Dual-Camera Digital Bokeh Masking for Unclassified Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The miniaturization of smartphone camera modules has increased the importance of digital bokeh technology, but existing methods struggle to accurately separate and blur background regions from object regions, particularly in unclassified areas.
Innovation Solution
An electronic device equipped with a first camera and a second camera of different focal lengths, along with a processor, identifies unclassified regions, obtains multiple zoom images, and generates a masking image to separate background and object regions, enabling improved digital bokeh performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If digital bokeh technology is used to separate and blur background regions, then the aesthetic quality and depth of field effect are improved, but the accuracy of separating object regions from background regions deteriorates, especially in unclassified areas
Solution Approach 1:
The patent segments the image processing task into multiple stages: first identifying unclassified regions, then obtaining zoom images specifically for those regions, generating masking candidate images through pixel-to-pixel XOR operations, and finally creating a composite masking image. This segmentation allows targeted processing of problematic areas without affecting the entire image, improving separation accuracy.
Solution Approach 2:
The patent performs preliminary actions by first identifying unclassified regions before the main processing, and by obtaining zoom images in advance for these regions. The masking candidate images are generated beforehand through XOR operations, and then averaged to create the final masking image. This preliminary processing ensures that object and background regions are properly distinguished before the final blur application.
2Measurement precision
If multiple zoom images are obtained and processed to improve classification accuracy, then the precision of object-background separation is improved, but the processing time and computational complexity increase
Solution Approach 1:
The patent applies partial action by obtaining zoom images only for unclassified regions rather than the entire image. The masking candidate images are generated only for these specific regions through pixel-to-pixel XOR operations, and then averaged to create the final masking image. This partial processing reduces computational time while maintaining high precision for the critical unclassified areas.
Data Source
AI summary
An electronic device includes: a first camera; a second camera having a second focal length that is different from that of the first camera; and at least one processor configured to: identify, within an input image obtained through the first camera, an unclassified region; obtain, through the second camera, a plurality of zoom images based on a zoom ratio of the second camera identified, by the at least one processor, to capture the unclassified region; identify a masking portion corresponding to an object, based on the plurality of zoom images; identify a background region image of the input image by determining the unclassified region as one of the background region and the object region based on the input image and the masking portion; and display an output image based on a blur processing for the background region image.