Authors: Alain Simons, Amar Aggoun and James Wood and Edmond Prakash
Photographic images, generally speaking, are represented by a rectangular array of pixels (picture elements). Each pixel is defined by its own color value. Describing the color value for each pixel is easy to compute but may end in large file sizes when using high pixel densities. With the evolution of HD screens an ever increasing amount of pixels needs to be displayed. In a modern world these screens are one way or another connected by the internet. As a result the internet is moving towards a bottleneck in terms of data transfer. Classical image compression techniques have reached their limitations, consumers are getting used to cutting edge images, cloud based patterns and daily use of online services. All of this indicates an urgent need to reduce online traffic where possible. As will be demonstrated the largest consumer of bandwidth is everything that involved images. The main goal of this research is to find other methods to define photographic images than those used today. Two requirements are set for the proposed method:
- To reduce file size.
- To equal or improve image quality.
To achieve these requirements a new technology is proposed to represent photographic images. The new technology, called VP (VectorPixels), is a vector based solution which relies heavily on the programmable OpenGL Graphic Pipeline. In short a photographic image will be converted to a vector based outline to be rendered again in the graphic pipeline.
Keywords: Pixel, VectorPixel, Photographic images, image quality, resolution independent, image enlargement