Study of Various Image Fusion Approaches for Extraction and Classification of Infrastructural Growth
- PDF / 1,268,144 Bytes
- 7 Pages / 547.087 x 737.008 pts Page_size
- 12 Downloads / 171 Views
SHORT ARTICLE
Study of Various Image Fusion Approaches for Extraction and Classification of Infrastructural Growth Jyoti Sarup & Akinchan Singhai
Received: 28 September 2010 / Accepted: 13 February 2012 / Published online: 29 March 2012 # Indian Society of Remote Sensing 2012
Abstract Estimation of infrastructural growth is the key issue for planning and resource management. In this regard it is highly required to have a proper database and documentation. Remotely sensed data and its processing techniques are most important parameter to achieve this goal. In developing countries, the planning and resource management is still dependent on traditional methods, but integration of satellite data of high resolution and of multiple spectral bands with appropriate processing techniques, makes it possible to get optimal result in limited fiscal resources. The merging of multi resolution sensor data can be the best option instead of using costly data for low budget planning and development. This study aims to analyze the potentials of image fusion of multispectral and panchromatic satellite data with high ground resolution images and evaluating their significance in infrastructural classification. While the accuracy assessment tests of classification result, suggest the appropriate classification techniques. The Relevance of image fusion in auto vectorization has also been discussed in this paper. J. Sarup (*) Department of Civil Engineering, Maulana Azad National Institute of Technology, Bhopal, India e-mail: [email protected] A. Singhai Centre for Remote Sensing and GIS, Department of Civil Engineering, Maulana Azad National Institute of Technology, Bhopal, India e-mail: [email protected]
Keyword Image fusion . Image classification . Accuracy assessment . Auto vectorization
Introduction For infrastructural planning, a comprehensive data base of the existing situation has to be established first (Rüdenauer & Schmitz 2010). Satellite imagery provides vital information related to infrastructural features and also helpful to categorize the existing infrastructure into urban settlement, industries, transportation routes etc. The high spatial resolution is necessary for an accurate description of feature geometry and structures, whereas high spectral resolution is better used to enhance the visual interpretation and feature display. An advanced image processing techniques like image fusion of multi sensor data can improve both spatial and spectral resolution which is beneficial in the limited fiscal environment and cost effective as well as compatible with costly high resolution data. These can be used for urban planning classification and resource mapping in developing countries. The present paper deals with the study to evaluate the significance of image fusion to classify infrastructural information more accurately. Different methods have been used to merge the IRS (Indian Remote Sensing) Satellite data Panchromatic (PAN) - high-spatial resolution and Linear Imaging Self Scanner (LISS III) -high-spectral reso
Data Loading...
