A Discriminative Feature Learning Approach for Remote Sensing Image Retrieval

Abstract

Effective feature representations play a decisive role in content-based remote sensing image retrieval (CBRSIR). Recently, learning-based features have been widely used in CBRSIR and they show powerful ability of feature representations. In addition, a significant effort has been made to improve learning-based features from the perspective of the network structure. However, these learning-based features are not sufficiently discriminative for CBRSIR. In this paper, we propose two effective schemes for generating discriminative features for CBRSIR. In the first scheme, the attention mechanism and a new attention module are introduced to the Convolutional Neural Networks (CNNs) structure, causing more attention towards salient features, and the suppression of other features. In the second scheme, a multi-task learning network structure is proposed, to force learning-based features to be more discriminative, with inter-class dispersion and intra-class compaction, through penalizing the distances between the feature representations and their corresponding class centers. Then, a new method for constructing more challenging datasets is first used for remote sensing image retrieval, to better validate our schemes. Extensive experiments on challenging datasets are conducted to evaluate the effectiveness of our two schemes, and the comparison of the results demonstrate that our proposed schemes, especially the fusion of the two schemes, can improve the baseline methods by a significant margin.

Distribution Consistency Loss for Large-Scale Remote Sensing Image Retrieval

“…When using the ResNet50 network framework, on the UCMD dataset, our experimental results have increased this indicators by more than 3% in mAP, compared to the reference Pool5 (ResNet50) [47], which achieves a mAP value of 98.76%. At the same time, our method achieves the value of 100% in P@5, 100% in P@10, 99.33% in P@50, 49.82% in P@100 and 3.98% in P@1000, which surpassed the recently published ResNet50 [27] (91.90 in P@5, 91.40% in P@10 and 84.50% in P@50). When on the PatternNet dataset, compared with the recently published ResNet50 [27], DCL provides a best result of 99.43% in Map, and achieves 100% in P@5, 100% in P@10, 99.89% in P@50, 99.66% in P@100 and 16.38% in P@1000.…”

“…Finally, remote sensing image retrieval is performed through the features extracted by the network [26]. In order to obtain more discriminative features, the two dimensions of channel and space are weighted to obtain significant features [27]. The pre-trained RSIR method uses a trained overfeat network to extract RS features.…”

“…For the NWPU-RESISC45 dataset, we follow the data splitting strategy of 80% training and 20% testing as per [25]. NWPU-RESISC45 dataset [27] consists of 31,500 images, which is a large-scale RS image archive. It contains 45 classes: airplane, airport, baseball diamond, basketball court, beach, bridge, chaparral, church, circular farmland, cloud, commercial area, dense residential, desert, forest, freeway, golf course, ground track field, harbor, industrial area, intersection, island, lake, meadow, medium residential, mobile home park, mountain, overpass, palace, parking lot, railway, railway station, rectangular farmland, river, roundabout, runway, sea ice, ship, snowberg, sparse residential, stadium, storage tank, tennis court, terrace, thermal power station and wetland.…”

An End-to-End Local-Global-Fusion Feature Extraction Network for Remote Sensing Image Scene Classification

“…Here, an attention structure is proposed in Reference [17] to generate several attention representations based on the high-level features, which is the first use of attention mechanism in remote sensing. While the method in Reference [17] generates an attention mask for each element in the high-level feature map, Reference [39] proposes a new attention module to emphasize the meaningful features along the channel and spatial dimensions based on the high-level features. The effectiveness of this attention module has subsequently been verified in the task of remote sensing image retrieval.…”

Context-Aware Human Activity and Smartphone Position-Mining with Motion Sensors

“…Multi-Task Learning (MTL) is a learning paradigm in machine learning and its purpose is to take advantage of useful information contributed by multiple related tasks to improve the generalization performance of all the tasks [11]. MTL has shown significant advantage to single-task learning because of its ability to facilitate knowledge sharing between tasks [31], e.g., bioinformatics and health informatics [32,33], web applications [34,35] and remote sensing [36][37][38].…”