Object-Based Convolutional Neural Network for High-Resolution Imagery Classification

Abstract

Timely and accurate classification and interpretation of high-resolution images are very important for urban planning and disaster rescue. However, as spatial resolution gets finer, it is increasingly difficult to recognize complex patterns in high-resolution remote sensing images. Deep learning offers an efficient strategy to fill the gap between complex image patterns and their semantic labels. However, due to the hierarchical abstract nature of deep learning methods, it is difficult to capture the precise outline of different objects at the pixel level. To further reduce this problem, we propose an object-based deep learning method to accurately classify the high-resolution imagery without intensive human involvement. In this study, high-resolution images were used to accurately classify three different urban scenes. The proposed method is built on a combination of a deep feature learning strategy and an object-based classification for the interpretation of high-resolution images. Specifically, high-level feature representations extracted through the convolutional neural networks framework have been systematically investigated over five different layer configurations. Furthermore, to improve the classification accuracy, an object based classification method also has been integrated with the deep learning strategy for more efficient image classification.

Existing System

Wavelet Transform, Sparse Support Vector Machine (SVM).

Proposed System

In this paper, we propose an effective way to classify high resolution images by combining deep features and image objects. Compared to the traditional classification methods, the proposed procedure utilizes deep CNN framework to automatically extract robust and discriminative features for complex urban objects classification (such as building roofs and cars). To evaluate the effectiveness of deep features, we tested the CNN framework with five different layer configurations for the classification of high-resolution imagery. However, as the CNN framework gets deeper, the generated features become more and more robust but often too abstract (overlooked the shape of the target objects) to describe boundary information. Complementary, the object-based classification method can preserve edge information in complex urban scenes which can be integrated with the highly abstracted deep features. As a solution, an object-based classification method is combined with deep features to promote the interpretation accuracy of high-resolution images.

Architecture

BLOCK DIAGRAM