view:42169 Last Update: 2020-1-8
Amir Salimi, Mansour Ziaii, Mahdieh Hosseinjani Zadeh, Ali Amiri, Sadegh Karimpouli
High performance of Support Vector Machine to classify hyperspectral data using limited data set
کارایی بالای ماشین بردار پشتیبان در طبقه بندی داده های ابر طیفی با استفاده از مجموعه داده های محدود
To prospect mineral deposits at regional scale, recognition and classification of hydrothermal alteration zones using remote sensing data is a popular strategy. Due to the large number of spectral bands, classification of the hyperspectral data may be negatively impressed by Hughes phenomenon. A practical way to handle Hughes problem is preparing a lot of training samples until the size of the training set be adequate and comparable with the number of the spectral bands. In order to gather adequate ground truth instances as training samples, a time consuming and costly ground survey operation is needed. In this situation that preparation enough field samples is not an easy task, using an appropriate classifier which can properly work with limited training data set is highly desirable. Among the supervised classification methods, Support Vector Machine is known as a promising classifier that can produce acceptable results even with limited training data. Here, this capability is evaluated when SVM is used to classify alteration zones of Darrehzar district. For this purpose, only 12 sampled instances from study area are utilized to classify Hyperion hyperspectral data with 165 useable spectral bands. Results demonstrate that if parameters of SVM, namely C and σ, are accurately adjusted, SVM can be successfully used to identify alteration zones when field data samples are not available enough.