Image Background Blurring via Foreground Contour Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image background blurring methods suffer from low blurring accuracy and poor blurring effects, particularly in complex scenarios where the distinction between the shooting subject and background is not accurately differentiated.
Innovation Solution
A method and device that divide an image into foreground and background regions using a preset classification model, detect the contour edge of the shooting subject, and perform blurring based on this edge to enhance the blurring effect, employing techniques like deep learning for semantic segmentation and edge detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional background blurring methods are used, then the processing speed is fast, but the blurring accuracy is low and the shooting subject and background cannot be accurately differentiated
Solution Approach 1:
The image is divided into foreground region and background region through contour detection and region segmentation. The algorithm identifies the shooting subject's contour edges and separates the image into distinct foreground and background areas, enabling precise control over which regions undergo blurring processing.
Solution Approach 2:
Different blurring intensities are applied to different regions of the image. The background region undergoes strong blurring while the foreground region containing the shooting subject remains sharp. This local differentiation ensures high blurring accuracy for the background while preserving the clarity of the main subject.
2Manufacturing precision
If conventional background blurring methods are used, then the processing is simple, but the blurring effect is poor at edges of the shooting subject
Solution Approach 1:
The blurring radius is dynamically adjusted based on the distance from each pixel to the shooting subject's contour edge. Pixels closer to the edge have smaller blurring radii while pixels farther away have larger blurring radii, creating a natural gradient transition that avoids harsh edges and produces realistic blurring effects.
Solution Approach 2:
The algorithm introduces a spatial dimension to the blurring process by calculating distances from contour edges and using these distance values to determine blurring intensity. This transforms the simple binary blur decision into a continuous, spatially-varying operation that naturally handles edge cases.
Data Source
AI summary
A method and device for blurring an image background, a storage medium, and an electronic apparatus. The method for blurring an image background comprises: using a preset classification model to divide an image to be processed into a foreground region and a background region (201); detecting a contour edge of a photographic subject in the foreground region obtained from the division (S202); and performing blurring on a background of the image according to the contour edge of the photographic subject (S203).