Dewarped Book Image Generation via Cylindrical 3D Surface Modeling

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for dewarping images from document pages using consumer-grade stereo cameras face challenges with perspective distortion and require complex hardware, limiting their effectiveness with low-cost devices and affecting OCR performance.

Innovation Solution

A method involving a multi-resolution pipeline that generates a cylindrical 3D geometric surface from stereo photos, allowing for high-resolution dewarping using consumer-grade cameras without relying on text content analysis, and simplifying camera calibration for use across different resolution settings.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If structured light is used to capture 3D information, then measurement precision is improved, but device complexity increases

Engineering Contradiction:
Improve3D information accuracyVSAvoidhardware complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces structured light (optical/mechanical system) with a computational approach using stereo camera images and cylindrical surface modeling. The 3D information is derived through image processing algorithms rather than physical structured light projection, thereby reducing hardware complexity while maintaining measurement capability.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces a cylindrical surface model as an intermediary between the stereo camera images and the final dewarped output. This mathematical model mediates the transformation from 2D images to 3D representation and back to flattened 2D output, enabling accurate dewarping without requiring complex hardware.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If multiple cameras are used to capture 3D information, then measurement precision is improved, but device complexity increases

Engineering Contradiction:
Improve3D information accuracyVSAvoidcamera system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent makes the stereo camera system universal by enabling it to perform both 3D capture and accurate dewarping through a unified cylindrical surface model. The same stereo camera that captures images also provides the geometric information needed for dewarping, eliminating the need for separate specialized equipment.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Manufacturing precision

If energy minimization and conformal mapping are used for dewarping, then manufacturing precision is improved, but device complexity increases

Engineering Contradiction:
Improvedewarping accuracyVSAvoidprocessing complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent changes the mathematical parameters from complex energy minimization and conformal mapping to a simpler cylindrical surface model with fixed geometric parameters. This parameter transformation simplifies the dewarping process while maintaining accuracy, as the cylindrical model provides a straightforward mathematical framework for distortion correction.

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentUS10798359B2Generating hi-res dewarped book images
Publication Date: 2020.10.06 FUJIFILM BUSINESS INNOVATION CORP
  • US10798359B2 patent drawing
  • US10798359B2 patent drawing
  • US10798359B2 patent drawing

AI summary

Systems and methods for generating high resolution dewarped images for an image of a document captured by a 3D stereo digital camera system, or a mobile phone camera capturing a sequence of images, which may improve OCR performance. Example embodiments include a compact stereo camera with two sensors mounted at fixed locations, and a multi-resolution pipeline to process and to dewarp the images using a three dimensional surface model based on curve profiles of the computed depth map. Example embodiments also include a mobile phone including a camera which captures a sequence of images, and a processor which computes a disparity map using the captured sequence of image frames, computes a model of the at least one document page by generating a cylindrical three dimensional geometric surface using the computed disparity map, and renders a dewarped image from the computed model.