Raster and Vector Differences
We get questions from customers asking if Imagener Unlimited can leave images in vector form instead of converting the images back to raster (pixel-based) format for output.
Vector graphics are not pixels. They’re path instructions given to a digital device like a computer about the shape of an object. Each path has a start and an end point, and groups of paths can create a somewhat complex object.
Because vector graphics are groups of paths, objects created with vector graphics can be enlarged to any size without loss of visual clarity. Vectors are resolution independent until the vector-created object is output. Imagine a curve made with a piece of string that could be enlarged to any size because the supply of string is theoretically infinite, but in reality, is limited by the amount of computer memory. Because of this property people naturally believe turning a pixel-based image into vector information would allow the image to be enlarged to any size.
While vectors are good for mathematical shaping they’re not well suited for visual output in most instances. Scanners and cameras capture pixels – not vectors – for this reason. The flower image below shows part of the reason why.
Raster and Vector Differences
This image shows a typical flower image on the top and left and a vectorized version of the flower on the right. Notice how the right image shows “paths of image information” but the image beauty is missing.
Looking at the top leaf in the image even more closely, you can see how vector graphics – while perfect for viewing simple shapes, often gives beauty in imagery a cartoon-like appearance:
After mapping the image into tiny vectors, Imagener Unlimited uses the dimension information the user input to enlarge the vector mesh, then map the result back into pixels for output. Vector mappings exist only as shape and size information, not a pixel-row mapping necessary for a printer, display or other output.
Kneson’s Unlimited enlargement engine is a layer-based approach to enlarging images. Imagener Unlimited converts pixels into a fine, highly complex vector mesh during one stage of a 3-stage process. The other two stages are needed to record and restore the beauty of the source image.
This is why Imagener is memory intensive. The program has to hold the original image, the enlargement information plus memory for computation and sharpening. This is why Imagener has memory information in the Info tear-off box in the program:
Turning a raster (pixel)-based image into vector information is only one of Imagener Unlimited’s 3-stages of processing to enlarge images. Imagener Unlimited’s image processing performs an extremely intensive operation on images. It is performing a 3 stage transformation of the image as opposed to a straight enlargement algorithm as Imagener Enhanced or Professional performs. As such, Imagener Unlimited may not be the best enlargement engine for images that have simple backgrounds or are too small. See this newsletter we wrote about the different enlargement engines in Imagener – click here.
For those unfamiliar with the various versions of Imagener, be aware that Imagener Unlimited has the enlargement engines of Imagener Enhanced and Imagener Professional which are also far superior to what can be achieved in other expensive commercially available programs. Each enlargement engine is very different and enlarges images using a different methodology from the others.
We suggest trying your enlargement using those enlargement engines as well. They are found in the drop-down box labeled “Resample Method” — choose any of three additional enlargement engines, Kneson Progressive++, Kneson Progressive, or Kneson. These are three additional distinct enlargement engines that each have unique abilities to enlarge images depending on image content.