Image Border Definition Using Edge Detection and Bezier Splines

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image editing tools face challenges in defining borders quickly and accurately, allowing for natural transitions from foreground to background, and correcting mistakes during the border definition process, especially when dealing with complex images like those with multiple colors or textures.

Innovation Solution

The solution involves generating a tunnel around the defined border with adjustable width and shape, allowing for modifications, and using an algorithm to determine alpha values for pixels based on sampled exterior pixels, enabling efficient foreground-to-background transitions and color correction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a user draws a border by manually moving the cursor along the desired path, then the border can be defined with precision, but the process becomes very time-consuming and difficult

Engineering Contradiction:
Improveborder definition precisionVSAvoidtime to define border
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces the manual mechanical cursor movement system with an intelligent automated system that uses image analysis algorithms. The system automatically detects edges, colors, and patterns in the image to suggest and refine border paths, substituting manual mechanical tracing with computational intelligence to achieve both precision and speed.

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

Solution Approach 2:

The border definition system performs self-service by automatically analyzing the image content and proposing border paths without requiring continuous manual intervention. The system uses algorithms to detect natural boundaries, color transitions, and edge features, allowing the border definition process to serve itself through intelligent automation rather than relying solely on manual cursor movement.

Inventive Principle:
Principle #25Self-service

2Device complexity

If a user must complete the entire border definition in one continuous action, then the process may be simpler to implement, but any mistake requires starting over which is frustrating and time-consuming

Engineering Contradiction:
Improveborder definition process complexityVSAvoidease of border correction
Core Design Contradiction:
Device complexityVSEase of operation

Solution Approach 1:

The border definition process is segmented into multiple independent sections or segments. Users can define portions of the border separately, and each segment can be corrected or modified independently without affecting the entire border. This segmentation allows users to make mistakes in one area without having to restart the entire border definition process.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The border definition system is made dynamic and flexible, allowing users to add, remove, or modify border segments at any point during the definition process. The system dynamically updates the border as users make corrections, enabling iterative refinement without requiring a complete restart, thus improving ease of operation.

Inventive Principle:
Principle #15Dynamics

3Ease of manufacture

If a simple softening effect is applied to the border edge, then the implementation is straightforward, but it does not create a natural transition from foreground to background

Engineering Contradiction:
Improveease of border processingVSAvoidnatural transition quality
Core Design Contradiction:
Ease of manufactureVSReliability

Solution Approach 1:

The system dynamically changes multiple parameters along the border edge including transparency, blending mode, edge width, and color interpolation. These parameter variations are applied based on local image characteristics such as color transitions, texture patterns, and edge orientation, creating natural-looking transitions that adapt to different regions of the image rather than applying a uniform softening effect.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

Different border processing qualities are applied to different local regions of the border. The system analyzes local image characteristics and applies appropriate transition effects - such as feathering, blending, or sharp edges - depending on the specific area. This local quality approach ensures natural transitions in complex areas while maintaining simplicity in uniform regions.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP2431942B1Defining a border for an image
Publication Date: 2013.09.11 APPLE INC
  • EP2431942B1 patent drawingFigure 1
  • EP2431942B1 patent drawingFigure 2
  • EP2431942B1 patent drawingFigure 3

AI summary

Some embodiments provide a method for selecting a portion of an image. The method identifies edges in the image. The method defines a border about the portion of the image by using the identified edges. The method represents the border as a deformable curve. In some embodiments, defining the border includes detecting a cursor moving over the image and defining the border along identified edges in the vicinity of the cursor. In some embodiments, the method searches for edges in the vicinity of the cursor and snaps the border to the edges. Identifying the edges of the image includes performing an edge detection algorithm in some embodiments. Identifying the edges further includes performing a de-noise algorithm in some embodiments. In some embodiments, the parametrizable curve is a bezier spline.