Image transformation techniques such as Principal Component Analysis, Minimum Noise Fraction and Independent Component Analysis were used for extracting information from the satellite image in various fields, but its application in landform mapping seems to be very less. The objective of the study is to demonstrate the viability of less explored image transformation techniques for delineating coastal dune complex and associated features using European satellite Sentinel 2 multispectral data. Based on the difference in lithology and vegetation cover the study sites has been chosen to extract coastal dune complex features such as dune ridges, dune with or without vegetation, swale with or without vegetation and water. Overall results of ICA shows enhanced visualization followed by PCA in bringing up the fine variations, identifying micro landforms within the dune complex from the optimum band combinations. Thus, sentinel data coupled with advanced image processing technique like ICA could identify maximum information about the boundaries and micro landform within the coastal sand dune complex in a cost-effective manner. However, in areas with dense vegetation cover none of these techniques has given good results in identifying the boundary or in identifying the micro landforms.

主成分分析，最小噪声分数和独立成分分析等图像转换技术用于从各个领域的卫星图像中提取信息，但在地形制图中的应用似乎很少。这项研究的目的是证明使用欧洲卫星Sentinel 2多光谱数据来描绘海岸沙丘复杂区和相关特征的较少探索的图像转换技术的可行性。根据岩性和植被覆盖率的差异，选择了研究地点以提取沿海沙丘的复杂特征，例如沙丘脊，有或没有植被的沙丘，有或没有植被和水的沼泽。ICA的整体结果显示出增强的可视化效果，随后PCA提出了细微的变化，从最佳波段组合中识别沙丘群内的微地形。因此，与诸如ICA的先进图像处理技术相结合的前哨数据可以以经济高效的方式识别有关沿海沙丘综合体中边界和微地形的最大信息。但是，在植被茂密的地区，这些技术在识别边界或识别微地形方面均未取得良好的结果。