An automatization of Barnsley's algorithm for the inverse problem of iterated function systems.

Area of Science:

• Fractals and Dynamical Systems
• Computer Vision and Image Processing
• Geometric Modeling

Background:

• Iterated Function Systems (IFS) are fundamental in fractal geometry, used to generate complex fractal patterns.
• The inverse problem of IFSs involves determining the system's parameters from its generated fractal attractor.
• Manual methods for solving the IFS inverse problem exist but are labor-intensive and complex.

Purpose of the Study:

• To automate Barnsley's manual algorithm for solving the inverse problem of Iterated Function Systems (IFSs).
• To develop an algorithm capable of retrieving the number of mappings and their parameters from a digital binary image representing an IFS attractor.
• To explore the application of solving the IFS inverse problem for efficient binary image representation and coding.

Main Methods:

• Implementation of an automatic algorithm searching a finite parameter space for affine mappings.
• Utilizing the collage theorem and Hausdorff metric to ensure accuracy in parameter retrieval.
• Testing the algorithm's efficacy on digital binary images approximating IFS attractors.

Main Results:

• Demonstration that the automated algorithm successfully solves the inverse problem of IFSs.
• Experimental validation showing the algorithm's ability to retrieve both the number of mappings and their parameters.
• Confirmation that the retrieved IFS parameters accurately reconstruct the original fractal attractor.

Conclusions:

• The developed automatic algorithm provides an efficient and effective solution to the inverse problem of Iterated Function Systems.
• This automation facilitates the use of IFSs for compact binary image representation and compression.
• The study confirms the practical applicability of automated IFS parameter retrieval from image data.