Mobile Camera Information Extraction from Printed Matter

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for extracting additional information from printed matter using a mobile terminal's built-in camera face challenges due to camera shake, distortion, and unstable image capturing conditions, leading to poor image quality and difficulty in accurately extracting embedded information, especially when the amount of information increases or the terminal is inclined.

Innovation Solution

A mobile information processing apparatus and method that employs an image forming apparatus to embed additional information in printed matter using error diffusion and quantization techniques, allowing for extraction via a mobile terminal's camera by adjusting image capturing conditions and using multiple captured images to compensate for incomplete data from unit blocks.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If additional information is embedded in printed matter using information hiding technique, then the information can be extracted by capturing with a camera, but the image quality deteriorates due to camera shake, distortion, and unstable capturing conditions when using a mobile terminal's built-in camera

Engineering Contradiction:
Improveadditional information extraction accuracyVSAvoidimage capturing stability
Core Design Contradiction:
Loss of informationVSReliability

Solution Approach 1:

The patent applies preliminary action by pre-embedding multiple types of markers (reference markers and correction markers) into the printed matter before capturing. These markers are positioned and designed in advance to provide reference information for later image correction and additional information extraction, enabling the system to compensate for unstable capturing conditions.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent uses feedback by analyzing the captured image to detect markers, calculating geometric transformation parameters based on marker positions, and applying corrections to the captured image. The system continuously adjusts the extraction process based on the actual image quality and marker detection results, improving extraction accuracy despite unstable capturing conditions.

Inventive Principle:
Principle #23Feedback

2Quantity of substance

If the amount of additional information increases, then more data can be embedded in the printed matter, but the basic unit area becomes larger making accurate extraction more difficult

Engineering Contradiction:
Improveamount of additional informationVSAvoidextraction accuracy
Core Design Contradiction:
Quantity of substanceVSMeasurement precision

Solution Approach 1:

The patent applies segmentation by dividing the additional information into multiple units and embedding them in separate unit blocks across the printed matter. Each unit block contains a portion of the additional information along with embedded markers. This segmentation allows the system to extract information from multiple distributed locations, maintaining precision even as the total amount of information increases.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from extracting additional information from a single location to extracting from multiple spatial locations by dividing the information across multiple unit blocks. This dimensional approach distributes the extraction task across different areas of the printed matter, enabling accurate extraction of larger amounts of information through spatial distribution.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Area of stationary object

If image capturing is performed at a longer distance from the printed matter, then the entire area can be captured, but the image quality decreases making additional information extraction difficult

Engineering Contradiction:
Improvecaptured areaVSAvoidimage quality
Core Design Contradiction:
Area of stationary objectVSMeasurement precision

Solution Approach 1:

The patent uses copying by creating a geometric transformation model that maps the captured image coordinates to the original printed matter coordinates. By detecting marker positions in the captured image and calculating transformation parameters, the system creates a virtual copy of the original layout, enabling accurate extraction of additional information even when captured from a distance with potential distortion.

Inventive Principle:
Principle #26Copying

Data Source

PatentEP3422690B1Information processing apparatus, control method thereof, and program
Publication Date: 2022.05.18 CANON KK
  • EP3422690B1 patent drawingFigure 1A~1B
  • EP3422690B1 patent drawingFigure 2A~2B
  • EP3422690B1 patent drawingFigure 3

AI summary

To make it possible to easily extract additional information from printed matter by using a built-in camera of a mobile terminal. An information processing apparatus (110) that reads additional information by capturing printed matter in which the additional information is embedded by using a built-in camera, including: a control unit (111, 210) configured to acquire a first captured image obtained by capturing the printed matter under a first image capturing condition and a second captured image obtained by capturing the printed matter under a second image capturing condition different from the first image capturing condition; and an extraction unit (112) configured to extract the additional information by analyzing the first captured image and the second captured image.