Iterative Contour Identification for Digital Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for identifying contours in digital images, especially curved contours, are limited in their ability to accurately detect contours relevant for subsequent automated digital-image processing tasks, such as optical character recognition, due to constraints on local curvature and intensity gradients.
Innovation Solution
The method identifies seed points in digital images, extends line segments to form initial contours, and iteratively adds segments while applying filters to selectively combine and refine contours, ensuring only relevant contours are identified for subsequent processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional contour identification methods are used, then the processing speed is maintained, but the accuracy of detecting curved contours is insufficient
Solution Approach 1:
The patent segments the contour detection process into multiple stages: identifying seed points, extending initial line segments, iteratively adding segments, and filtering/combining contours. This segmentation allows each stage to be optimized independently, improving overall accuracy without sacrificing processing speed.
Solution Approach 2:
The patent performs preliminary actions by first identifying seed points and extending initial line segments before performing the main contour detection. This preliminary structuring of the detection process enables more accurate curved contour detection while maintaining efficiency through pre-computed information.
2Reliability
If contour identification is constrained to orthogonal intensity gradients, then the relevance of detected contours for OCR is improved, but the number of detectable contours is reduced
Solution Approach 1:
The patent changes the parameter constraints for contour detection by applying orthogonal intensity gradient constraints selectively during the iterative extension process. This allows the system to maintain high relevance for OCR applications while still detecting sufficient contours through adaptive parameter adjustment during processing.
Solution Approach 2:
The patent applies partial constraints by enforcing orthogonal intensity gradient requirements only at critical stages of contour extension rather than throughout the entire process. This partial application maintains OCR relevance while detecting a sufficient quantity of contours for effective character recognition.
3Manufacturing precision
If iterative contour extension with filtering is applied, then the precision of relevant contour identification is improved, but the computational complexity increases
Solution Approach 1:
The patent segments the complex iterative process into distinct, manageable stages: seed point identification, initial segment extension, iterative contour growth, and final filtering/combination. This segmentation reduces algorithmic complexity by making each stage independently optimizable and easier to implement.
Solution Approach 2:
The patent performs preliminary filtering and combination operations during the iterative extension process rather than as a separate post-processing step. This preliminary action reduces the overall computational complexity by eliminating redundant contours earlier in the process.
Data Source
AI summary
The current document is directed to automated methods and systems, controlled by various constraints and parameters, that identify contours in digital images, including curved contours. Certain of these parameters constrain contour identification to those contours in which the local curvature of a contour does not exceed a threshold local curvature and to those contours orthogonal to intensity gradients of at least threshold magnitudes. The currently described methods and systems identify seed points within a digital image, extend line segments from the seed points as an initial contour coincident with the seed point, and then iteratively extend the initial contour by adding line segments to one or both ends of the contour. The identified contours are selectively combined and filtered in order to identify a set of relevant contours for use in subsequent image-processing tasks.