 This paper investigates the use of convolutional neural networks, CNNs, for generating powerful feature representations for high-resolution remote sensing image retrieval, HRSA. First, a CNN pre-trained on a different problem is used as a feature extractor since there are no sufficiently sized remote sensing datasets to train a CNN from scratch. Then, a novel CNN architecture based on convolutional layers and a three-layer perceptron is proposed to learn low-dimensional features from limited labeled images. Finally, the proposed schemes are evaluated on several challenging, publicly available datasets. The results show that the proposed schemes, particularly the novel CNN, achieve state-of-the-art performance. This article was authored by Weixiu and Zhou, Sean Newsom, Tso Minli, and others.