Image Processing Apparatus for Photo Book Embedding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques struggle to accurately read additional information embedded in images, particularly when the capturing angle is not perpendicular, leading to unreadable images and potential damage to photo books during correction.
Innovation Solution
An image processing apparatus and method that segment images into multiple embedding regions and select appropriate quantization conditions or multiplexing periods based on image information and region positions, allowing for precise reading of additional information regardless of capturing angle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If a tight back bookbinding method is used, then the photo book structure is stable and durable, but the gutter portion forms a valley fold that causes non-perpendicular capturing angles and makes additional information unreadable
Solution Approach 1:
The patent divides the image into multiple embedding regions (first embedding region and second embedding region) based on the valley fold position. Different quantization conditions are applied to each region, allowing the system to handle the geometric distortion caused by the valley fold while maintaining overall structure stability. This segmentation enables accurate information extraction despite the non-perpendicular capturing angle in the gutter portion.
Solution Approach 2:
The patent applies different quantization conditions to different embedding regions. The first quantization condition is used for the first embedding region and the second quantization condition is used for the second embedding region. This local differentiation allows optimal information embedding and extraction in each region, compensating for the varying geometric characteristics caused by the valley fold while preserving the stable tight back structure.
2Reliability
If additional information is embedded using high-frequency components, then the information can be hidden effectively, but the information becomes harder to read when captured at non-perpendicular angles
Solution Approach 1:
The patent dynamically adjusts the quantization conditions based on the embedding region position and the valley fold characteristics. Instead of using a fixed quantization approach, the system adapts the quantization parameters for different regions, allowing the high-frequency embedded information to remain hidden while maintaining readability even when captured at non-perpendicular angles due to the valley fold.
Solution Approach 2:
The patent changes the quantization parameters (first quantization condition and second quantization condition) based on the embedding region characteristics. By adjusting these parameters according to the local image properties and valley fold position, the system maintains the security of high-frequency information embedding while ensuring it remains readable under various capturing angles.
3Measurement precision
If the capturing angle is corrected by pushing open the gutter portion, then the capturing angle becomes perpendicular and information becomes readable, but the photo book may be destroyed
Solution Approach 1:
The patent performs preliminary analysis of the captured image to detect the valley fold and determine embedding region boundaries before information extraction. By pre-identifying the geometric distortion and dividing the image accordingly, the system can accurately extract information without requiring physical correction of the gutter portion, thus avoiding damage to the photo book while maintaining information readability.
Solution Approach 2:
The patent introduces an intermediary processing step that involves dividing the image into embedding regions and applying region-specific quantization conditions. This intermediary approach allows the system to compensate for the non-perpendicular capturing angle through computational methods rather than physical manipulation, extracting information accurately while preserving the photo book's structural integrity.
Data Source
AI summary
An image processing apparatus for embedding additional information in an image having a plurality of pixels and image information. A generating unit generates (a) the additional information to be embedded and (b) a bookbinding type of a medium containing the image. A holding unit holds a plurality of quantization conditions for embedding the additional information, including a quantization threshold. A selection unit segments the image into a plurality of embedding regions, and selects a condition to use in quantization based on the image information. A quantization condition control unit controls at least one of the plurality of quantization conditions for a predetermined region of the image based on the bookbinding type generated. An error diffusion processing unit distributes an error to peripheral pixels of a target pixel. A separating unit separates the embedded additional information from the image.