Partitioning Original Images into e x e Pixel Blocks for Fractal Analysis
- Login to Download
- 1 Credits
Resource Overview
Divide the original image into e x e pixel sub-blocks, identify maximum and minimum values in each block, calculate and quantize their differences, then compute fractal values using r x r sub-blocks with texture analysis implementation
Detailed Documentation
First, the original image is partitioned into sub-blocks of size e x e pixels. For each sub-block, the algorithm identifies the maximum and minimum pixel values, then calculates their difference range. This difference undergoes quantization processing to normalize the value range. Finally, fractal values are computed using r x r sized sub-blocks through dimension analysis algorithms. This methodology enables detailed texture feature analysis and extracts enhanced image information through multi-scale block processing. Implementation typically involves using matrix operations for block division, numpy.amax()/numpy.amin() functions for extremum detection, and differential quantization for value normalization before applying fractal dimension calculations.
- Login to Download
- 1 Credits