# US 2026/0288859 A1 · System

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

- Number: US 2026/0288859 A1
- Document ID: US20260288859A1
- Status: Published application
- Kind: A1
- Application number: 19/568,910
- Filed: 2026-03-17
- Published: 2026-09-24
- Earliest claimed: 2025-03-19
- Assignees: SoftBank Group Corp.
- Applicants: SoftBank Group Corp.
- Inventors: Hiroshi SHIBASAKI
- Web page: https://patentgrep.com/patent/US20260288859A1

## CPC

- G06F 16/535: Filtering based on additional data, e.g. user or group profiles
- G06F 16/24578: {using ranking}
- G06F 16/56: having vectorial format
- G06F 40/40: Processing or translation of natural language (natural language analysis G06F40/20; semantic analysis G06F40/30)
- G06T 7/0002: {Inspection of images, e.g. flaw detection}
- G06T 11/60: Creating or editing images; Combining images with text
- G06V 10/25: Determination of region of interest [ROI] or a volume of interest [VOI]
- G06V 10/764: using classification, e.g. of video objects
- G06V 10/82: using neural networks
- G06T 2207/30168: Image quality inspection
- G06V 40/174: {Facial expression recognition}
- G10L 25/63: for estimating an emotional state
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

A system includes a processor that is configured to receive an image of a stained portion, an image of an entire object including the stain, information indicating a material of the object, information indicating circumstances under which the stain occurred, and information indicating elapsed time since the stain occurred, analyze the image of the stained portion, the image of the entire object, the material information, the stain circumstance information, and the elapsed time information by using a generative artificial intelligence model, generate, based on results of the analyzing, a prompt for instructing the generative artificial intelligence model to calculate an optimal cleaning method, search, based on the generated prompt, a database for the optimal cleaning method and select the optimal cleaning method from the database, output the selected optimal cleaning method visually by using at least one of text, diagrams, images, and videos.

## Claims

### Claim 1

A system comprising:
circuitry configured to:
receive, via a communication interface coupled to a packet-switched network, first image data including a region of interest, second image data including an entire object incorporating the region of interest, material information, occurrence information, and elapsed-time information from a terminal device, and convert the received data into structured input data;
execute image analysis processing on the structured input data to estimate a region type, region characteristics, and affected area parameters, and combine the region characteristics with the material information and the elapsed-time information to generate a feature quantity;
generate, based on the feature quantity, a search condition for retrieving method information defining processing conditions and a processing procedure, perform an inquiry process to a storage device using the search condition to obtain a plurality of candidate method information items, and evaluate the plurality of candidate method information items based on at least material compatibility, processing effectiveness, and operability to select an optimal candidate;
generate a prompt sentence for instructing a generative AI model to generate processing guidance content based on the feature quantity and the selected optimal candidate, and transmit the prompt sentence to the generative AI model to obtain guidance content data; and
receive user state data from the terminal device, apply an emotion recognition model to the user state data to generate an emotional state representation, adjust a presentation format of the guidance content data based on the emotional state representation, and transmit the adjusted guidance content data to the terminal device via the communication interface.

### Claim 2 (depends on 1)

The system according to claim 1, wherein the circuitry is configured to
apply a convolutional neural network to the first image data and the second image data to extract visual feature vectors representing the region characteristics and the material properties, and use the visual feature vectors as inputs to the feature quantity generation.

### Claim 3 (depends on 2)

The system according to claim 2, wherein the circuitry is configured to
apply a region segmentation model to the first image data to identify a boundary of the region of interest, estimate an area proportion of the region of interest relative to the entire object, and incorporate the area proportion as a parameter in the feature quantity.

### Claim 4 (depends on 3)

The system according to claim 3, wherein the circuitry is configured to
apply a material classification model to the second image data to identify material type features independent of the material information received from the terminal device, and use the identified material type features to validate the material information.

### Claim 5 (depends on 4)

The system according to claim 4, wherein the circuitry is configured to
apply a time-decay function to the elapsed-time information to generate a time-weighted degradation parameter, and incorporate the time-weighted degradation parameter in the feature quantity to adjust candidate method information evaluation scores.

### Claim 6 (depends on 1)

The system according to claim 1, wherein the circuitry is configured to
generate the guidance content data comprising at least one of text instructions, diagram data, image data, and video data using the generative AI model, and transmit the guidance content data to the terminal device for visual presentation.

### Claim 7 (depends on 6)

The system according to claim 6, wherein the circuitry is configured to
apply a multimodal generation model to generate synchronized text and visual guidance content data, and assemble the text and visual components into a structured presentation format for transmission to the terminal device.

### Claim 8 (depends on 1)

The system according to claim 1, wherein the circuitry is configured to
evaluate the plurality of candidate method information items using a multi-criteria scoring function that assigns weighted scores to material compatibility, processing effectiveness, and operability parameters, and select the candidate with the highest aggregate score as the optimal candidate.

### Claim 9 (depends on 8)

The system according to claim 8, wherein the circuitry is configured to
apply a ranking model to re-rank the plurality of candidate method information items based on user preference information stored in the storage device, and present re-ranked candidates to the user via the terminal device prior to final selection.

### Claim 10 (depends on 1)

The system according to claim 1, wherein the circuitry is configured to
apply facial expression recognition processing to image data included in the user state data to classify the emotional state into predefined categories, and apply voice analysis processing to audio data included in the user state data to generate prosodic features, and combine the facial expression classification and the prosodic features in the emotion recognition model.

### Claim 11 (depends on 10)

The system according to claim 10, wherein the circuitry is configured to
adjust the presentation format of the guidance content data by selecting a simplified instruction format when a high-stress emotional state is detected and a detailed technical format when a calm emotional state is detected.

### Claim 12 (depends on 1)

The system according to claim 1, wherein the circuitry is configured to
receive the first image data and the second image data from an image acquisition device coupled to the terminal device, apply an image quality assessment model to verify that the first image data and the second image data satisfy minimum resolution and focus quality requirements, and request re-capture when the quality requirements are not met.

### Claim 13 (depends on 12)

The system according to claim 12, wherein the circuitry is configured to
apply an image preprocessing pipeline to the first image data and the second image data comprising color normalization, contrast enhancement, and noise reduction prior to executing the image analysis processing.

### Claim 14 (depends on 1)

The system according to claim 1, wherein the circuitry is configured to
store the structured input data, the feature quantity, and the selected optimal candidate in the storage device indexed by region type and material information, and use stored historical data to improve candidate evaluation scoring over time.

### Claim 15 (depends on 14)

The system according to claim 14, wherein the circuitry is configured to
receive outcome feedback information from the terminal device indicating a result of applying the guidance content data, and update the candidate evaluation scoring parameters based on the outcome feedback information.

### Claim 16 (depends on 1)

The system according to claim 1, wherein the circuitry is configured to
apply a natural language processing model to the occurrence information received from the terminal device to extract circumstance parameters and temporal context, and incorporate the circumstance parameters and temporal context as additional fields in the structured input data.

### Claim 17 (depends on 16)

The system according to claim 16, wherein the circuitry is configured to
apply a context classification model to the circumstance parameters to assign an occurrence category, and use the occurrence category as an additional search parameter in the search condition for retrieving candidate method information items.

### Claim 18

A system comprising:
circuitry configured to:
receive, via a communication interface coupled to a packet-switched network, first image data including a region of interest, second image data including an entire object, material information, occurrence information, and elapsed-time information from a terminal device;
apply a convolutional neural network to the first image data and the second image data to extract visual feature vectors, apply a region segmentation model to identify a boundary of the region of interest, apply a time-decay function to the elapsed-time information to generate a time-weighted degradation parameter, and combine the visual feature vectors, material information, and time-weighted degradation parameter to generate a feature quantity;
generate search conditions from the feature quantity, retrieve a plurality of candidate method information items from a storage device, apply a multi-criteria scoring function to evaluate the candidates, and select an optimal candidate based on evaluation scores;
generate a prompt sentence incorporating the feature quantity and the selected optimal candidate, input the prompt sentence to a generative AI model to obtain guidance content data comprising text, diagram, and image components; and
apply an emotion recognition model to user state data received from the terminal device to generate an emotional state representation, adjust a presentation format of the guidance content data based on the emotional state representation, and transmit the adjusted guidance content data to the terminal device.

### Claim 19 (depends on 18)

The system according to claim 18, wherein the circuitry is configured to
receive outcome feedback information from the terminal device indicating a result of applying the guidance content data, update candidate evaluation scoring parameters in the storage device based on the outcome feedback information, and apply updated scoring parameters in subsequent candidate method information evaluation.

### Claim 20

A method comprising:
receiving, via a communication interface coupled to a packet-switched network, first image data including a region of interest, second image data including an entire object incorporating the region of interest, material information, occurrence information, and elapsed-time information from a terminal device, and converting the received data into structured input data;
executing image analysis processing on the structured input data to estimate a region type, region characteristics, and affected area parameters, and combining the region characteristics with the material information and the elapsed-time information to generate a feature quantity;
generating, based on the feature quantity, a search condition for retrieving method information defining processing conditions and a processing procedure, performing an inquiry process to a storage device using the search condition to obtain a plurality of candidate method information items, and evaluating the plurality of candidate method information items to select an optimal candidate;
generating a prompt sentence for instructing a generative AI model to generate processing guidance content based on the feature quantity and the selected optimal candidate, and transmitting the prompt sentence to the generative AI model to obtain guidance content data; and
receiving user state data from the terminal device, applying an emotion recognition model to the user state data to generate an emotional state representation, adjusting a presentation format of the guidance content data based on the emotional state representation, and transmitting the adjusted guidance content data to the terminal device via the communication interface.
