On the Selective and Invariant Representation of DCNN for High-Resolution Remote Sensing Image Recognition

TL;DR

This paper investigates how deep convolutional neural networks (DCNNs) develop selective and invariant features for high-resolution remote sensing image recognition, using AlexNet and various analysis tools to understand their internal representations.

Contribution

It provides an in-depth analysis of the properties of DCNNs' feature representations in remote sensing, highlighting the importance of selectivity and invariance for image classification and understanding.

Findings

DCNNs exhibit strong invariance in HRRS image recognition.

Selective and invariant features are crucial for accurate remote sensing tasks.

Analysis tools reveal how neurons respond to different ground objects.

Abstract

Human vision possesses strong invariance in image recognition. The cognitive capability of deep convolutional neural network (DCNN) is close to the human visual level because of hierarchical coding directly from raw image. Owing to its superiority in feature representation, DCNN has exhibited remarkable performance in scene recognition of high-resolution remote sensing (HRRS) images and classification of hyper-spectral remote sensing images. In-depth investigation is still essential for understanding why DCNN can accurately identify diverse ground objects via its effective feature representation. Thus, we train the deep neural network called AlexNet on our large scale remote sensing image recognition benchmark. At the neuron level in each convolution layer, we analyze the general properties of DCNN in HRRS image recognition by use of a framework of visual stimulation-characteristic response combined with feature coding-classification decoding. Specifically, we use histogram statistics, representational dissimilarity matrix, and class activation mapping to observe the selective and invariance representations of DCNN in HRRS image recognition. We argue that selective and invariance representations play important roles in remote sensing images tasks, such as classification, detection, and segment. Also selective and invariance representations are significant to design new DCNN liked models for analyzing and understanding remote sensing images.