RSI-CB: A Large-Scale Remote Sensing Image Classification Benchmark Using Crowdsourced Data

Abstract

Remote sensing image classification is a fundamental task in remote sensing image processing. In recent years, deep convolutional neural network (DCNN) has seen a breakthrough progress in natural image recognition because of three points: universal approximation ability via DCNN, large-scale database (such as ImageNet), and supercomputing ability powered by GPU. The remote sensing field is still lacking a large-scale benchmark compared to ImageNet and Place2. In this paper, we propose a remote sensing image classification benchmark (RSI-CB) based on massive, scalable, and diverse crowdsource data.Using crowdsource data, such as Open Street Map (OSM) data, ground objects in remote sensing images can be annotated effectively by points of interest, vector data from OSM, or other crowdsource data. The annotated images can be used in remote sensing image classification tasks. Based on this method, we construct a worldwide large-scale benchmark for remote sensing image classification. This benchmark has two sub-datasets with 256 × 256 and 128 × 128 sizes because different DCNNs require different image sizes. The former contains 6 categories with 35 subclasses of more than 24,000 images. The latter contains 6 categories with 45 subclasses of more than 36,000 images. The six categories are agricultural land, construction land and facilities, transportation and facilities, water and water conservancy facilities, woodland, and other lands, and each has several subclasses. This classification system of ground objects is defined according to the national standard of land-use classification in China and is inspired by the hierarchy mechanism of ImageNet. Finally, we conduct many experiments to compare RSI-CB 1 with the SAT-4, SAT-6, and UC-Merced datasets on handcrafted features, such as scale-invariant feature transform, color histogram, local binary patterns, and GIST, and classical DCNN models, such as AlexNet, VGGNet, GoogLeNet, andResNet. In addition, we show that DCNN models trained by RSI-CB have good performance when transferred to another dataset, that is, UC-Merced, and good generalization ability.Experiments show that RSI-CB is more suitable as a benchmark for the remote sensing image classification task than other benchmarks in the big data era and has potential applications.

AID++: An Updated Version of AID on Scene Classification

“…In order to compare the generalization capabilities of different data sets training CNN models, we have designed multiple sets of transferring experiments. Several ResNet-50 models are trained on AID++, ImageNet [9], NWPU [10], RSI-CB [11], AID [3] respectively. And the pretrained models will be regarded to as a feature extractor classify the images of WHU-19, which is a small-scale remote sensing data set.…”

Remote Sensing Big Data Classification with High Performance Distributed Deep Learning

“…Currently, there are different solutions to secure the same performance level achieved on a single node with small mini-batch sizes despite the essential increase of the effective mini-batch size during distributed training. In the core of the simplest solutions is the tuning of the learning rate schedule that uses warm-up phases before the training, scales the learning rate with the number of distributed workers, and reduces the rate according to a fixed factor after a fixed number of epochs [6,44,50]. More sophisticated strategies to deal with very large batch sizes (for ImageNet, for instance, greater than 2 13 = 8192) use adaptive learning rates that are tuned dependent on network layer depth and the value of computed gradients and progress of training, such as that employed in LARS (Layer-wise Adaptive Rate Scaling)-an adaptive optimizer dedicated to large-scale distributed training setting [45,51].…”

“…From 1 December 2017 to 30 November 2018, the Sentinel Data Access System had a publication rate of over 26,500 products/day with an average daily download volume of 166 TB (https://sentinels.copernicus.eu/web/sentinel/news/-/article/2018-sentinel-data-accessannual-report). The large-scale, high-frequency monitoring of the Earth requires robust and scalable Machine Learning (ML) models trained over annotated (i.e., not raw) time series of multisensor images at global level [6,7] (e.g., acquired by Landsat 8 and Sentinel-2). However, these data do not exist yet.…”

Recent Advances and Opportunities in Scene Classification of Aerial Images with Deep Models

“…In other words, building largescale datasets for scene classification is reasonably desirable, which can reach the full potentials of deep CNNs on one hand, and accelerate the widespread use of new-generation deep learning models on the other hand. Very recently, it is noteworthy that tremendous efforts have been made to build large-scale benchmark datasets for scene classification., such as the AID [20], NWPU-RESISC45 [21] and RSI-CB [22], the total number of samples being 10, 000, 31, 500, and ∼ 36, 000 respectively. Compared with previous prevalent datasets, these new datasets occupy obvious superiority both in the total number and diversity of image samples.…”