Search

Tax mapping is an essential element for efficient management of real estate property and tax management, paving the way for effective implementation of g-governance practices in India. Conventional mapping of individual buildings and their attributes involves large effort and is time-consuming. The use of high-resolution UAV imagery facilitates the process of automatic extraction of land parcel, roof area and height by providing accurate and latest building and landuse information for improved tax assessment and collection. Image segmentation has been performed on UAV imagery applied to differentiate rooftop surfaces from other features such as roads and vegetation using gray level thresholding. An example-based feature extraction algorithm was adopted to extract the land parcel and rooftop surfaces from the segmented image. Further, DSM and DTM derived from UAV imagery were used to determine the height of the buildings by extraction of terrain points and rooftop points. The extracted building height was used to estimate the numbers of floors in the building using thresholding. The number of floors and rooftop area were used to derive the total floor area of each building. The property tax to be levied for each building was calculated automatically using total floor area in efficient and scientific manner. The UAV imagery used in the study enabled rapid mapping of buildings parcels for tax assessment compared to conventional mapping methods.

The biomass content of vegetation has great significance for efficient balancing of biomass cycle. Estimation of biomass in vegetation requires quantification of tree canopy. Automatic extraction of tree canopy helps in faster and efficient estimation of the biomass content of a vegetative area in comparison to manual mapping methods. An agricultural farm at Papparapati Village in Dharmapuri, District, Tamil Nadu, with an Area of 1.2 sq.km was chosen as study area and multispectral UAV imagery at 4-cm spatial resolution with 4 Bands viz., NIR, Green, Red Edge, RED was acquired. In this study, an attempt has been made to automatically extract tree canopy using simple deep learning convolution neural network (CNN) using training sites as input from the band spectral information. The extracted tree canopy was refined using linear cluster algorithm to delineate the exact tree crown. The biomass of the tree is estimated using the tree properties (crown diameter, height) derived from UAV data and using the allometric equations. A field survey was conducted to validate the classification results achieved from UAV data of individual sample trees. It is observed that the accuracy of extracted tree properties was in agreement with the field measurements of sample trees.

Synthetic aperture radar (SAR) has operational applications in crop mapping and monitoring in many countries due to the distinct backscatter signature at various stages of crop growth. Polarimetric analysis of SAR data from different satellites was used for information extraction from different types of scatters in imaged terrain. The scattering processes were analyzed through the received scatter matrix derived from the target decomposition of SAR data. Three decomposition techniques, namely Freeman–Durden, Cloude–Pottier and Touzi decomposition of the ALOS PALSAR-1 data, were used in this study to extract land use/cover information with a specific emphasis on agriculture. The decomposed output parameters from these techniques were classified with supervised classifier of support vector machine (SVM) using region of interest (ROI) selected land use/cover classes. An accuracy assessment for the classified output was carried out using the ROI. The Ramgarh village in Jaisalmer district of Rajasthan with the predominance of agricultural land, sand dunes and settlements was chosen as the study area. Freeman–Durden decomposition resulted in the highest overall accuracy of about 85% in the land use/cover classification among the three decomposition techniques adopted in the study. It was also observed that the accuracy of land use/cover mapping derived from Cloude–Pottier and Touzi decompositions improved with the use of eigenvalues in the SVM classification. Higher accuracies in the classification of agriculture land were noted with all the three decomposition techniques. The four parameters of Cloude–Pottier (H, A, α, β) and Touzi (α s, Φ s, ψ, τ) decompositions improved the classification accuracy for all the classes due to eigenvalues. The overall classification accuracy was above 88% for both the decomposition techniques with four parameters. The soil moisture values for agriculture land and sand dunes were validated through soil moisture maps generated using Oh 1992 and 2004 models.