Classification of Hyperspectral Images with CNN in Agricultural Lands

Hyperspectral images (HSI) offer detailed spectral reflectance information about sensed objects through provision of on hundreds narrow bands. HSI have a leading role in broad range applications, such as forestry, agriculture, geology, and environmental sciences. The monitoring management agricultural lands is great importance for meeting the nutritional other needs rapidly continuously increasing world population. In relation to this, classification an effective way creating land use cover maps quickly accurately. recent years, using convolutional neural networks (CNN), which sub-field deep learning, has become very popular research topic several CNN architectures been developed by researchers. aim this study was investigate performance model scenes. For purpose, 3D-2D framework well-known support vector machine (SVM) were compared Indian Pines Salinas Scene datasets that contain crop mixed vegetation classes. As result study, it confirmed offers superior classifying datasets.