Header menu link for other important links
An efficient image classification technique for synthetic aperture radar (SAR) images
S. Rajkumar,
Published in World Academy of Research in Science and Engineering
Volume: 9
Issue: 3
Pages: 3015 - 3022
The data processing in satellite is mostly treated in digital image processing techniques. Since, any satellite data are typically the digital image data. Further using the digital image processing algorithm, the remotely sensed information is measured for different applications. In recent years, the insinuation of extracting the Synthetic Aperture Radar (SAR) image features and classification are increasing. The classification in SAR images from remote sensing is the most effective scheme to extract and process the remote sensing data. Basically, the image classification is the process of assigning the individual pixels of an image into different categories and is one of the vital roles in satellite image processing today. Classifying the SAR image is the prerequisite for detecting and predicting the changes in geographical applications. Accordingly, this paper focuses the features such as correlation, texture, color etc. in the spectral feature space. Here, a novel image classification method called ‘Adaptive Classifier (AC)’ is developed that accurately classifies the water body and non-water body regions in various places of India such as Berijam Lake (Kodaikanal) and Kochi over a period of 3 years from 2016 to 2018. In this paper, there are 8 different regions in SAR images such as lake, river, sea, forest, barren land, building, agricultural land and transport were identified and classified. Through, qualitative and quantitative measures the classified result is compared with Nearest Neighbor (NN), k-Nearest Neighbor (KNN) and Support Vector Machine (SVM) and further proved that the proposed classification outperforms the existing techniques. © 2020, World Academy of Research in Science and Engineering. All rights reserved.
About the journal
JournalInternational Journal of Advanced Trends in Computer Science and Engineering
PublisherWorld Academy of Research in Science and Engineering