Raster to Vector Conversion for Maps
The Difference between Raster and Vector Maps
A raster map is basically an image that has been acquired by scanning a paper map with a large format scanner. The resulting file is normally stored in a TIF format and is comprised of individual pixels, with each individual pixel value creating the spatial location for a specific attribute, such as elevation, color, or some other identifying characteristic.
By comparison, vector maps are generally “drawn” maps with data that comes in the form of points and lines. The points and lines are related geometrically and mathematically, with points stored using the x-y coordinates (for two-dimensional) or x-y-z coordinates, when height or depth is taken into account.
Today, the most mapping is accomplished with the use of a Geographical Information System (GIS). Most of the major GIS software systems have some capability for handling either type of data (raster or vector), although each individual software package may more efficient at handling one or the other type of data. So, it is best to make sure that the GIS software package that you choose is most efficient with the type of data you will use.
File Sizes —In general, vector data produces a smaller file size than raster data, because a raster image needs space for all pixels while only point coordinates are stored in vector representation. Additionally, vector data is easier to handle than raster data because it has fewer data items and it is more easily adjusted for different size scales in a mapping application. For this reason, vector data is the prevalent choice for most mapping, GIS (Geographic Information System) systems.
Converting from Raster to Vector
Because the vector data structure is the more usual choice for handling graphical GIS data, acquiring accurate vector data from a scanned map raster file is essential. This means starting with an accurate scan, which can be a challenge. Often old maps have been stored in a damp basement or some other area that has not provided proper preservation. They may have gotten damp, water-marked and wrinkled. Or, maps that are on thermal paper may have badly faded images. Some older maps are very fragile with fraying edges that fall away as they are handled. This means that the very first, and one of the most important, steps in converting maps is to ensure that they are properly scanned so that the TIF file that will be used for the conversion from raster to vector is as accurate as it can be at the outset.
Scott Shuppert of CAD / CAM services recommends high accuracy scanners, like those manufactured by Contex. He also recommends that you not skimp on the technological expertise of the scanning personnel. Often times, companies with huge scanning jobs decide to purchase a scanner and hire inexperienced temporary employees to do the scanning. While this approach may save a few dollars at the beginning of the project, it nearly always results in wasted time and money when it comes to the raster to vector conversion process, because the scanned raster files require additional cleanup that should have been handled during the scanning process.
Challenges in Doing the Raster to Vector Conversion
While vector data provides a simpler and more abstract data representation than a raster image, it is not easy to do an automatic conversion from raster to vector. This vectorization process, especially with regards to tight contour lines in maps, can be quite challenging. Like the scanning process, raster to vector conversion of maps is best left to the experts.
A complete raster to vector conversion process includes image acquisition (scanning), pre-processing (cleaning, despeckling, rubber-sheeting, etc.), line tracing, text extraction (OCR), shape recognition, topology creation and attribute assignment.
The image acquisition process generates the initial raster image at a certain spatial resolution. The quality and resolution of the raster image are key factors for the quality and accuracy of the vectorized data. It is always recommended to start with clean and sharp originals and scan at a reasonable resolution. The scanning resolution should match the resolution at which the original image source was created. If scanning resolution is set too high than the original image source, it not only uses an unnecessary amount of system resource to process but also noise and artifacts are scanned, then need to be cleaned out of the image.
When the color classification and color separation are required, it’s important to remember that these processes are extremely sensitive to the color quality of the scanned image. Additionally, other types of color images — such as satellite and aerial photos — may be incorporated into a GIS and used to directly create vector data, such as boundaries and roads. This type of imagery generally consists of very large raster files that require more system resources and a higher level of expertise to handle.
How to Get the Raster to Vector Expertise You Need
Find a raster to vector conversion service that can offer you case studies and references documenting their expertise in GIS raster to vector conversions.