Image fractal interpolation method for optimizing particle swarm

Abstract

The present invention relates to an image fractal interpolation method for optimizing a particle swarm. The method comprises the following steps: 1) extracting a feature quantity from a to-be-interpolated image L(x, y); 2) performing fractal interpolation on the to-be-interpolated image; and 3) optimizing the interpolation parameter. The particle swarm-fractal method used in the present inventionhas high optimization and self-adaptability, and can perform an interpolation operation according to the feature of the image, the interpolation accuracy can completely satisfy the following image decomposition requirements, and the interpolation efficiency can also meet the real-time requirements of subsequent image decomposition; and thus, the image interpolation method based on the optimized particle swarm fractal can obtain good interpolation results, and also has good exploration value for the development of related theories and technologies.

Description

technical field [0001] The invention relates to an image processing method, in particular to an image fractal interpolation method for optimizing particle groups. Background technique [0002] Since the empirical mode decomposition algorithm was proposed, many scholars have continuously enriched and developed its interpolation algorithm. The interpolation algorithms currently used mainly include: polynomial interpolation, Akima interpolation, piecewise Hermit interpolation and spline interpolation. It is relatively simple to fit the extreme points obtained by interpolation in one-dimensional space, but the interpolation and fitting involved in the decomposition of two-dimensional empirical mode needs to complete related operations in two-dimensional space, and these problems become complicated. [0003] Two-dimensional empirical mode decomposition initially interpolates image signals according to rows and columns, then synthesizes the interpolation results, and then decompos...

AI Technical Summary

Problems solved by technology

This type of algorithm mainly includes: Nunes et al. proposed an envelope surface interpolation fitting method based on morphological operations and radial basis function interpolation (Radial Basis Functions, RBF) in 2003, but the calculation efficiency is low

Nunes then proposed an improved version of the algorithm in 2005, which achieved certain results, but did not fundamentally solve the problems of interpolation accuracy and interpolation efficiency; in 2004, Liu et al. from the Institute of Automation, Chinese Academy of Sciences proposed the DEMD algorithm. It has been improved, and the texture image processing effect is also better, but the calculation efficiency cannot be guaranteed

In 2011, Lei from Chongqing University proposed two-dimensional B-spline interpolation, which improved the interpolation effect, but still did not solve the problem of interpolation efficiency.

Therefore, the future development direction of the two-dimensional image signal interpolation algorithm should be reasonable, accurate and fast. This is because the pros and cons of the two-dimensional empirical mode decomposition interpolation algorithm directly affect its promotion and application, so the academic community pays more attention to the interpolation algorithm research. Rationality, accuracy and speed. For example, in 2016, Saad et al. from the University of Regensbug in Germany proposed a fast interpolation algorithm based on the Green function to achieve rapid image decomposition, but the interpolation performance is poor; in 2016, Xu from China University of Geosciences The kriging envelope interpolation method was proposed to try to solve the interpolation problem and apply it to geochemical identification, but the calculation efficiency is low

In summary, so far, no relatively good fast adaptive image interpolation algorithm has been proposed for the interpolation problem of the 2D empirical mode decomposition algorithm

Images