intensity image is generated based on the first and third registered polarization images of the bridging neck region, including: Linear polarization images were acquired in the bridging neck region to obtain the first polarization image, the second polarization image, the third polarization image, and the fourth polarization image; Image registration is performed on the first polarization image, the second polarization image, the third polarization image, and the fourth polarization image to obtain the first registered polarization image, the second registered polarization image, the third registered polarization image, and the fourth registered polarization image; The total intensity image is generated by adding the pixel values ​​of the first and third registered polarization images.

4. The blockchain-based supply chain end-to-end traceability and trusted data sharing system according to claim 3, characterized in that, The polarization overlay map is obtained based on the total intensity image, including: A first polarization difference image is generated based on the subtraction of pixel values ​​between the first and third registered polarization images. A second polarization difference image is generated based on the pixel values ​​of the second and fourth registered polarization images. A linear polarization degree image is generated based on the first polarization difference image, the second polarization difference image, and the total intensity image. Channel mapping fusion is performed on the linear polarization degree image and the total intensity image to obtain a polarization overlay map.

5. A blockchain-based supply chain end-to-end traceability and trusted data sharing system according to claim 4, characterized in that, A fused image is generated based on the dark field scattering development map and the polarization overlay map, including: Dark-field azimuth complementary acquisition and scattering development fusion processing were performed on the bridging neck region under the holding state to obtain the dark-field scattering development map under the holding state. A scattering response map is generated based on the absolute value of the difference between the same pixel coordinates of the dark field scattering development map and the dark field scattering development map under pressure. Polarization pressure imaging was performed on the bridging neck region under the pressing state to obtain the polarization pressure image under the pressing state. A polarization response map is generated based on the absolute value of the difference between the same pixel coordinates of the polarization pressure map and the polarization pressure map under pressure conditions. Establish the particle parameter vector for the particle swarm optimization algorithm; wherein the particle parameter vector includes the first fusion weight, the second fusion weight, and the filter control parameters; Based on the first fusion weight and the second fusion weight, the scattering response map and the polarization response map are fused to obtain a fused image.

6. A blockchain-based supply chain end-to-end traceability and trusted data sharing system according to claim 5, characterized in that, Based on the fused image, the particle parameter vector is iteratively updated to obtain a pseudo-closed contact response map, including: Based on the filtering control parameters, deterministic filtering is performed on the fused image to obtain candidate contact response maps; Calculate the image evaluation value of the candidate contact response map; Based on the image evaluation values ​​of the candidate contact response maps, the particle parameter vector is iteratively updated to obtain the pseudo-closed contact response map.

7. A blockchain-based supply chain end-to-end traceability and trusted data sharing system according to claim 6, characterized in that, Generate a normal encoding map, including: Under natural attachment conditions, multi-directional illumination images were acquired in the bridging neck region to obtain a group of multi-directional illumination images. Photometric stereo processing is performed on the multi-directional illumination image group to obtain the surface normal vector; Image encoding is performed on the surface normal vectors to obtain the normal encoded map.

8. A blockchain-based supply chain end-to-end traceability and trusted data sharing system according to claim 7, characterized in that, The geometric evidence map is obtained based on the normal encoding map, including: Under natural attachment conditions, stripe projection images are acquired on the bridging neck region to obtain a stripe projection image sequence. Phase demodulation is performed on the stripe projection image sequence to obtain a phase map; A linear transformation is performed on the phase map to obtain the height map; Image encoding is performed on the height map to obtain a height-coded map; The normal encoding map and the height encoding map are spliced ​​together in a fixed format to obtain the geometric evidence map.

9. A blockchain-based supply chain end-to-end traceability and trusted data sharing system according to claim 8, characterized in that, Image digest barcode images are obtained based on geometric evidence maps and pseudo-closed contact response maps, including: By splicing the pseudo-closed contact response map and the geometric evidence map in a fixed format, the first credible shared visual credential map is obtained. The image digest is obtained by performing digest calculation on the first trusted shared visual credential image. The image digest is encoded using a two-dimensional barcode to obtain the image digest barcode image.

10. A blockchain-based supply chain end-to-end traceability and trusted data sharing system according to claim 9, characterized in that, Based on image digest barcode images, trusted data sharing is achieved for supply chain traceability events, including: The image digest barcode image is embedded into the embedding region of the first trusted shared visual credential image to obtain the second trusted shared visual credential image; Identify the supply chain traceability events corresponding to RFID tamper-evident seals; Write the identifiers of supply chain traceability events into the blockchain ledger; Based on the second trusted shared visual credential diagram and the blockchain ledger, data is shared in a trusted manner for supply chain traceability events.

Images