Statistical Based Vectorization for
Standard Vector Graphics

 


SVGStat is a novel algorithm for raster to vector conversion. The technique is mainly devoted to vectorize digital picture maintaining an high degree of photorealistic appearance specifically addressed to the human visual system. The algorithm makes use of an advanced segmentation strategy based on statistical region analysis together with a few ad-hoc heuristics devoted to track boundaries of segmented regions. The final output is rendered by Standard Vector Graphics. Experimental results confirm the effectiveness of the proposed approach both in terms of perceived and measured quality. Moreover, the final overall size of the vectorized images outperforms existing methods.


Main Goal

To represent raster images using SVG primitives instead of a grid of pixels.

Similar approaches: Vector Eye, Autotrace, Kvec, SVGenie, SVGWave, SWaterG, VISTA.

Our Proposal: SVGStat

A Raster2Vector method based on:

  1. Partitioning of the original image in polygonal through Statistical Region Merging (SRM), a segmentation algorithm that belongs to the family of region growing techniques with statistical test for region fusion.
  2. Boundary representation of each region using SVG expressions:

Experimental Results

 

PSNR and bpp comparison

For sake of comparison SVG images obtained with our algorithm have been compared with other techniques such as Vector Eye, SWaterG, SVGenie, SVGWave.
The parameter of different techniques have been set in order to obtain images visually agreeable.

Visual Comparison


SVGenie

SVGStat

Vector Eye

 


VISTA

SVGStat