Under-display Camera Image Reconstruction via Pixel Gap Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The placement of front-facing cameras behind display surfaces in mobile devices often results in degraded image quality due to obstruction by display pixels, which limits the display area and light path.
Innovation Solution
A method and device that capture image data with blind regions caused by display pixels, processing first and second data segments to fill in these blind regions, using spatial alignment and interpolation to create a continuous image dataset, and adjusting image data based on the color spectrum of the display pixels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If the front-facing camera is placed behind the display surface, then the display area is increased, but the image quality is degraded due to obstruction by display pixels
Solution Approach 1:
The patent segments the image data into multiple portions corresponding to different regions captured through gaps between display pixels. By dividing the image capture task across multiple segments and processing them separately, the system recovers image data that would otherwise be blocked by display pixels, thereby maintaining image quality while keeping the camera behind the display.
Solution Approach 2:
The patent moves the camera from a traditional side-by-side arrangement with the display to a layered arrangement where the camera is positioned behind the display surface. This spatial reconfiguration in the depth dimension allows the display area to expand to the full front surface while the camera captures light through gaps in the display pixel structure.
2Area of stationary object
If the display pixels are placed in front of the camera, then the display can be positioned behind the camera, but blind regions are created in the field of view
Solution Approach 1:
The patent segments the field of view into visible regions (through gaps) and blind regions (blocked by pixels). By capturing multiple images from different positions and segmenting the data accordingly, the system reconstructs the complete image including previously blind regions, eliminating information loss.
Solution Approach 2:
The patent uses software processing and image synthesis as an intermediary to bridge the gap between the physical obstruction of display pixels and the desired complete image capture. The processing layer reconstructs blind regions by combining data from multiple captures, effectively mediating between the physical constraint and the information requirement.
3Manufacturing precision
If multiple images are captured to fill blind regions, then image quality is improved, but the time required for image capture increases
Solution Approach 1:
The patent employs periodic capture of multiple images in quick succession, where each capture is offset to reveal different portions of the scene through gaps in the display pixels. This periodic sampling approach allows efficient reconstruction of the complete image by combining multiple rapid captures rather than requiring a single prolonged exposure.
Solution Approach 2:
The patent performs preliminary capture of multiple offset images before final image synthesis. By pre-capturing the necessary data segments from different positions and storing them temporarily, the system prepares all required information in advance, enabling efficient reconstruction without requiring time-consuming processing during the capture moment itself.
Data Source
AI summary
A computer implemented method, device and a computer program product are provided. The method is under control of one or more processors configured with specific executable program instructions. The method displays content on a display layer of an electronic device. The display layer includes display pixels separated by gaps. The display layer overlays a camera in the electronic device such that the display pixels create blind regions in a field of view of the camera. The method captures, at the camera, image data in connection with a scene in a field of view of the camera. The image data includes first and second data segments that correspond to visible regions of the scene that are visible to the camera through the gaps. The first and second data segments are separated by a blind region in the scene that is blocked by the display pixels and is not visible to the camera. The method processes the first and second data segments to fill in the blind region therebetween to form a resultant data set that defines a continuous region of the scene spanning the visible regions and the blind region of the scene.