An automated soil line identification method using relevance vector machine

摘要: The soil line is an important concept that describes the linear relationship between reflectance of bare soils in near-infrared (NIR) and red (R) spectral bands. Bare parameters (slope intercept) are used calculating several vegetation indices. Previous studies have proposed both manual empirical procedures estimating parameters. Manual introduce some amount subjectivity identifying line. Empirical methods often suffer because variations caused by type, moisture, organic matter contents. existence non-bare pixels also affects these procedures. In this study, we automated supervised learning algorithm using relevance vector machine (RVM) for extracting from Landsat images. 10-fold cross validation (10-fold CV) indicated 92% accuracy distinguishing other image. area under receiver operating characteristic (ROC) curve reached a value ...