GEO-REFERENCING OF NON-CO-ORDINATED MAPS

Presenter : PRAVEEN KUMAR SINHA

Type: Poster

A major effort on softcopy conversion of Mineral Exploration Report by out sourcing was casted during the FS 2015-16 in Southern Region. The softcopy conversion involved digitisation of all text pages, tables, images and map plates of the reports. Map plates of the reports are scanned, geo-referenced and the geo-referenced Large Scaled and Detailed Maps were digitised, in thematically associated and geometrically segregated vector layers. In the process of digitisation a total 1723 nos. of reports are digitised.

The mineral exploration is carried out in small mineralised areas, happen to be “Potential Mineralised Blocks”. These block are geologically mapped by developing a grid of peg/ pillar points in Local Co-ordinate System (LCS) around a marked origin (X=0 m, Y=0) in the respective field and by developing a base line (normally oriented in E-W and N-S directions). The grid (at an interval usually 50 m x 50 m or 100 m x 100 m) is developed at right angle to the base line.  Dispositions of geological features, available road-network, village location, drainage lines etc. – base features lying into grids are mapped to corresponding paper grids, by available local measurements of distance and direction from nearby grid points. These maps are locally referenced and so they cannot instantly be brought to a Global Co-ordinate System for convenience of viewing localisation of blocks in respect of all other block areas in a single platform of space. Provided the authors have put an extra effort to the map by supplying the corner co-ordinates to the map. By geo-referencing maps on the provided co-ordinates on a map can be digitised to GCS made viewable along with other blocks.

The softcopy conversion is carried out with an objective to assemble the generated data-files to a server-setup as repository and bring these report through GSI Portal in public domain (policy in pipeline) so that the general public as individual, stake holders & industry, academia and government may have their access to the contents of the report for achieving respective pursuits. A general guideline is formulated by the GSI and an instruction is issued to all the Regional Centers, State Units and all participating Divisions of GSI that the text pages, tables, images and map plates of the reports need to be composed to searchable PDF file format and the map plate to GIS enabled, geo-referenced, vector shapefile format.

In conversion process the objectives of making the digital files available for the textual, tabular, image & photographical elements, scanned maps, plates, geo-referenced maps (if GCPs are provided) are achieved for the 1723 reports, however, out of the specified number of reports only 198 map plate of LSM/ DM nature with author provided co-ordinate references could be digitised.

Ground Control Points (GCP): If a map has 3 or more than 3 GCPs, the map can go for direct registration to Geographic Co-ordinate System (GCS) and specific geometry and thematically segregated layers need to be digitised directly on the GCS.

Non-Coordinated Maps: Majority of un-coordinated maps are left behind for absence of co-ordinates referenced to a global system. This is a paper upon how to meet the digitisation of those maps which do not have co-ordinates provided by authors to geo-reference them to a global system.

Whether a non-coordinated map (LCS) can be geo-referenced? Why not. When genetically non-associated conceptual model of spokes for longitude and latitude on GCS can be wrapped round the earth then a scaled map on LCS can also be brought to a GCP. This document provides a brief methodology/ steps to bring a map on LCS to GCP.

The conversion from LCS to GCS would require scalar transformations so the nature of scalar nature needs to be known on the maps. On survey five types of the detailed maps (DM) and large scaled maps (LSM) on LCS are recognised with the mineral exploration reports:

  • Maps with bar-scale,
  • Gridded maps: where grid point are joined with two set of parallel lines, mutually perpendicular to each other. The interval of the parallel lines from the base lines are marked on the map,
  • Map with only factor scale (i.e., 1: 10000),
  • Maps with village locations, segment/s road and drainage system as carto base for the geology on the map and there are a few maps
  • where none of the ways for depicting a scale is used nor carto base is marked on the maps, these maps have depiction of geological features

A good thing has happened in process of softcopy conversion, is that all the plate including map plates are scanned and put to designated sub-folder  “Scanned-Maps” with a designation to scan file ‘Map-Plate-Number” in the storage location (folder) for a report [identified by order of strings for, Generating center-Mission-Library Accession No.- Field Season, e.g.,  ‘SR-KG-M-II-256-1970-72’, stands  for a report of FS 1970-72   in the activity field covered by Mission-II (Mineral Exploration) of GSI, generated from State Unit : Karnataka & Goa of Southern Region, having: 256 – as its library accession No.]. The plates having 3 or more ground control points (GCPs) marked on the map have under gone geo-referencing. The rectified-resampled image of maps is saved in sub-folder for Geo-Referenced Maps. These geo-referenced maps (in GCS) are digitised and generated layers are saved in sub folder named: “Shapefiles”.

Drawing the extent-polygon for non-coordinated maps: Here the focus lies on the non-coordinated maps, as these maps have no co-ordinates marked over so it cannot be instantly referred to mapped piece of land and Imageries. These maps can be viewed in display device with unknown co-ordinates of a GIS. A conversion factor (f) may be arrived for the measured length of bar scale or grid interval on unknown co-ordinates (uc) and real world length of the bar scale or grid interval [i.e., f = real length/ measured length (uc)]. The factor may be utilised for conversion of measured map frame lines on unknown co-ordinate system to their lengths in Real World Units (RWU) aligned along N-S (RWU(N-S) = m.uc(N-S) *f) and E-W (RWU(E-W) = m.uc(E-W)*f). With the known length and width of a map frame in real world units, a polygon can be drawn (the polygon may be defined in WKT format as “POLYGON((0  0, 0 Xmax, Xmax  Ymax, 0  Ymax, 0  0))”. However, for the third category where scale is textually expressed in terms of ratio, the paper map would be required for measurement of the frame lengths and conversion to real world units.

Feature to feature matching between map and imagery / toposheet: Once the toposheet of detailed or large scale mapping is known by referring the text and locality index of a report, feature to feature matching, between map and imagery / toposheet would be a resort to an easy and direct operationalisation. For the fifth category of maps the geological features of detailed map to map50K geological features matching may bring the maps to closer geo-referenced positioning.

Crude Placement of the map & its frame to projected co-ordinate system and fine tuning of geo-referenceation: Now, the extent defining polygon may be placed into a cell of the grid developed for array of toposheets of the region (present context Southern Region of GSI). The placement may also be done on a plot of locality in projected co-ordinate system whose longitude & latitude are read from the text of the report. If the north pointer of the map depicts an angularity from true north, the drawn frame may be wended to the rotation. So that the frame acquires a rotated directional disposition without imposing an aerial deformity to the dimensions. This is a crude coordinated positioning of the frame where map image may be referenced by drawn 4 tie-lines, one by one between the corner point image to respective corner point of frame. Rectified map and its frame can be projection defined and re-projected to GCS.

Fine tuning of geo-reference & Digitisation of extents for mapped themes: Further refinement in the scan map may be achieved by matching the resembling features on scan map to that on Open Source Geomatics Resources, i.e., Google map, Street Map Layers or Scanned & geo-referenced toposheet. By the method of matching the features or by assigning the co-ordinates read on Google Imagery/ Scanned & geo-referenced toposheet in back ground, fine tuning of the geo-reference may be attained. After achieving the best possible fine tuning, the frame polygon needs to be re-digitised / re-positioned to corrected coordinates. Now the features of the map may be digitised.

Accurate positioning of the map / map-image: Field visit with a crudely referenced map and reading GPS for identifiable GCPs / DGPS can restore a geo-reference with precision to a map on LCS. However, for a large number of maps already described steps, i.e., frame drawing, frame positioning in identified toposheet / locality, fine tuning by feature matching with corresponding feature on imagery would a flow path to achieve geo-reference to the maps.