Merging of IRS LISS III and pan data - evaluation of various methods for a predominantly agricultural area

Page 1 of 1

Merging of IRS LISS III and pan data - evaluation of various methods for a predominantly agricultural area

S. S. Ray and S. Panigrahy
Agro-Ecology and Management Division,
ARG/RESA, Space Applications Centre,
Ahmedabad - 380 015

Merging of multi-sensor image data has become a widely used procedure because of the complementary nature of various data sets. High spatial resolution is necessary for an accurate description of shapes, features and structures, whereas high spectral resolution is better used for land cover classification. Hence merging of these two types of data, to get multi-spectral images with high spatial resolution, is beneficial for various applications like vegetation, land-use, precision farming and urban studies. Various techniques are available for merging of multi-sensor image data. Most of these techniques have been used for SPOT Panchromatic and multi-spectral and Landsat TM data. Ideally, the method used to merge data sets with high-spatial resolution and high-spectral resolution should not distort the spectral characteristics of the spectral high spectral resolution data, particularly with respect to digital classification accuracy.

In the present study, six different methods have been used to merge the IRS PAN (high-satial resolution) and LISS III (high-spectral resolution) data for a predominantly agricultural area. This area covered a potato farm in Jalandhar, Punjab. The methods tried for merging were Intensity-Hue-Saturation (HIS), Principal Component Analysis (PCA), High Pass Filter (HPF), Brovey, Synthetic Variable Ration (SVR) and cubic spline wavelet technique. The LISS III data, which was of same date as that of Pan data, was registered to the Pan data taking GCPs and using cubic convolution method. Merged data products were generated using above-mentioned six techniques. The merged products were evaluated on three aspects, i.e. statistically, graphically and by comparing classification accuracy. The visual impressions of various merged products were also studied. The study could help to rank the suitability of vrious merging methods for agricultural landuse application.