Edge-Guided Morphological Closing for Video Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional morphological closing operations in digital video processing often alter well-defined foreground subject edges and fail to preserve small spaces or features, leading to the disappearance of real holes in the foreground subject.
Innovation Solution
An edge-guided morphological closing method is introduced, where the number of neighboring pixels considered for processing is dependent on the presence of edges in the video frame, with dilation and erosion operations controlled by the edges of the foreground subject to prevent alteration of defined contours and preserve small features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional morphological closing operations are applied to fill holes in foreground subjects, then small holes and artifacts are removed, but well-defined edges and small spaces are altered or lost
Solution Approach 1:
The patent applies different processing behaviors to different regions of the image based on local characteristics. Edge pixels are identified and protected from morphological operations, while non-edge pixels undergo normal closing operations. This local differentiation allows holes to be filled in homogeneous regions without affecting the integrity of edge structures.
Solution Approach 2:
The image is segmented into edge pixels and non-edge pixels based on gradient magnitude thresholds. This segmentation allows the morphological closing operation to be selectively applied only to non-edge pixels, thereby preserving edge structures while still filling holes in the foreground subjects.
2Productivity
If morphological closing with fixed neighborhood size is used, then processing is simple and fast, but small features and spaces may disappear
Solution Approach 1:
The patent makes the neighborhood size adaptive rather than fixed. The size of the neighborhood considered during morphological closing varies locally based on whether a pixel is identified as an edge pixel or not. Non-edge pixels use a larger neighborhood for effective hole filling, while edge pixels use a restricted neighborhood to preserve features.
3Manufacturing precision
If edge detection is added to guide morphological closing, then edge preservation improves, but device complexity increases
Solution Approach 1:
The patent replaces complex edge-aware morphological operations with a simpler gradient-based edge detection mechanism. By using gradient magnitude calculations to identify edge pixels, the system achieves edge preservation without requiring sophisticated edge detection algorithms or complex conditional logic during the morphological operation itself.
Data Source
AI summary
A method of image processing, includes receiving at least one video frame of a video sequence, the at least one video frame including at least one foreground subject and a background, and processing the at least one video frame so as to separate the at least one foreground subject from the background. The processing includes: generating a pixel mask indicating whether a pixel of the at least one video frame belongs to the foreground subject or to the background, applying morphological closing to the pixel mask, wherein the applying morphological closing includes, for each pixel of the pixel mask, conditioning a pixel value in the mask to values of neighboring pixels. The conditioning includes: determining at least edges of the at least one foreground subject in the at least one video frame; and, for the generic pixel under processing, determining the neighboring pixels based on the determined edges.