Under-screen Camera Image Reconstruction via Radial Basis Function
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Under-screen cameras in smartphones suffer from image degradation due to light diffraction, resulting in blurry images, and existing methods like deep learning networks are time-consuming for image reconstruction.
Innovation Solution
An electronic device trains a radial basis function network based on sample and reconstructed images, converting it into an image reconstruction model using a deconvolution algorithm to improve image clarity and efficiency, reducing reconstruction costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If deep learning networks are used for image reconstruction, then image clarity is improved, but reconstruction time and computational cost increase
Solution Approach 1:
The patent changes the algorithmic parameters by switching from deep learning networks to radial basis function networks combined with deconvolution algorithms. This parameter change maintains image reconstruction quality while significantly reducing computational complexity and processing time, directly resolving the contradiction between image clarity and reconstruction time
Solution Approach 2:
The patent substitutes the mechanical/computational system of deep learning networks with a different computational approach using radial basis function networks and deconvolution. This substitution preserves the functional outcome of image reconstruction while eliminating the time-consuming nature of deep learning methods
2Measurement precision
If deep learning networks are used for image reconstruction, then image quality is improved, but computational cost increases
Solution Approach 1:
The patent changes the computational parameters by adopting radial basis function networks with deconvolution algorithms instead of deep learning networks. This parameter change maintains high image quality while reducing computational cost through more efficient mathematical operations and simpler network architectures
Solution Approach 2:
The patent extracts and isolates the essential function of image reconstruction from the complex deep learning framework, implementing it through a simplified radial basis function network combined with deconvolution. This extraction removes unnecessary computational complexity while preserving the core image quality improvement function
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The method effectively enhances image clarity and reduces reconstruction time and costs for under-screen camera images by simulating point spread functions and applying a deconvolution algorithm.
Implementation Method 1
the projected light onto the under-screen camera has diffraction, which causes point sources of light to spread out when projected onto the under-screen camera
Data Source
AI summary
A method for reconstructing images, an electronic device and a storage medium is provided. In the method, the electronic device obtains an image to be reconstructed of an object captured by a target camera device, and obtains a sample image of the object. A radial basis function network is trained based on the image to be reconstructed and the sample image. An image reconstruction model is obtained by converting the radial basis function network based on a deconvolution algorithm, the image reconstruction model is configured for reconstructing images captured by the target camera device. By performing the method, a clarity of the image to be reconstructed can be improved.