# US 2025/0085691 A1 · System and method for ai-generated patterns integrating multi-food material extrusion, 3d printing, and laser applications

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0085691 A1
- Document ID: US20250085691A1
- Status: Published application
- Kind: A1
- Application number: 18/818,665
- Filed: 2024-08-29
- Published: 2025-03-13
- Earliest claimed: 2023-09-12
- Assignees: The Hong Kong University of Science and Technology
- Applicants: The Hong Kong University of Science and Technology
- Inventors: Guijun LI; Kong Wai LEE; Yang XU; Yee Him Timothy CHAN; Miao TANG; Huan LIU; Yexin PAN; Qiaoyaxiao YUAN
- Web page: https://patentgrep.com/patent/US20250085691A1

## CPC

- G05B 19/4099: Surface or curve machining, making three-dimensional [3D] objects, e.g. desktop manufacturing
- B33Y 80/00: Products made by additive manufacturing
- A21C 9/066: {using two separate sheets or pieces of dough with filling in between, e.g. crimped at their edges; Ravioli-making machines (A21C11/163 takes precedence)}
- G05B 2219/32128: Gui graphical user interface
- G05B 2219/33002: Artificial intelligence AI, expert, knowledge, rule based system KBS
- G05B 2219/35134: 3-D cad-cam
- G05B 2219/49023: 3-D printing, layer of powder, add drops of binder in layer, new powder
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)
- B33Y: ADDITIVE MANUFACTURING, i.e. MANUFACTURING OF THREE-DIMENSIONAL [3D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVE AGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3D PRINTING, STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING
- A21C: MACHINES OR EQUIPMENT FOR MAKING OR PROCESSING DOUGHS; HANDLING BAKED ARTICLES MADE FROM DOUGH

## Abstract

A method using AI models to generate patterns or images on a pastry is provided. The method at least includes steps as follows: providing a UI module for user input of keywords and descriptions; generating black and white vector-style images based on this input using an AI-based image module; processing the 2D image into digital instructions via an image-transformer module; sending these instructions to a laser patterning module; and using the laser patterning module to apply the pattern to the cookie's surface.

## Claims

### Claim 1

A system for utilizing AI models to generate pattern or images on a surface of a mooncake, comprising:
a user interface (UI) module serving as an entry component for the system and configured to facilitate user interaction and input for image generation and to allow a user to enter keywords and descriptions for define an image for a mooncake;
an AI-based image module connected to the UI module and configured to generates at least one black and white vector-style image based on user's input fed by the UI module;
an image-transformer module connected with the UI module and the AI-based image module and configured to process a two-dimension pattern or image generated by the AI-based image module, so as to enable three-dimension printing extrusion, determining a desired volume with millimetre resolution to accurately translate patterns into a three-dimension structure for the mooncake;
a 3D printer connected to the image-transformer module and receiving the digital instructions from the image-transformer module for 3D printing, wherein the 3D printer is configured to form an outer layer, an inner filling, and a top layer for the mooncake using different food ingredients; and
a laser patterning module connected to the image-transformer module and receiving the digital instructions from the image-transformer module for laser patterning, wherein the laser patterning module is configured to emit a laser beam to a surface of the top layer of the mooncake to create an image or a pattern on the surface of the top layer of the mooncake, involving selectively heating and browning the surface according to the two-dimension pattern or image generated by the AI-based image module which resulting from the user's input.

### Claim 2 (depends on 1)

The system according to claim 1, wherein the UI module is further configured to provide a dedicated input screen for the user, offering an interface for entering positive and negative keywords.

### Claim 3 (depends on 2)

The system according to claim 2, wherein the AI-based image module is implemented in a Python-based engine, and wherein the dedicated input screen of the UI module is configured to show an AI-generated image produced by the AI-based image module for the user, allowing for interactive adjustments and real-time feedback to fine-tune images.

### Claim 4 (depends on 1)

The system according to claim 1, wherein the image-transformer module is further configured to convert the two-dimension pattern or image generated by the AI-based image module into G-code instructions for the 3D printer for defining specific movements and actions of the 3D printer, including an extrusion path and layering of printed materials.

### Claim 5 (depends on 1)

The system according to claim 1, wherein the 3D printer is equipped with a multi-nozzle material extrusion system and a multi-nozzle head system, allowing the 3D printer to extrude up to four materials from extrusion syringes.

### Claim 6 (depends on 5)

The system according to claim 5, wherein the 3D printer is equipped with four material storage slots, each configured to store a different material or ingredient, the 3D printer comprises dedicated pipelines for transporting the material or ingredient from material storage slots to the multi-nozzle head system, and wherein the multi-nozzle head system is connected to four nozzle heads, each capable of extruding a different material, and works in conjunction with a rotating platform to apply the materials or ingredients to the mooncake.

### Claim 7 (depends on 1)

The system according to claim 1, wherein the image-transformer module further comprises a database configured to store at least one dimensions of the mooncake's top layer, outer layer, and inner filling, and wherein the database works in conjunction with the digital instructions from the image-transformer module to facilitate the image generation of various parameters needed for mooncake production.

### Claim 8

A system for utilizing AI models to generate pattern or images on a surface of a cookie, comprising:
a user interface (UI) module serving as an entry component for the system and configured to facilitate user interaction and input for image generation and to allow a user to enter keywords and descriptions for define an image for a cookie;
an AI-based image module connected to the UI module and configured to generates at least one black and white vector-style image based on user's input fed by the UI module;
an image-transformer module connected with the UI module and the AI-based image module and configured to process a two-dimension pattern or image generated by the AI-based image module, so as to format the two-dimension pattern or image into a scalable vector graphics (SVG) file; and
a laser patterning module connected to the image-transformer module and receiving the digital instructions from the image-transformer module for laser patterning, wherein the laser patterning module is configured to follow a path defined in the SVG file fed by the image-transformer module and engrave a pattern or image design onto a surface of the cookie, so as to cause localized caramelization of the cookie's surface, adding depth and contrast to the cookie's surface.

### Claim 9 (depends on 8)

The system according to claim 8, wherein the laser patterning module further comprises a controller for storing parameter setting, and wherein the parameter setting comprises number of passes, speed, power, frequency, pulse duration, laser on delay, laser off delay, laser end delay, and laser polygon delay.

### Claim 10 (depends on 9)

The system according to claim 9, wherein the controller is configured to apply the parameter settings to the laser patterning module, such that the laser patterning module performs the laser patterning with a wavelength of 1066 nm, a power output set at 10%, a speed of 1000 mm/s, a frequency of 30 kHz, and a pulse duration of 10 nanoseconds.

### Claim 11 (depends on 10)

The system according to claim 10, wherein the laser patterning module further includes a database that records the dimensions of the cookies to be processed, ensuring that during the laser patterning process, the laser beam does not exceed the size boundaries of the cookies.

### Claim 12 (depends on 8)

The system according to claim 8, wherein the AI-based image module further comprises AI text-to-image or image-to-image models using Python-based model for analyzing the keywords and descriptions fed by the UI module.

### Claim 13

A method for utilizing AI models to generate pattern or images on a surface of a pastry, comprising:
providing a user interface (UI) module for serving as an entry component for a user;
allowing, by the UI module, the user to enter keywords and descriptions for define an image for a pastry;
facilitating, by the UI module, user interaction and input for image generation;
generating, by an AI-based image module, at least one black and white vector-style image based on user's input fed by the UI module;
processing, by an image-transformer module, a two-dimension pattern or image generated by the AI-based image module, for generating digital instructions at least for a laser patterning module;
receiving, by a laser patterning module, the digital instructions from the image-transformer module for laser patterning; and
emitting, by the laser patterning module, a laser beam to a surface of the pastry to create an image or a pattern on the surface of the pastry according to the two-dimension pattern or image generated by the AI-based image module which resulting from the user's input.

### Claim 14 (depends on 13)

The method according to claim 13, wherein the pastry is a mooncake, and the method further comprises:
processing the two-dimension pattern or image to enable three-dimension printing extrusion;
determining a desired volume with millimetre resolution to accurately translate patterns into a three-dimension structure for the mooncake;
forming, by a 3D printer, an outer layer, an inner filling, and a top layer for the mooncake using different food ingredients; and
controlling the laser beam for selectively heating and browning the surface of the top layer of the mooncake.

### Claim 15 (depends on 14)

The system according to claim 14, further comprising:
showing, by a dedicated input screen, an AI-generated image produced by the AI-based image module for the user, allowing for interactive adjustments and real-time feedback to fine-tune images.

### Claim 16 (depends on 14)

The system according to claim 14, further comprising:
extruding, by the 3D printer, up to four materials from extrusion syringes, wherein the 3D printer is equipped with four material storage slots, each configured to store a different material or ingredient, the 3D printer comprises dedicated pipelines for transporting the material or ingredient from the material storage slots to the multi-nozzle head system, and wherein the multi-nozzle head system is connected to four nozzle heads, each capable of extruding a different material, and works in conjunction with a rotating platform to apply the materials or ingredients to the mooncake.

### Claim 17 (depends on 13)

The method according to claim 13, wherein the pastry is a cookie, and the method further comprises:
formatting the two-dimension pattern or image into a scalable vector graphics (SVG) file;
controlling the laser beam by having the laser beam follow a path defined in the SVG file; and
engraving, by the laser patterning module, a pattern or image design onto a surface of the cookie, so as to cause localized caramelization of the cookie's surface, adding depth and contrast to the cookies' surface.

### Claim 18 (depends on 17)

The method according to claim 17, further comprising:
storing, by a controller, parameter setting for the laser patterning module, wherein the parameter setting comprises number of passes, speed, power, frequency, pulse duration, laser on delay, laser off delay, laser end delay, and laser polygon delay.

### Claim 19 (depends on 18)

The method according to claim 18, further comprising:
applying, by the controller, the parameter settings to the laser patterning module, such that the laser patterning module performs the laser patterning with a wavelength of 1066 nm, a power output set at 10%, a speed of 1000 mm/s, a frequency of 30 kHz, and a pulse duration of 10 nanoseconds.
