Sky Segmentation via Canny Edge Convex Hull
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sky segmentation methods, whether edge-based or content-based, face challenges such as incomplete pixel-by-pixel segmentation, high computational costs, and inefficiency with limited labeled data and real-time constraints, particularly in varying weather and environmental conditions.
Innovation Solution
A method that involves detecting edges in an image, connecting them using morphological closing, identifying the largest contour, determining the convex hull of this contour, and generating a second image comprising the convex hull, which represents the ground region, allowing for accurate sky segmentation without requiring extensive training data or high computational resources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If content-based sky segmentation methods are used, then pixel-by-pixel segmentation accuracy is improved, but computational cost and data requirements increase significantly
Solution Approach 1:
The patent segments the image processing task into distinct stages: edge detection using Canny algorithm, morphological closing operations, contour identification, and convex hull computation. This segmentation allows each stage to be optimized independently, achieving accurate sky segmentation without requiring the full computational resources of content-based methods.
Solution Approach 2:
The patent extracts only the essential features needed for sky segmentation: edges detected by Canny algorithm and contours formed by morphological closing. By extracting and processing only these critical features rather than analyzing all pixel content, the method achieves accurate segmentation with reduced computational cost compared to full content-based approaches.
2Productivity
If traditional edge-based sky segmentation methods are used, then computational speed is improved, but segmentation completeness deteriorates
Solution Approach 1:
The patent merges multiple edge-based techniques into a unified approach: Canny edge detection identifies initial edges, morphological closing operations connect fragmented edges into complete contours, and convex hull computation finalizes the sky region segmentation. This combination maintains computational efficiency while achieving complete and accurate segmentation that individual edge-based methods cannot accomplish alone.
Solution Approach 2:
The patent performs preliminary edge detection using the Canny algorithm before main segmentation. This preliminary action identifies all potential edge locations, which are then processed by morphological closing to form complete contours. This staged approach ensures no segmentation details are missed while maintaining real-time processing capability.
3Measurement precision
If morphological closing with larger kernel area is used, then edge connection completeness is improved, but processing time increases
Solution Approach 1:
The patent dynamically adjusts the morphological closing kernel size based on image characteristics and processing requirements. By making the kernel area adaptable rather than fixed, the system can optimize the balance between edge connection completeness and processing time for different scenarios, achieving effective segmentation without unnecessary computational overhead.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A method includes receiving a first image. The method also includes detecting a plurality of edges in the first image. The method also includes connecting the edges. The method also includes identifying a contour in the first image based at least partially upon the connected edges. The method also includes determining a convex hull of the contour. The method also includes generating a second image comprising the convex hull.