WO2022263920A1 - A system that provides solutions by establishing invisible bridges in physical objects/elements using laser marking technique - Google Patents

A system that provides solutions by establishing invisible bridges in physical objects/elements using laser marking technique Download PDF

Info

Publication number
WO2022263920A1
WO2022263920A1 PCT/IB2021/058789 IB2021058789W WO2022263920A1 WO 2022263920 A1 WO2022263920 A1 WO 2022263920A1 IB 2021058789 W IB2021058789 W IB 2021058789W WO 2022263920 A1 WO2022263920 A1 WO 2022263920A1
Authority
WO
WIPO (PCT)
Prior art keywords
unique
server
coded watermark
serial number
watermark
Prior art date
Application number
PCT/IB2021/058789
Other languages
French (fr)
Inventor
Serra DEMİR
Mete KAYALAR
Original Assignee
Cosmodot Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from TR2021/009908 external-priority patent/TR2021009908A2/en
Application filed by Cosmodot Inc. filed Critical Cosmodot Inc.
Publication of WO2022263920A1 publication Critical patent/WO2022263920A1/en

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N1/00—Scanning, transmission or reproduction of documents or the like, e.g. facsimile transmission; Details thereof
    • H04N1/32—Circuits or arrangements for control or supervision between transmitter and receiver or between image input and image output device, e.g. between a still-image camera and its memory or between a still-image camera and a printer device
    • H04N1/32101—Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title
    • H04N1/32144—Display, printing, storage or transmission of additional information, e.g. ID code, date and time or title embedded in the image data, i.e. enclosed or integrated in the image, e.g. watermark, super-imposed logo or stamp
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Definitions

  • Present invention relates to a deep neural network system that learns the encode/decode algorithm that works compatible with the file formats of vector and/or all laser marking machines placed on all objects and elements (physical) that can be laser marked using the laser marking technique; provides solutions to the problems of traceability, recycling, sustainability, measurement, resistance to external conditions (dust, hot cold temperature differences), lighting, shadows, perspective, closure and viewing distance, thanks to the unique digital coded watermark placed on the products, independent of the design and size; and provides robust encoding and decoding of arbitrary bridge multi-bit sequences by overlaying objects/elements in a way that approaches perceptual invisibility; and that learns an encoding/decoding algorithm that is a learned steganographic algorithm being resistant to distortions caused by real production and external conditions.
  • Present invention also relates to a system that provides real-time decoding of hidden hyperlinks to objects/elements and connecting to remote resources.
  • marking is performed only in an area of 8 mm square, and sometimes the marking disappears due to the small dimensions of the product and sometimes the temperature differences undergone by the product.
  • a system related to a data network and used for physical or digital objects to connect with associated remote resources is disclosed in the patent application document no. TR2020/08748 mentioned in the state of the art.
  • the system disclosed in the patent application document numbered TR2020/08748 and the operating structures of the new invention according to this disclosure are quite similar.
  • work is carried out on the watermark, and in the new invention according to the description, the unique digital coded watermark is managed.
  • a unique digital coded watermarked file is produced and the file is distributed to predetermined parties.
  • the unique digitally coded watermark is specially created for at least one object of any type and for every different type of template with laser marking technique, such as metals (aluminum, steel, zinc, lead, copper, etc.), platinum, diamond, gold, palladium, bronze, tin, tungsten, rhodium, brass, titanium, cobalt, glass, ceramic, rubber, plastic (polyethylene, etc.), stone (marble) , granite, travertine, basalt and andesite), wood, cement, aggregates, brick, clay, concrete, PVC ground plane, vehicle tire and so on.
  • Unique digitally coded watermark is composed of many dots that are difficult to see with bare eyes.
  • the unique digitally coded watermark is a special array matrix and is applied by laser marking technique.
  • the unique digitally coded watermark contains a mathematical algorithm consisting of at least one and each printed dots with their own unique sequence, making it clear at which point of which template the design is made. For all these reasons, the system described in the patent application document numbered TR2020/08748 and the invention according to this description have differences.
  • Patent application document numbered W02019067730A1 belonging to Digimarc Corporation mentioned in the state of the art discloses the watermark detection methods and relevant embodiments.
  • the watermark detection methods and embodiments in the relevant patent application document are carried out in different ways from the invention that is the subject of this specification.
  • the unique digitally coded watermark specially created for at least one object of any type and for every different type of template with laser marking technique, such as metals (aluminum, steel, zinc, lead, copper, etc.), platinum, diamond, gold, palladium, bronze, tin, tungsten, rhodium, brass, titanium, cobalt, glass, ceramic, rubber, plastic (polyethylene, etc.), stone (marble) , granite, travertine, basalt and andesite), wood, cement, aggregates, brick, clay, concrete, PVC ground plane, vehicle tire and so on.
  • metals aluminum, steel, zinc, lead, copper, etc.
  • platinum diamond
  • the invention is inspired by current situations and aims to solve the above- mentioned disadvantages.
  • Object of the present invention is to provide a deep neural network system that learns the encode/decode algorithm that works compatible with the file formats of vector and/or all laser marking machines placed on all objects and elements (physical) that can be laser marked using the laser marking technique; provides solutions to the problems of traceability, recycling, sustainability, measurement, resistance to external conditions (dust, hot cold temperature differences), lighting, shadows, perspective, closure and viewing distance, thanks to the unique digital coded watermark placed on the products, independent of the design and size; and provides robust encoding and decoding of arbitrary bridge multi-bit sequences by overlaying objects/elements in a way that approaches perceptual invisibility; and that learns an encoding/decoding algorithm that is a learned steganographic algorithm being resistant to distortions caused by real production and external conditions.
  • Another object of the present invention is to set forth a system that provides real time decoding of hidden hyperlinks to objects/elements and connecting to remote resources.
  • the system according to the invention can detect all analog inputs formed on all surfaces and floors with a unique digital coded watermark compatible with vector and/or file formats of all laser machines, placed on all objects and elements (physical) that can be laser marked using laser marking technique, these inputs and the pattern on the floor where the located inputs are perceived, either based on the region(s) or independent of the region(s).
  • the transmission of unique digitally coded watermark and input information or detection is conducted by means of a smartphone and/or a camera in the transmission terminal that only reads the patterns.
  • the transmitted unique digital coded watermark information is interpreted and processed thereof.
  • a data network was created in the system according to the invention in order to establish a connection between objects and elements, related resources and remote resources, and to assign a unique identity.
  • the system has a structure that is used to connect related and physical objects with relevant remote resources.
  • System is used to connect any object to a corresponding network or local resource.
  • System comprises four basic processes; recording, embedding, detection and responding.
  • Figure 1 is the schematic illustration of the system according to the invention.
  • Present invention relates to a deep neural network system (1) that learns the encode/decode algorithm that works compatible with the file formats of vector and/or all laser marking machines placed on all objects and elements (A) (physical) that can be laser marked using the laser marking technique; provides solutions to the problems of traceability, recycling, sustainability, measurement, resistance to external conditions (dust, hot cold temperature differences), lighting, shadows, perspective, closure and viewing distance, thanks to the unique digital coded watermark (C) placed on the products, independent of the design and size; and provides robust encoding and decoding of arbitrary bridge multi-bit sequences by overlaying objects/elements (A) in a way that approaches perceptual invisibility; and that learns an encoding/decoding algorithm that is a learned steganographic algorithm being resistant to distortions caused by real production and external conditions.
  • A physical
  • C unique digital coded watermark
  • Present invention also relates to a system (1 ) that provides real-time decoding of hidden hyperlinks to objects/elements (A) and connecting to remote resources (B).
  • System (1 ) comprises of many new features and applications.
  • system (1 ) is used for connecting a laser-marked product to an electronic content.
  • Technology used in the system (1 ) is not limited to these features.
  • the system (1 ) is used to connect any object (A) (physical) to a corresponding network or remote resource (B).
  • the system (1 ) according to the invention schematically illustrated in Figure 1 includes the following;
  • ⁇ starting device (2) which detects data from an input object (A), processes this data and extracts the unique digitally coded watermark information (C) inside the object (A);
  • the router (3) which receives the processed data from the starting device (2) and processes it thereof, records the information about the processed data in the database (6) and transmits at least a certain part of this processed data;
  • ⁇ product processor (4) that receives at least a certain part of the processed data from the router (3), which is distinguished and detected by the router (3), and by checking this data, compares the unique digital coded watermark (C) information obtained from the object (A) with the information belonging to the previously recorded object (A), and enables the establishment of a connection between object (A) and remote resources (B) as a result of this comparison;
  • ⁇ server (5) which saves the content of the identity of the object (A) and the product's identification information, in order to assign an identity to the object (A) and associate this identity with one or more responses;
  • ⁇ database (6) that stores the data recorded by the server (5) and enables this data to be used by the product processor (4).
  • System (1 ) comprises four basic processes; recording, embedding, detection and responding.
  • the unique digitally coded watermark (C) is first recorded in the database (6) by the server (5).
  • the server (5) must save the identity information of the object (A) in relation to the unique digitally coded watermark (C) in the database (6) for the system (1 ) to identify the response (e.g., a URL) corresponding to an object (A) identifier related to a unique digitally coded watermark (C).
  • the server (5) captures some basic identification information in the registration process of the unique digitally coded watermark (C). This information is then used to verify the incoming message, and the associated information/action is defined thereof.
  • Identity information includes these examples:
  • the object (A) defined by the server (5) to the database (6) and the associated attributes include the object (A) watermarked with their unique digital coded watermark (C).
  • the object (A) features defined by the server (5) to the database (6) generally include the unique identification number of the object (A), its name, description, and a list of accounts authorized to access the record of the object (A).
  • Action defines the response intended by the user when the unique digitally coded watermark (C) is detected.
  • Response may vary by product, but the explanatory edit includes returning some additional information about the object (A) with the unique digitally coded watermark (C).
  • the action in the system (1 ) is to return a URL or HTML or object (A) identification code that will be used to display a web page associated with the object (A) with the unique digitally coded watermark (C).
  • Registered unique serial number is the last component of the registration.
  • the registered unique serial number is the assigned product identifier and the product- specific identifier, allowing the system (1 ) to obtain the specific information/action for the object (A) in question.
  • Recording refers to the process of assigning an identity to an object (A) and associating that identity with a corresponding action or response. Recording process may include additional steps such as registering the registrant's name and/or organization, product name and product-specific identifier and so on.
  • the server (5) is configured to generate the file with a unique digitally coded watermark (C) and distribute the file to predetermined parties.
  • the unique digitally coded watermark (C) is specially created for at least one object (A) of any type and for every different type of template with laser marking technique, such as metals (aluminum, steel, zinc, lead, copper, etc.), platinum, diamond, gold, palladium, bronze, tin, tungsten, rhodium, brass, titanium, cobalt, glass, ceramic, rubber, plastic (polyethylene, etc.), stone (marble) , granite, travertine, basalt and andesite), wood, cement, aggregates, brick, clay, concrete, PVC ground plane, vehicle tire and so on.
  • Unique digitally coded watermark (C) is composed of many dots that are difficult to see with bare eyes.
  • the unique digitally coded watermark (C) is a special array matrix and is applied by laser marking technique.
  • the unique digitally coded watermark (C) contains a mathematical algorithm consisting of at least one and each printed dots with their own unique sequence, making it clear at which point of which template the design is made.
  • the key code of the unique digital coded watermark (C) is primarily used to create the constellation cell, which will be processed into the picture with an error correction system based on the Reed- Solomon algorithm.
  • the server (5) makes some geometric additions to the content of this cell in order to increase the readability of the code by the browser.
  • the server (5) decides the footprint occupied by this cell in the picture, depending on the resolution provided by the print or digital media to be used at this stage.
  • the server (5) blends the repetitive constellation pattern produced based on this decision with a high transparency on the original image. With this process, the server (5) optionally creates alterations in saturation or brightness that cannot be captured with bare eyes. In this way, the server (5) generates the file with the unique digital coded watermark (C) through the scanning application in the constellation pattern and distributes the file to the predetermined parties.
  • C unique digital coded watermark
  • the server (5) is also configured to generate an encapsulation file and distribute the file to predetermined parties.
  • Embedding is performed after recording. Embedding refers to the encoding process of a digital identifier object (A) (for example, a unique digitally coded watermark (C) that carries a unique serial number in its payload).
  • A digital identifier object
  • C unique digitally coded watermark
  • the unique digitally coded watermark (C) embedded in the object (A) is detected by the starting device (2).
  • Detection is the complementary process of embedding; that is, it enables a digital identifier to be distinguished from an object (A).
  • Starting device (2) is used by the user.
  • Starting device (2) can have a camera.
  • Starting device (2) may be a wireless mobile device such as a mobile phone, tablet, and the like.
  • Starting device (2) can have 2D image sensor.
  • Starting device (2) may include a steganographic unique digitally coded watermark (C) detector configured to decipher steganographically encoded information in physical objects (A).
  • C steganographic unique digitally coded watermark
  • Starting device (2) analyzes the object (A) and extracts the unique digitally coded watermark (C) embedded in the object (A). Starting device (2) then transmits a certain portion or all of the data of this embedded unique digitally coded watermark (C) to the router (3).
  • Router (3) decodes the incoming data and analyzes the product identification information and product information in the data. Router (3) then compares this information with the data in the database (6). Router (3) then transfers this information to the relevant product processor (4). Responding is performed finally in the system (1 ). Responding refers to the action taken based on the distinguished identifier.
  • the product processor (4) receives the transmitted message from the router (3) and makes effort to match the unique serial number of the detected unique digitally coded watermark (C) with the unique serial number of a registered unique digitally coded watermark (C) previously stored in the database (6). If a match is found, the product processor (4) performs the intended action. For example, these actions include returning a URL for web redirection, serving an HTML page for initial user navigation, initiating software downloads, and so on. If no match is found, the product processor (4) sends an error code and message to the starting device (2). If a match is found, but the corresponding action is unavailable, incomplete, inactive, or invalid, the product processor (4) sends an error code and message to the starting device (2).
  • the product processor (4) processes the requisitions received from the starting device (2) via the internet and the router (3) and transmits the requested information/actions to the remote source (B).
  • the requested information is the URL and this URL is associated with the unique digitally coded watermark (C) payload transferred by the starting device (2).
  • C unique digitally coded watermark
  • Product processor (4) is configurated to perform the following process steps:
  • ⁇ verify the received identifier (for example, unique digital coded watermark (C) unique serial number) against the list of active identifiers, if the unique serial number is not found, return an error message to the calling session and log the error in an error handling routine;
  • C unique digital coded watermark
  • ⁇ respond to a valid registration request package upload request (appropriate format, an existing unique serial number, an account ID, and a valid corresponding account password) by returning a valid registration package for a unique digitally coded watermark (C) unique serial number provided;
  • Perspective Warp uses four different test models: Perspective Warp, Motion and Focal Blur, Color Manipulation, and Vibration and Shaking. 1 .
  • Perspective Warp uses four different test models: Perspective Warp, Motion and Focal Blur, Color Manipulation, and Vibration and Shaking. 1 .
  • Perspective Warp uses four different test models: Perspective Warp, Motion and Focal Blur, Color Manipulation, and Vibration and Shaking. 1 .
  • Server (5) associates any two images of the same planar surface with a holography with the camera model assumption.
  • Server (5) generates a random homography to simulate the effect of a camera not fully aligned with the encoded image marker.
  • Server (5) keeps the four vertex positions of the marker within a fixed range (max ⁇ 40 pixels, i.e., ⁇ 10%) to sample a homography, and then decodes the homography; mapping the original vertices to their new positions.
  • Server (5) therefore re-samples the original image in two-cycle in order to create the perspective warped image.
  • Blur can be caused by both camera movement and incorrect autofocus.
  • Server (5) samples a random angle to simulate motion blur and generates a straight-line blur kernel with a width between 3 and 7 pixels.
  • Server (5) uses a randomly sampled Gaussian blur kernel with a standard deviation of 1 to 3 pixels to simulate misfocusing.
  • Displays have a limited gamut compared to the full RGB color space. Cameras manipulate their output using exposure settings, white balance, and a color correction matrix. Server (5) estimates these disruptions with a sequence of random Ran color reversal (stable in all the image) as follows.
  • server (5) assumes standard non-photon starved imaging conditions using a Gaussian noise model (by sampling the standard deviation).
  • the server (5) uses image noise (s ⁇ U[0, 0.2]) to account.
  • copy detection technology such as fragile unique digitally code watermarks (C), authentication hashes embedded in a unique digitally code watermark (C), or special authentication keys used to generate the unique digitally code watermark (C) can be used to authenticate the printed object (A).
  • the unique digitally coded watermark (C) may carry information used to access and index information in a database (6) or computer network as described above.
  • a direct, undetectable code is embedded in the elements / objects (A).
  • this code automatically redirects an associated web browser to a destination selected by the print media manufacturer.
  • This target for example, a web page may provide additional information or services that are timelier and/or more comprehensive than those provided by printed material.
  • All objects and elements (A) (physical or electronic) with a unique digitally coded watermark (C) in vector and/or all laser machine file formats placed with laser marking can be linked to URL/Links.
  • AR representation can be made on all objects and elements (A) (physical) with a unique digital coded watermark (C) in vector and/or all laser machine file formats placed with laser marking.
  • Augmented reality screenings can be made live and/or anime (static or dynamic).
  • C All objects and elements (A) (physical) with unique digitally coded watermarks (C) in vector and/or all laser machine file formats placed with laser marking can be counted in warehouses, at salespoints on production lines, etc. (On Shelf and Warehouse)
  • Product traceability can be made on all objects and elements (A) (physical) with a unique digital coded watermark (C) in vector and/or all laser machine file formats placed with laser marking.
  • Product recycling information can be achieved through all objects and elements (A) (physical) with a unique digital coded watermark (C) in vector and/or all laser machine file formats placed with laser marking.
  • Production optimization can be achieved by associating sustainable information on all objects and elements (A) (physical) with a unique digital coded watermark (C) in vector and/or all laser machine file formats placed with laser marking.
  • All product-related information can be accessed through all objects and elements (A) (physical) with a unique digital coded watermark (C) in vector and/or all laser machine file formats placed with laser marking.
  • Bill of Materials information associated with the product can be accessed through all objects and elements (A) (physical) with a unique digitally coded watermark (C) in vector and/or all laser machine file formats placed with laser marking. i. The authenticity of the product can be checked over all objects and elements (A) (physical) with a unique digital coded watermark (C) in vector and/or all laser machine file formats placed with laser marking.
  • J. T raceability, sustainability and originality follow-ups can be achieved in all sectors below through all objects and elements (A) (physical) with a unique digitally coded watermark (C) in vector and/or all laser machine file formats placed with laser marking.
  • A (physical) with a unique digitally coded watermark (C) in vector and/or all laser machine file formats placed with laser marking.
  • a. Automotive Industry and Parts b. Passenger and Commercial Vehicle Tires c. Display d. Computer Parts and Chips e. Smart Factories (Industry 4.0)
  • f. Consumer Durables and White Appliances g. Aerospace h. Commercial Vehicles and Transportation i. Tool and Mold Manufacturing j. Construction Industry and Building Materials k. Electronic Products and Spare Parts I. Machine and System Manufacturing m. Air Conditioner and Energy Technology n. Medical Technology o. Watch and Jewelry Industry p. Metal Industry q. Extrusion, Manufacturing and Conversion Industries r. Marinership s. Jewelry t. Space and Space Products

Landscapes

  • Engineering & Computer Science (AREA)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Editing Of Facsimile Originals (AREA)

Abstract

Present invention relates to a deep neural network system (1) that learns the encode/decode algorithm that works compatible with the file formats of vector and/or all laser marking machines placed on all objects and elements (A) (physical) that can be laser marked using the laser marking technique; provides solutions to the problems of traceability, recycling, sustainability, measurement, resistance to external conditions (dust, hot cold temperature differences), lighting, shadows, perspective, closure and viewing distance, thanks to the unique digital coded watermark (C) placed on the products, independent of the design and size; and provides robust encoding and decoding of arbitrary bridge multi-bit sequences by overlaying objects/elements (A) in away that approaches perceptual invisibility; and that learns an encoding/decoding algorithm that is a learned steganographic algorithm being resistant to distortions caused by real production and external conditions. Present invention also relates to a system (1) that provides real-time decoding of hidden hyperlinks to objects/elements (A) and connecting to remote resources (B).

Description

DESCRIPTION
A SYSTEM THAT PROVIDES SOLUTIONS BY ESTABLISHING INVISIBLE BRIDGES IN PHYSICAL OBJECTS/ELEMENTS USING LASER MARKING
TECHNIQUE
Technical Field
Present invention relates to a deep neural network system that learns the encode/decode algorithm that works compatible with the file formats of vector and/or all laser marking machines placed on all objects and elements (physical) that can be laser marked using the laser marking technique; provides solutions to the problems of traceability, recycling, sustainability, measurement, resistance to external conditions (dust, hot cold temperature differences), lighting, shadows, perspective, closure and viewing distance, thanks to the unique digital coded watermark placed on the products, independent of the design and size; and provides robust encoding and decoding of arbitrary bridge multi-bit sequences by overlaying objects/elements in a way that approaches perceptual invisibility; and that learns an encoding/decoding algorithm that is a learned steganographic algorithm being resistant to distortions caused by real production and external conditions.
Present invention also relates to a system that provides real-time decoding of hidden hyperlinks to objects/elements and connecting to remote resources.
State of the Art
Currently, by using solutions such as QR, Datamatrix and laser marking technique, it is possible to connect objects to remote sources with product traceability and assigning a unique identity to the applied product. When these applications are applied on the objects, variations in lighting, shadows, perspective, closure and viewing distance that may occur on the object can provide connections to remote sources, provided that they are resistant to certain conditions and conditions.
With currently use laser marking, a digital identity is assigned to the product only under certain conditions without being applied to a design with QR and Datamatrix options. Moreover, these applications completely lose their connection to remote sources in cases such as deterioration, wear, temperature change, dusting, which will occur from any external cause in the application area of the object. To exemplify; when a marking such as QR or Datamatrix applied with laser marking technique is made on a metal sheet and a deterioration occurs on the surface due to oxidation, all readers are no longer able to decode this coding in case of corruption or destruction at any point of this transferred QR or Datamatrix.
As another example; especially on small parcels in the industrial and automotive fields, marking is performed only in an area of 8 mm square, and sometimes the marking disappears due to the small dimensions of the product and sometimes the temperature differences undergone by the product.
In summary, resistance to external conditions (dust, hot cold temperature differences), lighting, shadows, perspective, closure and variations in viewing distance, visual distortions, light differences, temperature differences and most importantly, QR and Datamatrix options in the laser marking technique applied to the materials that are not flat, such as curved, oval, etc. have remained inoperable and unable to produce a solution thereof.
A system related to a data network and used for physical or digital objects to connect with associated remote resources is disclosed in the patent application document no. TR2020/08748 mentioned in the state of the art. The system disclosed in the patent application document numbered TR2020/08748 and the operating structures of the new invention according to this disclosure are quite similar. However, in the patent application document numbered TR2020/08748, work is carried out on the watermark, and in the new invention according to the description, the unique digital coded watermark is managed. In the invention according to this description, a unique digital coded watermarked file is produced and the file is distributed to predetermined parties. The unique digitally coded watermark is specially created for at least one object of any type and for every different type of template with laser marking technique, such as metals (aluminum, steel, zinc, lead, copper, etc.), platinum, diamond, gold, palladium, bronze, tin, tungsten, rhodium, brass, titanium, cobalt, glass, ceramic, rubber, plastic (polyethylene, etc.), stone (marble) , granite, travertine, basalt and andesite), wood, cement, aggregates, brick, clay, concrete, PVC ground plane, vehicle tire and so on. Unique digitally coded watermark is composed of many dots that are difficult to see with bare eyes. The unique digitally coded watermark is a special array matrix and is applied by laser marking technique. The unique digitally coded watermark contains a mathematical algorithm consisting of at least one and each printed dots with their own unique sequence, making it clear at which point of which template the design is made. For all these reasons, the system described in the patent application document numbered TR2020/08748 and the invention according to this description have differences.
Patent application document numbered W02019067730A1 belonging to Digimarc Corporation mentioned in the state of the art discloses the watermark detection methods and relevant embodiments. The watermark detection methods and embodiments in the relevant patent application document are carried out in different ways from the invention that is the subject of this specification. In the relevant application document, there is no mention of the unique digitally coded watermark specially created for at least one object of any type and for every different type of template with laser marking technique, such as metals (aluminum, steel, zinc, lead, copper, etc.), platinum, diamond, gold, palladium, bronze, tin, tungsten, rhodium, brass, titanium, cobalt, glass, ceramic, rubber, plastic (polyethylene, etc.), stone (marble) , granite, travertine, basalt and andesite), wood, cement, aggregates, brick, clay, concrete, PVC ground plane, vehicle tire and so on.
In conclusion, due to the above-mentioned negativities and the inadequacy of the existing solutions on the subject, it was deemed necessary to make an improvement in the relevant technical field.
Summary of the invention
The invention is inspired by current situations and aims to solve the above- mentioned disadvantages.
Object of the present invention is to provide a deep neural network system that learns the encode/decode algorithm that works compatible with the file formats of vector and/or all laser marking machines placed on all objects and elements (physical) that can be laser marked using the laser marking technique; provides solutions to the problems of traceability, recycling, sustainability, measurement, resistance to external conditions (dust, hot cold temperature differences), lighting, shadows, perspective, closure and viewing distance, thanks to the unique digital coded watermark placed on the products, independent of the design and size; and provides robust encoding and decoding of arbitrary bridge multi-bit sequences by overlaying objects/elements in a way that approaches perceptual invisibility; and that learns an encoding/decoding algorithm that is a learned steganographic algorithm being resistant to distortions caused by real production and external conditions.
Another object of the present invention is to set forth a system that provides real time decoding of hidden hyperlinks to objects/elements and connecting to remote resources.
The system according to the invention can detect all analog inputs formed on all surfaces and floors with a unique digital coded watermark compatible with vector and/or file formats of all laser machines, placed on all objects and elements (physical) that can be laser marked using laser marking technique, these inputs and the pattern on the floor where the located inputs are perceived, either based on the region(s) or independent of the region(s).
In the system according to the invention, the transmission of unique digitally coded watermark and input information or detection is conducted by means of a smartphone and/or a camera in the transmission terminal that only reads the patterns.
In the system according to the invention, the transmitted unique digital coded watermark information is interpreted and processed thereof. In conclusion, inputs produced in the system that can be managed, that can accomplish results and that need to be directed to other systems and software are transferred to the intended platforms in accordance with all requirements.
A data network was created in the system according to the invention in order to establish a connection between objects and elements, related resources and remote resources, and to assign a unique identity. With this data network, the system has a structure that is used to connect related and physical objects with relevant remote resources.
Technology used in the system according to the invention is not limited to these features. More generally, the system is used to connect any object to a corresponding network or local resource. System comprises four basic processes; recording, embedding, detection and responding.
The structural and characteristic features of the invention and all the advantages thereof will be understood more clearly with the figure given below and the detailed explanation with reference to this figure; therefore the evaluation should be made by taking this figure and detailed explanation into consideration.
Figure for A Better Understanding of the Invention
Figure 1 is the schematic illustration of the system according to the invention.
List of the Reference Numbers 1. System
2. Starting device
3. Router
4. Product processor
5. Server 6. Database
A. Object/element
B. Remote sources
C. Unique digitally coded watermark Detailed Description of the Invention In this detailed description, preferred embodiments of the system (1) according to the invention is disclosed only for a better understanding of the subject of the invention.
Present invention relates to a deep neural network system (1) that learns the encode/decode algorithm that works compatible with the file formats of vector and/or all laser marking machines placed on all objects and elements (A) (physical) that can be laser marked using the laser marking technique; provides solutions to the problems of traceability, recycling, sustainability, measurement, resistance to external conditions (dust, hot cold temperature differences), lighting, shadows, perspective, closure and viewing distance, thanks to the unique digital coded watermark (C) placed on the products, independent of the design and size; and provides robust encoding and decoding of arbitrary bridge multi-bit sequences by overlaying objects/elements (A) in a way that approaches perceptual invisibility; and that learns an encoding/decoding algorithm that is a learned steganographic algorithm being resistant to distortions caused by real production and external conditions.
Present invention also relates to a system (1 ) that provides real-time decoding of hidden hyperlinks to objects/elements (A) and connecting to remote resources (B).
System (1 ) comprises of many new features and applications. For example, system (1 ) is used for connecting a laser-marked product to an electronic content. Technology used in the system (1 ) is not limited to these features. More generally, the system (1 ) is used to connect any object (A) (physical) to a corresponding network or remote resource (B).
The system (1 ) according to the invention schematically illustrated in Figure 1 , includes the following;
❖ starting device (2) which detects data from an input object (A), processes this data and extracts the unique digitally coded watermark information (C) inside the object (A);
❖ the router (3), which receives the processed data from the starting device (2) and processes it thereof, records the information about the processed data in the database (6) and transmits at least a certain part of this processed data;
❖ product processor (4) that receives at least a certain part of the processed data from the router (3), which is distinguished and detected by the router (3), and by checking this data, compares the unique digital coded watermark (C) information obtained from the object (A) with the information belonging to the previously recorded object (A), and enables the establishment of a connection between object (A) and remote resources (B) as a result of this comparison; ❖ server (5), which saves the content of the identity of the object (A) and the product's identification information, in order to assign an identity to the object (A) and associate this identity with one or more responses;
❖ database (6) that stores the data recorded by the server (5) and enables this data to be used by the product processor (4).
System (1 ) comprises four basic processes; recording, embedding, detection and responding.
In the system (1 ), the unique digitally coded watermark (C) is first recorded in the database (6) by the server (5). The server (5) must save the identity information of the object (A) in relation to the unique digitally coded watermark (C) in the database (6) for the system (1 ) to identify the response (e.g., a URL) corresponding to an object (A) identifier related to a unique digitally coded watermark (C). The server (5) captures some basic identification information in the registration process of the unique digitally coded watermark (C). This information is then used to verify the incoming message, and the associated information/action is defined thereof. Identity information includes these examples:
❖ Object (A) identity code,
❖ Object (A) and associated attributes (unique ID, name, description, etc.)
❖ Action and
❖ Registered Unique serial number (for registration updates)
The object (A) defined by the server (5) to the database (6) and the associated attributes include the object (A) watermarked with their unique digital coded watermark (C). The object (A) features defined by the server (5) to the database (6) generally include the unique identification number of the object (A), its name, description, and a list of accounts authorized to access the record of the object (A).
Action defines the response intended by the user when the unique digitally coded watermark (C) is detected. Response may vary by product, but the explanatory edit includes returning some additional information about the object (A) with the unique digitally coded watermark (C). The action in the system (1 ) is to return a URL or HTML or object (A) identification code that will be used to display a web page associated with the object (A) with the unique digitally coded watermark (C).
Registered unique serial number is the last component of the registration. The registered unique serial number is the assigned product identifier and the product- specific identifier, allowing the system (1 ) to obtain the specific information/action for the object (A) in question.
Recording refers to the process of assigning an identity to an object (A) and associating that identity with a corresponding action or response. Recording process may include additional steps such as registering the registrant's name and/or organization, product name and product-specific identifier and so on.
The server (5) is configured to generate the file with a unique digitally coded watermark (C) and distribute the file to predetermined parties. The unique digitally coded watermark (C) is specially created for at least one object (A) of any type and for every different type of template with laser marking technique, such as metals (aluminum, steel, zinc, lead, copper, etc.), platinum, diamond, gold, palladium, bronze, tin, tungsten, rhodium, brass, titanium, cobalt, glass, ceramic, rubber, plastic (polyethylene, etc.), stone (marble) , granite, travertine, basalt and andesite), wood, cement, aggregates, brick, clay, concrete, PVC ground plane, vehicle tire and so on. Unique digitally coded watermark (C) is composed of many dots that are difficult to see with bare eyes. The unique digitally coded watermark (C) is a special array matrix and is applied by laser marking technique. The unique digitally coded watermark (C) contains a mathematical algorithm consisting of at least one and each printed dots with their own unique sequence, making it clear at which point of which template the design is made. The key code of the unique digital coded watermark (C) is primarily used to create the constellation cell, which will be processed into the picture with an error correction system based on the Reed- Solomon algorithm. The server (5) makes some geometric additions to the content of this cell in order to increase the readability of the code by the browser. The server (5) decides the footprint occupied by this cell in the picture, depending on the resolution provided by the print or digital media to be used at this stage. The server (5) blends the repetitive constellation pattern produced based on this decision with a high transparency on the original image. With this process, the server (5) optionally creates alterations in saturation or brightness that cannot be captured with bare eyes. In this way, the server (5) generates the file with the unique digital coded watermark (C) through the scanning application in the constellation pattern and distributes the file to the predetermined parties.
The server (5) is also configured to generate an encapsulation file and distribute the file to predetermined parties.
Embedding is performed after recording. Embedding refers to the encoding process of a digital identifier object (A) (for example, a unique digitally coded watermark (C) that carries a unique serial number in its payload). The unique digitally coded watermark (C) embedded in the object (A) is detected by the starting device (2).
Detection is the complementary process of embedding; that is, it enables a digital identifier to be distinguished from an object (A).
To associate multi-bit digital data with a physical object (A) in embedding and detection operations, special software is being developed that enables embedding in the object (A) with laser marking, QR, Datamatrix, barcodes, data glyphs, metadata, file header information, RF ID, UV/IR identifiers, organic transistors, mobile phones and other 2D image reading devices; that is, technologies such as decoding are used thereof.
Starting device (2) is used by the user. Starting device (2) can have a camera. Starting device (2) may be a wireless mobile device such as a mobile phone, tablet, and the like. Starting device (2) can have 2D image sensor. Starting device (2) may include a steganographic unique digitally coded watermark (C) detector configured to decipher steganographically encoded information in physical objects (A).
Starting device (2) analyzes the object (A) and extracts the unique digitally coded watermark (C) embedded in the object (A). Starting device (2) then transmits a certain portion or all of the data of this embedded unique digitally coded watermark (C) to the router (3).
Router (3) decodes the incoming data and analyzes the product identification information and product information in the data. Router (3) then compares this information with the data in the database (6). Router (3) then transfers this information to the relevant product processor (4). Responding is performed finally in the system (1 ). Responding refers to the action taken based on the distinguished identifier.
The product processor (4) receives the transmitted message from the router (3) and makes effort to match the unique serial number of the detected unique digitally coded watermark (C) with the unique serial number of a registered unique digitally coded watermark (C) previously stored in the database (6). If a match is found, the product processor (4) performs the intended action. For example, these actions include returning a URL for web redirection, serving an HTML page for initial user navigation, initiating software downloads, and so on. If no match is found, the product processor (4) sends an error code and message to the starting device (2). If a match is found, but the corresponding action is unavailable, incomplete, inactive, or invalid, the product processor (4) sends an error code and message to the starting device (2).
The product processor (4) processes the requisitions received from the starting device (2) via the internet and the router (3) and transmits the requested information/actions to the remote source (B). For example, the requested information is the URL and this URL is associated with the unique digitally coded watermark (C) payload transferred by the starting device (2). In other embodiments, further actions and/or information may be requested.
Product processor (4) is configurated to perform the following process steps:
❖ verify the received identifier (for example, unique digital coded watermark (C) unique serial number) against the list of active identifiers, if the unique serial number is not found, return an error message to the calling session and log the error in an error handling routine;
❖ detect the corresponding active primary action from the database (6) for each unique digital coded watermark (C) unique serial number received;
❖ perform an alternate "default" action for each valid unique digital coded watermark (C) unique serial number received, if it detects that the corresponding primary action is not currently active; ❖ return URL for application use in redirect (roundtrip approach) or serve HTML page found in call session if the received valid unique digitally coded watermark (C) detects an active primary action associated with the unique serial number;
❖ return URL for application usage during redirect (roundtrip approach) or serve found call session HTML page if it does not detect an active primary action associated with the unique serial number of the current unique digital coded watermark (C) received, but detects an associated default action thereof;
❖ return an error message to the call session and log in the error in the error handling routine if it cannot detect a valid, active primary or default action associated with the unique serial number of the unique digital coded watermark (C);
❖ respond to a valid registration request package upload request (appropriate format, an existing unique serial number, an account ID, and a valid corresponding account password) by returning a valid registration package for a unique digitally coded watermark (C) unique serial number provided;
❖ respond to an invalid registration request packet, note the error by providing an error message to the call session;
❖ respond to a local transaction cache clear request by writing locally cached transactions to the transaction log (database (6))
❖ respond to multiple URL requests by first returning the URL associated with the unique serial number provided, then any other active serial numbers and URLs for publication
Server (5) uses four different test models: Perspective Warp, Motion and Focal Blur, Color Manipulation, and Vibration and Shaking. 1 . Perspective Warp
Server (5) associates any two images of the same planar surface with a holography with the camera model assumption. Server (5) generates a random homography to simulate the effect of a camera not fully aligned with the encoded image marker. Server (5) keeps the four vertex positions of the marker within a fixed range (max ± 40 pixels, i.e., ±10%) to sample a homography, and then decodes the homography; mapping the original vertices to their new positions. Server (5) therefore re-samples the original image in two-cycle in order to create the perspective warped image.
2. Motion and Focal Blur
Blur can be caused by both camera movement and incorrect autofocus. Server (5) samples a random angle to simulate motion blur and generates a straight-line blur kernel with a width between 3 and 7 pixels. Server (5) uses a randomly sampled Gaussian blur kernel with a standard deviation of 1 to 3 pixels to simulate misfocusing.
3. Color Manipulation
Displays have a limited gamut compared to the full RGB color space. Cameras manipulate their output using exposure settings, white balance, and a color correction matrix. Server (5) estimates these disruptions with a sequence of random afin color reversal (stable in all the image) as follows.
❖ Tone wrapping: [-0.1 , 0.1].
❖ Desaturation: Random linear interpolation between the full RGB image and its grayscale equivalent
❖ Brightness and contrast: affine histogram rescaling mx + b & m ~ U[0,5, 1 ,5] and b ~ U[-0.3, 0.3]
4. Vibration and Shaking
Vibration and shake caused by camera systems are well studied and include photon jitter, dark image and shooting jitter. As a solution, server (5) assumes standard non-photon starved imaging conditions using a Gaussian noise model (by sampling the standard deviation). The server (5) uses image noise (s ~ U[0, 0.2]) to account.. Exemplary Embodiments for Invention:
Examples of how the system (1 ) can be adapted for various objects (A) are described hereinbelow. In each of these cases, copy detection technology, such as fragile unique digitally code watermarks (C), authentication hashes embedded in a unique digitally code watermark (C), or special authentication keys used to generate the unique digitally code watermark (C) can be used to authenticate the printed object (A). In addition, the unique digitally coded watermark (C) may carry information used to access and index information in a database (6) or computer network as described above.
In accordance with the exemplary embodiment of the present invention, a direct, undetectable code is embedded in the elements / objects (A). When a properly activated phone is recognized by the camera, this code automatically redirects an associated web browser to a destination selected by the print media manufacturer. This target, for example, a web page may provide additional information or services that are timelier and/or more comprehensive than those provided by printed material. With this embodiment, consumers are provided with the application example benefits, examples of which are given below, and advertisers with more effective tools to connect readers to e-commerce outlets.
System (1 ) according to the invention can be used in many applications. For example;
A. All objects and elements (A) (physical or electronic) with a unique digitally coded watermark (C) in vector and/or all laser machine file formats placed with laser marking can be linked to URL/Links.
(Youtube, Url, Social media accounts, campaign websites, E-commerce sites, Spotify, Steam etc.)
B. Augmented Reality (AR) representation can be made on all objects and elements (A) (physical) with a unique digital coded watermark (C) in vector and/or all laser machine file formats placed with laser marking.
Augmented reality screenings can be made live and/or anime (static or dynamic). C. All objects and elements (A) (physical) with unique digitally coded watermarks (C) in vector and/or all laser machine file formats placed with laser marking can be counted in warehouses, at salespoints on production lines, etc. (On Shelf and Warehouse)
D. Product traceability can be made on all objects and elements (A) (physical) with a unique digital coded watermark (C) in vector and/or all laser machine file formats placed with laser marking.
E. Product recycling information can be achieved through all objects and elements (A) (physical) with a unique digital coded watermark (C) in vector and/or all laser machine file formats placed with laser marking.
F. Production optimization can be achieved by associating sustainable information on all objects and elements (A) (physical) with a unique digital coded watermark (C) in vector and/or all laser machine file formats placed with laser marking.
G. All product-related information (based on product type or content such as model year, price, engine displacement, mileage, fuel consumption) can be accessed through all objects and elements (A) (physical) with a unique digital coded watermark (C) in vector and/or all laser machine file formats placed with laser marking.
H. Bill of Materials information associated with the product can be accessed through all objects and elements (A) (physical) with a unique digitally coded watermark (C) in vector and/or all laser machine file formats placed with laser marking. i. The authenticity of the product can be checked over all objects and elements (A) (physical) with a unique digital coded watermark (C) in vector and/or all laser machine file formats placed with laser marking.
J. T raceability, sustainability and originality follow-ups can be achieved in all sectors below through all objects and elements (A) (physical) with a unique digitally coded watermark (C) in vector and/or all laser machine file formats placed with laser marking. a. Automotive Industry and Parts b. Passenger and Commercial Vehicle Tires c. Display d. Computer Parts and Chips e. Smart Factories (Industry 4.0) f. Consumer Durables and White Appliances g. Aerospace h. Commercial Vehicles and Transportation i. Tool and Mold Manufacturing j. Construction Industry and Building Materials k. Electronic Products and Spare Parts I. Machine and System Manufacturing m. Air Conditioner and Energy Technology n. Medical Technology o. Watch and Jewelry Industry p. Metal Industry q. Extrusion, Manufacturing and Conversion Industries r. Marinership s. Jewelry t. Space and Space Products

Claims

1. A deep neural network system (1 ) that comprises the following components;
❖ starting device (2) which detects data from an input object (A), processes this data and extracts the unique digitally coded watermark information (C) inside the object (A);
❖ the router (3), which receives the processed data from the starting device (2) and reprocesses data thereof, records the information about the processed data in the database (6) and transmits at least a certain part of this processed data;
❖ product processor (4) that receives at least a certain part of the processed data from the router (3), which is distinguished and detected by the router (3), and by checking this data, compares the unique digital coded watermark (C) information obtained from the object (A) with the information belonging to the previously recorded object (A), and enables the establishment of a connection between object (A) and remote resources (B) as a result of this comparison;
❖ server (5), which saves the content of the identity of the object (A) and the product's identification information, in order to assign an identity to the object (A) and associate this identity with one or more responses;
❖ database (6) that stores the data recorded by the server (5) and enables this data to be used by the product processor (4), and that learns the encode/decode algorithm that works compatible with the file formats of vector and/or all laser marking machines placed on all objects and elements (A) (physical) that can be laser marked using the laser marking technique; provides solutions to the problems of traceability, recycling, sustainability, measurement, resistance to external conditions (dust, hot cold temperature differences), lighting, shadows, perspective, closure and viewing distance, thanks to the unique digital coded watermark (C) placed on the products, independent of the design and size; and provides robust encoding and decoding of arbitrary bridge multi-bit sequences by overlaying objects/elements (A) in a way that approaches perceptual invisibility; and that learns an encoding/decoding algorithm that is a learned steganographic algorithm being resistant to distortions caused by real production and external conditions, characterized by comprising;
❖ a product processor (4) receiving the transmitted message from the router (3), making effort to match the unique serial number of the detected unique digital coded watermark (C) with the unique serial number of a registered unique digitally coded watermark (C) previously stored in the database (6), performing the requested action (for example, returning a URL for a web redirect, serving an HTML page for initial user navigation, initiating software downloads, and so on) if a match is found, and sending an error code and message to the starting device (2) if a match is not found, and sending an error code and message to the starting device (2) if a match is found but the corresponding action is unavailable, incomplete, inactive or invalid, and
❖ a server (5) distribute the file to predetermined parties and that is configured to generate the file with a unique digitally coded watermark (C) o that is specially created for at least one object (A) of any type and for every different type of template with laser marking technique, such as metals (aluminum, steel, zinc, lead, copper, etc.), platinum, diamond, gold, palladium, bronze, tin, tungsten, rhodium, brass, titanium, cobalt, glass, ceramic, rubber, plastic (polyethylene, etc.), stone (marble) , granite, travertine, basalt and andesite), wood, cement, aggregates, brick, clay, concrete, PVC ground plane, vehicle tire and so on, o that is composed of many dots that are difficult to see with bare eyes, o that is a special array matrix and applied with laser marking technique, o that contains a mathematical algorithm consisting of at least one and each printed dots with their own unique sequence, making clear at which point of which template the design is made.
2. The system (1) according to the Claim 1 , characterized by comprising; a server (5) that creates the constellation cell that will be processed into the picture with an error correction system based on the Reed-Solomon algorithm with the key code server (5) uses, and makes certain geometric additions to the content of this cell in order to increase the readability of the code by the scanner, that decides the footprint of the cell in this picture at this stage, based on the resolution provided by the print or digital media to be used, that blends the repetitive constellation pattern produced based on this decision with a high transparency on the original image, that creates alterations in saturation or brightness that cannot be perceived with bare eyes, and thereby generating the unique digitally coded watermarked (C) file via the constellation-pattern browser application and distributing the file to predetermined parties.
3. The system (1) according to the Claim 1 , characterized by comprising; a server (5) configured to generate an encapsulation file and distribute the file to predetermined parties thereof.
4. The system (1) according to the Claim 1 , characterized by comprising; the server (5) which saves the identification information belonging to the object (A) in the database (6) as in relation to the unique digitally coded watermark (C), which captures some basic identification information in the registration process of the unique digitally coded watermark (C) for identifying the response (e.g., a URL) corresponding a unique digital coded watermark (C) to an object (A) identifier by the system (1), wherein the identification information is ❖ Object (A) identity code,
❖ Object (A) and associated attributes (unique ID, name, description, etc.)
❖ Action and
❖ Registered Unique serial number (for registration updates) and so on.
5. The system (1 ) according to the Claim 1 , characterized by comprising a starting device (2) with a camera.
6. The system (1 ) according to the Claim 1 , characterized by comprising the starting device (2), which is a wireless mobile device such as a mobile phone, tablet, and the like.
7. The system (1 ) according to the Claim 1 , characterized by comprising a starting device (2) with a 2D image sensor.
8. The system (1 ) according to the Claim 1 , characterized by comprising a starting device (2) which may include a steganographic unique digitally coded watermark (C) detector configured to decipher steganographically encoded information in physical objects (A).
9. The system (1 ) according to the Claim 1 , characterized by comprising a router (3) that decodes the received data, analyzes the product identification information and product information in the data, compares this information with the data in the database (6), and then sends this information to the relevant product processor (4).
10. The system (1 ) according to the Claim 1 , characterized by comprising a product processor (4), which processes the requests from the starting device (2) via the internet and the router (3) and transmits the requested information/actions to the remote resource (B).
11. The system (1 ) according to the Claim 1 , characterized by comprising product handler (4) configured to perform the following processing steps; ❖ verify the received identifier (for example, unique digital coded watermark (C) unique serial number) against the list of active identifiers, if the unique serial number is not found, return an error message to the calling session and log the error in an error handling routine;
❖ detect the corresponding active primary action from the database (6) for each unique digital coded watermark (C) unique serial number received;
❖ perform an alternate "default" action for each valid unique digital coded watermark (C) unique serial number received, if product handler (4) detects that the corresponding primary action is not currently active;
❖ return URL for application use in redirect (roundtrip approach) or serve HTML page found in call session if the received valid unique digitally coded watermark (C) detects an active primary action associated with the unique serial number;
❖ return URL for application usage during redirect (roundtrip approach) or serve found call session HTML page if product handler (4) does not detect an active primary action associated with the unique serial number of the current unique digital coded watermark (C) received, but detects an associated default action thereof;
❖ return an error message to the call session and log in the error in the error handling routine if product handler (4) cannot detect a valid, active primary or default action associated with the unique serial number of the unique digital coded watermark (C);
❖ respond to a valid registration request package upload request (appropriate format, an existing unique serial number, an account ID, and a valid corresponding account password) by returning a valid registration package for a unique digitally coded watermark (C) unique serial number provided; ❖ respond to an invalid registration request packet, note the error by providing an error message to the call session;
❖ respond to a local transaction cache clear request by writing locally cached transactions to the transaction log (database (6))
❖ respond to multiple URL requests by first returning the URL associated with the unique serial number provided, then any other active serial numbers and URLs for publication.
12. The system (1) according to the Claim 1 , characterized by comprising a server (5) using four different test models; Perspective Warp, Motion and Focal Blur, Color Manipulation, and Vibration and Shaking.
13. The system (1) according to the Claim 12, characterized by comprising a server (5) that generates a random homography to simulate the effect of a camera not perfectly aligned with the encoded image marker, that solves the homography that keeps the four vertices positions of the marker within a fixed range (max ± 40 pixels, i.e. ±10%) to sample a homography and then maps the original vertices to their new positions, and that re-samples the original image in two-cycle in order to create the perspectively warped image that associates any two images of the same planar surface with a homography with the camera model assumption in the Perspective Warp test model.
14. The system (1) according to the Claim 12, characterized by comprising a server (5) that samples a random angle to simulate motion blur caused by both camera movement and faulty autofocus in the Motion and Focal Blur test model, and that generates a straight line blur kernel with a width of between 3 and 7 pixels and uses a randomly sampled Gaussian blur kernel with a standard deviation of 1 to 3 pixels to simulate misfocusing.
15. The system (1) according to the Claim 12, characterized by comprising a server (5) that estimates the distortions in the Color Manipulation test model on displays with a limited gamut compared to the full RGB color space, cameras whose output changes exposure settings using white balance and a color correction matrix, ❖ Tone wrapping: [-0.1 , 0.1].
❖ Desaturation: Random linear interpolation between the full RGB image and its grayscale equivalent
❖ Brightness and contrast: affine histogram rescaling mx + b & m ~ U[0,5, 1 ,5] and b ~ U[-0.3, 0.3] with a series of random affine color reversals (constant across the entire image).
16. The system (1) according to Claim 12, characterized by comprising a server (5) that assumes standard non-photon starved imaging conditions, using a Gaussian noise model (by sampling the standard deviation), which is well- studied and includes photon jitter, dark image and shooting jitter, and as a solution to the vibration and jolt induced by camera systems in the vibration and shaking test model and which uses image noise to account for (s ~ U[0, 0.2]).
PCT/IB2021/058789 2021-06-17 2021-09-27 A system that provides solutions by establishing invisible bridges in physical objects/elements using laser marking technique WO2022263920A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
TR2021009908 2021-06-17
TR2021/009908 TR2021009908A2 (en) 2021-06-17 A System That Provides Solutions by Creating Invisible Bridges on Physical Objects/Elements Using Laser Marking Technique

Publications (1)

Publication Number Publication Date
WO2022263920A1 true WO2022263920A1 (en) 2022-12-22

Family

ID=84527200

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2021/058789 WO2022263920A1 (en) 2021-06-17 2021-09-27 A system that provides solutions by establishing invisible bridges in physical objects/elements using laser marking technique

Country Status (1)

Country Link
WO (1) WO2022263920A1 (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1758045A1 (en) * 2004-04-26 2007-02-28 Technology Transfer Service Corp. Method for forming 2-dimensional code by laser marking and laser marking device
EP1923830A2 (en) * 1999-05-19 2008-05-21 Digimarc Corporation Methods and systems for controlling computers or linking to internet resources from physical and electronic objects
US20100142756A1 (en) * 2008-12-10 2010-06-10 Canon Kabushiki Kaisha Document security method
KR101535534B1 (en) * 2014-08-29 2015-07-13 대한민국 A Creating and Verifying Method Of A Document Having Printed Means Of Preventing From Forging/Manipulating
US20150228143A1 (en) * 2012-09-21 2015-08-13 Orell Füssli Sicherheitsdruck Ag Security document with microperforations

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1923830A2 (en) * 1999-05-19 2008-05-21 Digimarc Corporation Methods and systems for controlling computers or linking to internet resources from physical and electronic objects
EP1758045A1 (en) * 2004-04-26 2007-02-28 Technology Transfer Service Corp. Method for forming 2-dimensional code by laser marking and laser marking device
US20100142756A1 (en) * 2008-12-10 2010-06-10 Canon Kabushiki Kaisha Document security method
US20150228143A1 (en) * 2012-09-21 2015-08-13 Orell Füssli Sicherheitsdruck Ag Security document with microperforations
KR101535534B1 (en) * 2014-08-29 2015-07-13 대한민국 A Creating and Verifying Method Of A Document Having Printed Means Of Preventing From Forging/Manipulating

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
ANATOLY BELIKOV ET AL: "GoodPoint: unsupervised learning of keypoint detection and description", ARXIV.ORG, CORNELL UNIVERSITY LIBRARY, 201 OLIN LIBRARY CORNELL UNIVERSITY ITHACA, NY 14853, 1 June 2020 (2020-06-01), XP081689851 *

Similar Documents

Publication Publication Date Title
Tancik et al. Stegastamp: Invisible hyperlinks in physical photographs
US8379908B2 (en) Embedding and reading codes on objects
US7044395B1 (en) Embedding and reading imperceptible codes on objects
US20200073535A1 (en) Device, Method And Server For Retrieving Information Using A Machine Readable image
US8594364B2 (en) Batch identifier registration and embedding in media signals
US10506128B1 (en) Encoded signal systems and methods to ensure minimal robustness
CN102663600B (en) Anti-counterfeiting system based on digital watermarks and barcodes, anti-counterfeiting method for anti-counterfeiting system and application of anti-counterfeiting system
US8750556B2 (en) Watermark systems and methods
US20160117061A1 (en) System and method for image based interactions
CN103942585B (en) Method for converting information into two-dimensional codes
US20070055884A1 (en) User control and activation of watermark enabled objects
WO2006136958A9 (en) System and method of improving the legibility and applicability of document pictures using form based image enhancement
WO2002033954A2 (en) User control and activation of watermark enabled objects
JP2005174317A5 (en)
US20200228675A1 (en) Encoded signal systems and methods to ensure minimal robustness
WO2003001435A1 (en) Image based object identification
US9177199B2 (en) Semantic magazine pages
CN106233303A (en) Machine readable watermark in image and bar code
CN104732182A (en) Two-dimension-code-based communication method on webpage
WO2022263920A1 (en) A system that provides solutions by establishing invisible bridges in physical objects/elements using laser marking technique
US11120520B2 (en) Method for marking visuals of information for subsequent identification or authentication
CN105760912A (en) Encoding/decoding of information in graphic information units
WO2001041056A1 (en) Embedding and reading imperceptible codes on objects
JP2009159474A (en) Authentication system, authentication device, authentication program and authentication method
Yamada et al. A method for detecting watermarks in print using smart phone: finding no mark

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21835367

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21835367

Country of ref document: EP

Kind code of ref document: A1