Lightweight Parallel Convolutional Neural Network With SVM Classifier for Satellite Imagery Classification

Abstract

Satellite image classification is crucial for various applications, driving advancements in convolutional neural networks (CNNs). While CNNs have proven effective, deep models often encounter overfitting issues as the network's depth increases since the model has to learn many parameters. Besides this, traditional CNNs have the inherent difficulty of extracting fine-grained details and broader patterns simultaneously. To overcome these challenges, this article presents a novel approach using a lightweight parallel CNN (LPCNN) architecture with a support vector machine (SVM) classifier to classify satellite images. At first, preprocessing such as resizing and sharpening is used to improve image quality. Each branch within the parallel network is designed for specific resolution characteristics, spanning from low (emphasizing broader patterns) to high (capturing fine-grained details), enabling the simultaneous extraction of a comprehensive set of features without increasing network depth. The LPCNN incorporates a dilation factor to expand the network's receptive field without increasing parameters, and a dropout layer is introduced to mitigate overfitting. SVM is used alongside LPCNN because it is effective at handling high-dimensional features and defining complex decision boundaries, which improves overall classification accuracy. Evaluation of two public datasets (EuroSAT dataset and RSI-CB256 dataset) demonstrates remarkable accuracy rates of 97.91% and 99.8%, surpassing previous state-of-the-art models. Finally, LPCNN, with less than 1 million parameters, outperforms high-parameter models by effectively addressing overfitting issues, showcasing exceptional performance in satellite image classification.