Blur Sampling Layout for Faster High-Resolution Image Effects
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Applying image effects like blur to digital images is processor-intensive due to the detailed resolution, requiring either long processing times or powerful hardware, which varies with device capabilities.
Innovation Solution
A method involving accessing source image data, selecting target pixels, determining sample points with a central skew towards the target pixel, and generating output pixels based on sample point appearance values to apply a blur effect, which reduces the computational intensity by using a Gaussian blur effect, which reduces the computational intensity by determining output pixels based on sample point appearance values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If pixel-level processing is applied to digital images with high detailed resolution, then image processing precision is improved, but processing time increases and processor requirements increase
Solution Approach 1:
The patent segments the image processing task by introducing intermediate sample points between source pixels and target pixels. Instead of directly processing all source pixels for each target pixel, the system samples intermediate points to represent the blur kernel, reducing the computational burden while maintaining processing precision.
Solution Approach 2:
The patent introduces sample points as intermediary elements that mediate between source pixels and target pixels. These sample points serve as a compromise representation that captures blur characteristics without requiring direct pixel-level processing of the entire image, thus reducing processing time.
2Manufacturing precision
If pixel-level processing is applied to digital images with high detailed resolution, then image processing precision is improved, but processor power requirements increase
Solution Approach 1:
The patent segments the computational workload by using a reduced set of sample points to represent the blur kernel. This segmentation allows less powerful processors to handle the processing by dividing the complex pixel-level operations into manageable samples, reducing the overall processor power requirement.
Solution Approach 2:
The sample points act as intermediaries that reduce the computational complexity between source and target pixels. By using these intermediary samples rather than direct pixel mapping, the system achieves acceptable processing precision with lower processor power requirements.
3Ease of manufacture
If conventional blur processing methods are used, then image blur effect is achieved, but processing time becomes excessively long
Solution Approach 1:
The patent applies partial action by using a limited number of sample points to represent the blur kernel. Instead of processing all possible pixel combinations, the system uses a representative subset of sample points that provides sufficient blur effect with significantly reduced processing time.
Solution Approach 2:
The sample points serve as intermediaries that enable the blur effect to be applied efficiently. By mediating between source pixels and target pixels through these sampled points, the system achieves the desired blur effect without the excessive processing time associated with conventional methods.
Data Source
AI summary
Described herein is a computer implemented method including: accessing source image data defining a set of source pixels for an image; selecting a first target pixel from a set of one or more target pixels, the set of one or more target pixels corresponding to one or more unique source pixels of the set of source pixels; and generating a first output pixel corresponding to the first target pixel. Generating the first output pixel is by: determining a set of sample points having a sample point distribution within a blur sample area corresponding to an area of the image, wherein the sample point distribution has a central skew towards the first target pixel; determining a set of sample point appearance values based on the sample points and the image; determining a first blur appearance value based on the set of sample point appearance values; and generating the first output pixel based on the first blur appearance value.