APPLICATION OF GEOINFORMATICS IN CADASTRAL MAPPING

Abstract

The existing cadastral maps in India date back to 80-90 years and these cadastral maps were prepared using the conventional techniques like plane table survey and chain survey. The manual digitization process or the survey/resurvey of the land parcels is a time-consuming and costly affair involving lots of resources. It is a challenging task for the administrators of the State Revenue Departments to conduct the survey for updation of the cadastral land records. It demands a cost-effective solution with an acceptable level of accuracy. Remote Sensing based Geomatics can be an answer for this. Traditionally the pixel is considered as the smallest unit in remote sensing and the same has been used for all sorts of analysis in remote sensing. Recently the technical enhancements facilitate to consider a group of pixels as a single unit and make the analysis of the images. The approach is known as object-based image analysis (OBIA). The OBIA consists of grouping pixels as objects, based on homogeneity criteria. Then these image objects are processed to generate the spatial and textural properties of these image objects. These properties are also known as object attributes. A suitable set of these object attributes are then processed for the classification of the image. This results in a better classification accuracy than the traditional pixel-based analysis as it takes both spatial and spectral characteristics into account. The OBIA approach will also facilitate the updating of cadastral maps with the High-Resolution Satellite Imagery (HRSI) and Unmanned Aerial Vehicle (UAV) Data Set and will help in quick conversion to vector formats. In this research, the OBIA analysis is used for the extraction of the cadastral boundaries from the existing cadastral maps, HRSI, and UAV data. The semi-automated approach has been proposed and the quality of the extraction is also determined to access the accuracy of the extraction. In this study, a mechanism has been explored to vectorize the cadastral maps after scanning using the OBIA technique. In the first step, the cadastral map is scanned. After scanning the raster map is cleaned by removing the plot numbers and other texts as written by the Patwaries. The discontinuities or the gaps found in the scanned lines were filled by image processing techniques. After the removal of the text and filling the gaps in the lines, the scanned image is treated as a remote sensing image. The OBIA techniques like segmentation, attribution, and classification are then applied to identify the parcels. The parcels were then exported as a vector layer.