Vectorization Of Contours From Scanned Topographical Maps

Vectorization Of Contours From Scanned Topographical Maps

SHARE


Dr.K.Ananthanarayanan
Associate Professor
Department of Civil Engineering,
Indian Institute of Technology Madras,
Chennai -600036, INDIA.


Dr Koshy Varghese
Professor
Department of Civil Engineering,
Indian Institute of Technology Madras,
Chennai -600036, INDIA.

Abstract:
It is proposed to develop an automated scheme for the extraction and vectorization of contour lines from a scanned topographic map. A methodology for the development and implementation of the proposed scheme is formulated. It consists of data source selection, image processing and raster to vector conversion.

Generation of Digital Elevation Models and selection of optimal route for cross country transmission line projects is a long standing demand of the Construction Industry. The Geographical Information System is really helpful to find a visually intuitive and efficient solution. For effective real time use of Geographical Information System for such applications, it is essential for the data to be available in vector format rather than raster format. Vectorization is the process of converting raster data into vector form. The most widespread source of height (elevation) data for digital elevation model generation is the contour lines in the topographical maps. One of the preliminary processes in Transmission Line Routing is the preparation of optimal routes and feasibility reports using a GIS. This is carried out by map digitization, preparation of digital elevation model for GIS and selection of the optimal route.

The following limitations exist in the present process

  1. A GIS cannot extract information about the different topographical features (data layers like contours, river systems, existing power lines etc.) Using scanned topographical maps in their raster form.
  2. The scanned topographical maps being utilized at present are in the form of raster images, which need to be vectorized before the GIS can use them. This activity, carried out manually using AUTOCAD, which is a time consuming process.
  3. The size of the raster bitmap file is huge which requires large amounts of computer memory storage capacity. Vectorization is a good alternative that can reduce the topo map files size, while maintaining data integrity.

The above limitations can be overcome by the use of Image Processing techniques to convert the raster Bit Maps into a Vector Format. The vectorized topographical map can then be input through the GIS and using computer run algorithms, an optimal transmission line route may be finalized.

SCANNING THE MAP

Scanner Parameter
Scanning is the first step towards obtaining a raster image before it is used for vectorization. The aim is to have a raster image which can be converted to a vector format automatically rather than manually. The parameters like resolution, Bit depth, Threshold value and Gamma correction of the scanner will influence the quality of the raster output image.

Resolution: resolution may be defined as the density of a raster image. The number of pixels that are present in a given unit region (inch) of the image which is defined as dots/pixels per inch (dpi). The Survey of India maps (fig1) which are often drafted with 0.2 mm tip ink pens, the actual colour lines are around 0.5 thick and the distance between adjacent contours is generally greater than the contour width itself. However the contour line thickness also makes a difference. A resolution of 300 dpi is sufficient to differentiate adjacent contours, most cases the raster contour line thickness varies from 3 to 8 pixels


Fig 1 A Scanned Survey of India Map             Fig 2. Sample Histogram