Classification of Oil Rigs in SAR Images Using RPCA-Based Preprocessing

• This paper uses a signal separation method called Robust Principal Component Analysis (RPCA) as a pre-processing technique to improve the classification of oil rigs in Synthetic Aperture Radar (SAR) images. After the pre-processing method, features are extracted from the images using the VGG-16 convolutional neural network. These features guide classification through Support Vector Machine (SVM), Neural Networks, and Logistic Regression algorithms. The experiments used SAR images from the Sentinel-1 system, C-band, and VH polarization. Early results highlight that preprocessing improves classification accuracy compared to conventional methods. © VDE VERLAG GMBH ∙ Berlin ∙ Offenbach.

Subject headings

TEKNIK OCH TEKNOLOGIER  -- Elektroteknik och elektronik -- Signalbehandling (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Electrical Engineering, Electronic Engineering, Information Engineering -- Signal Processing (hsv//eng)

Keyword

Image classification

Image enhancement

Neural networks

Principal component analysis

Radar imaging

Synthetic aperture radar

Convolutional neural network

Feature guides

Oil-rigs

Pre-processing method

Pre-processing techniques

Robust principal component analysis

Separation methods

Signal separation

Support vectors machine

Synthetic aperture radar images

Support vector machines

Publication and Content Type

ref (subject category)

kon (subject category)