Image Blur Sampling with Random Pixel Offsets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing blur algorithms often result in lattice or layering effects in blurred images due to insufficient sampling points when using larger sampling step sizes, which can't be effectively addressed without increasing computational load.
Innovation Solution
A method involving random offset processing to determine an offset pixel, followed by weighted summation of the color value, and the blurred color value, and the random offset pixel, and the resulting technical solution is the use of random offset processing to determine an offset pixel, followed by weighted summation of the color values of sampled points to achieve a blurred image without increasing the number of sampled points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a larger sampling step size is used in existing blur algorithms, then the computational load is reduced, but lattice or layering effects appear in the blurred image
Solution Approach 1:
The patent applies preliminary action by performing random offset processing on pixel positions before the blur sampling process. This pre-processing step randomly displaces pixel positions, which prevents the regular sampling pattern that causes lattice and layering effects, allowing larger sampling step sizes to be used without degrading image quality
Solution Approach 2:
The patent changes the parameter of pixel position by applying random offset processing. This transforms the original fixed pixel positions into randomly displaced positions, which breaks the regular sampling pattern and eliminates the lattice and layering artifacts while maintaining acceptable image quality with reduced computational load
Data Source
AI summary
Embodiments of the present disclosure provide an image processing method and apparatus and relate to image processing technologies. The method comprises: acquiring a pixel position of a first pixel point in an initial image, the first pixel point being any pixel point in the initial image; performing random offset processing on the pixel position of the first pixel point to acquire an offset position corresponding to the first pixel point; determining an offset pixel point corresponding to the first pixel point according to the offset position corresponding to the first pixel point; performing a blur processing on the offset pixel point corresponding to the first pixel point to obtain a blurred color value corresponding to the first pixel point; assigning the blurred color value corresponding to the first pixel point to a color value of the first pixel point to generate a blurred image corresponding to the initial image.