Image Distortion Rectification via Vanishing Point Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Digital images often suffer from distortion due to involuntary camera rotations, causing parallel features to appear non-parallel and features to be misaligned, which existing technologies have not effectively addressed.

Innovation Solution

A method that extracts edgelets from an image, identifies vanishing points based on edgelet intersections, and uses these points to perform affine rectification and alignment transformations to correct image distortion, allowing parallel features to become more parallel and features to align with the image axes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If involuntary camera rotation is used to capture images, then image capture convenience is improved, but image geometric accuracy deteriorates

Engineering Contradiction:

Improveimage capture convenienceVSAvoidimage geometric accuracy

Core Design Contradiction:

Ease of operationVSManufacturing precision

Solution Approach 1:

The system automatically detects vanishing points and performs rectification transformations without user intervention. The image processing algorithm self-corrects the distortion by identifying geometric features (edgelets, vanishing points) and applying the appropriate affine transformation, making the convenience of handheld shooting compatible with geometric accuracy.

Inventive Principle:

Principle #25Self-service

2Manufacturing precision

If vanishing point-based transformation is applied, then image geometric accuracy is improved, but processing complexity increases

Engineering Contradiction:

Improveimage geometric accuracyVSAvoidprocessing complexity

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The image processing is broken down into discrete, manageable steps: edgelet extraction, vanishing point detection through intersection calculation, transformation matrix computation, and final image warping. This segmentation allows each step to be implemented independently and optimized, reducing overall processing complexity while maintaining geometric accuracy.

Inventive Principle:

Principle #1Segmentation

3Manufacturing precision

If aggressive rectification transformation is applied, then distortion correction is improved, but risk of introducing new artifacts increases

Engineering Contradiction:

Improvedistortion correction accuracyVSAvoidtransformation artifacts

Core Design Contradiction:

Manufacturing precisionVSObject-affected harmful factors

Solution Approach 1:

The system performs correctness checks on the transformed image to detect potential artifacts or failures. This feedback mechanism allows the system to evaluate the quality of rectification and potentially adjust parameters or reject transformations that would introduce harmful artifacts, balancing correction accuracy with image quality preservation.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS9064309B2Automatic rectification of distortions in images

Publication Date: 2015.06.23 GOOGLE LLC

AI summary

Implementations relate to relate to rectification of distortion in an image. In some implementations, a method includes extracting edgelets from an image, each edgelet defined by a location of a pixel having an edge depicted in the image and defined by a direction of the edge. The method finds at least one vanishing point in the image based on intersections of pairs of the edges of the edgelets. Each vanishing point is selected based on a plurality of the edgelets in the image having edges approximately aligned with the vanishing point. Found vanishing points are used to transform the image to reduce image distortion.