Web Browser Selection Overlay Using Polygonal Approximation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing web browser graphical object selection techniques struggle with precision when dealing with overlapping objects, as they rely on rectangular bounding boxes that can include transparent pixels, making it difficult to select non-rectangular objects and those partially occluded.
Innovation Solution
Implementing a selection overlay using IMAGEMAP elements with polygonal approximation of selection regions, allowing for finer-grained selection by defining shapes that accurately represent the visible area of graphical objects, even when they overlap or are occluded.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If rectangular bounding boxes are used for graphical object selection, then the selection process is simple to implement, but the selection precision deteriorates when objects overlap or are non-rectangular
Solution Approach 1:
The patent divides the selection process into two distinct phases: a coarse selection phase using rectangular bounding boxes for initial object identification, and a fine selection phase using polygonal approximations for precise object boundary definition. This segmentation allows the system to benefit from both the simplicity of rectangular selection and the precision of custom-shaped selection regions.
Solution Approach 2:
The selection region dynamically transitions from a static rectangular bounding box to a dynamic polygonal approximation that adapts to the actual shape of the graphical object. The polygon vertices are calculated based on the object's visible pixels, allowing the selection boundary to conform precisely to non-rectangular shapes while maintaining ease of implementation through automated calculation.
2Ease of operation
If rectangular bounding boxes include transparent pixels, then the selection area is simplified, but the ability to distinguish overlapping objects deteriorates
Solution Approach 1:
The patent applies different quality criteria to different parts of the selection process. The rectangular bounding box provides a coarse selection area that is easy to define, while the polygonal approximation provides a fine-grained selection boundary that precisely follows object edges. This local differentiation allows transparent pixels to be excluded from the final selection region, improving object discrimination while maintaining ease of operation through the two-stage approach.
Solution Approach 2:
The patent introduces an intermediary calculation step that processes the rectangular bounding box selection to generate a polygonal approximation. This intermediary process analyzes the visible pixels within the bounding box and constructs a polygon that accurately represents the object's true boundary, thereby mediating between the simplicity of rectangular selection and the precision needed to distinguish overlapping objects.
3Measurement precision
If polygonal approximation is used for selection regions, then selection precision is improved, but the computational complexity increases
Solution Approach 1:
The patent implements a partial action approach by first performing a coarse rectangular selection that captures the general object area, then applying polygonal approximation only to the selected region rather than to all objects on the page. This partial application of the computationally intensive polygon calculation significantly reduces overall computational complexity while maintaining high selection precision where it is most needed.
Solution Approach 2:
The patent performs preliminary action by first establishing the rectangular bounding box selection region before calculating the polygonal approximation. This preliminary step defines a limited area for subsequent detailed processing, reducing the computational scope of the polygon calculation to only the necessary region and thereby lowering overall computational complexity while preserving selection precision.
4Adaptability or versatility
If traditional selection methods are used, then browser compatibility is maintained, but the ability to select occluded objects deteriorates
Solution Approach 1:
The patent implements a universal selection system that maintains compatibility with traditional rectangular bounding box selection methods while adding the capability for polygonal approximation selection. The system can operate in both modes depending on the situation, allowing it to work with existing browsers and tools that expect rectangular selections while also providing enhanced functionality for selecting occluded and non-rectangular objects when needed.
Data Source
AI summary
Various approaches for selection of graphical objects in a web browser-based application are described. In one approach, a selection overlay is stored in a web document. The selection overlay encompasses a plurality of user-selectable graphical objects in the web document occupies a highest position in layers of graphical objects in the web document and includes a plurality of selection regions. Each selection region is respectively associated with one of the plurality of graphical objects and defines a region of selectability for the associated graphical object. In response to the web browser-based application detecting entry of a user-directed pointer into one of the selection regions in the selection overlay, the graphical object associated with the one selection region is selected for user manipulation and displayed.