Scanned Document AI Prompt Registration for One-Touch Image Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Users need to repeatedly input instruction sentences for generative AI services when processing images from scanned documents, which is cumbersome and inefficient.

Innovation Solution

A system is implemented where a one-touch button registers an instruction sentence template for generative AI, allowing automatic processing of scanned documents without repeated user input, utilizing a networked computer, image processing apparatus, and generative AI server to streamline the workflow.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If users input instruction sentences manually each time for AI service processing, then the AI service can be executed with user-specific instructions, but the operation becomes troublesome and time-consuming

Engineering Contradiction:
Improveease of operationVSAvoidtime loss
Core Design Contradiction:
Ease of operationVSLoss of time

Solution Approach 1:

The system performs preliminary action by automatically generating and transmitting instruction sentences to the AI service without requiring user input each time. The image processing apparatus pre-configures the instruction sentence based on the image data being processed, so that when processing is initiated, the instruction is already prepared and ready to be sent, eliminating the need for repeated manual input.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements self-service by enabling the image processing apparatus to automatically create and manage instruction sentences for AI service processing. The apparatus itself generates the appropriate instruction text based on the processing task, transmits it to the AI service, and manages the entire workflow without requiring external user intervention for each processing operation.

Inventive Principle:
Principle #25Self-service

2Productivity

If the same instruction sentence is used for multiple images, then processing efficiency improves, but the system lacks adaptability to different processing needs

Engineering Contradiction:
ImproveproductivityVSAvoidadaptability
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The system applies dynamics by making the instruction sentence configurable and adaptable rather than fixed. Users can define different instruction sentence templates for different processing scenarios, and the system dynamically selects and transmits the appropriate instruction based on the specific image processing task. This allows both efficient batch processing with consistent instructions and flexible adaptation when different processing requirements arise.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP4672030A1Information processing apparatus, program, and information processing method
Publication Date: 2025.12.31 CANON KK
  • EP4672030A1 patent drawingFigure 1
  • EP4672030A1 patent drawingFigure 2A~2C
  • EP4672030A1 patent drawingFigure 3A~3B

AI summary

An image processing apparatus includes a receiving unit configured to receive selection of a button from a user, an acquisition unit configured to acquire image data by reading an original document according to reception of the selection by the receiving unit, and a transmission unit configured to transmit an instruction sentence registered in association with the button by the user and to be used for processing that is performed using the image data by an artificial intelligence (AI) service and the image data to the AI service.