作者: Bisma Yousuf , Aparna Shukla , Manoj K. Arora , Ankit Bindal , Avtar S. Jasrotia
DOI: 10.1109/JSTARS.2019.2955955
关键词:
摘要: Subpixel classification (SPC) extracts meaningful information on land-cover classes from the mixed pixels. However, major challenges for SPC are to obtain reliable soft reference data (RD), use apt input data, and achieve maximum accuracy. This article addresses these issues applies support vector machine (SVM) retrieve subpixel estimates of glacier facies (GF) using high radiometric-resolution Advanced Wide Field Sensor (AWiFS) data. Precise quantification GF has fundamental importance in glaciological research. Efficacy approach was first tested synthetic followed by AWiFS MultiSpectral Instrument including ancillary resulted overall accuracy (OA) 95%, proving merit SVM. Classification is inversely related glacier's surface heterogeneity. Reducing number enhanced OA ∼18%. Source timing RD invariably controls improved ∼5% addressing issue temporal gap between RD. ∼11% increase with inclusion confirmed their positive effect Input fractional area were strongly correlated ( r > 0.9) each other substantiating results.