8We present an analysis of three possible strategies for exploiting the power of existing convolutional neural net-9 works (ConvNets or CNNs) in different scenarios from the ones they were trained: full training, fine tuning, and 10 using ConvNets as feature extractors. In many applications, especially including remote sensing, it is not feasible to 11 fully design and train a new ConvNet, as this usually requires a considerable amount of labeled data and demands 12 high computational costs. Therefore, it is important to understand how to better use existing ConvNets. We perform 13 experiments with six popular ConvNets using three remote sensing datasets. We also compare ConvNets in each 14 strategy with existing descriptors and with state-of-the-art baselines. Results point that fine tuning tends to be the best 15 performing strategy. In fact, using the features from the fine-tuned ConvNet with linear SVM obtains the best results. 16 We also achieved state-of-the-art results for the three datasets used.

In this paper, we evaluate the generalization power of deep features (ConvNets) in two new scenarios: aerial and remote sensing image classification. We evaluate experimentally ConvNets trained for recognizing everyday objects for the classification of aerial and remote sensing images. ConvNets obtained the best results for aerial images, while for remote sensing, they performed well but were outperformed by low-level color descriptors, such as BIC. We also present a correlation analysis, showing the potential for combining/fusing different ConvNets with other descriptors or even for combining multiple ConvNets. A preliminary set of experiments fusing ConvNets obtains state-of-the-art results for the well-known UCMerced dataset.

Semantic segmentation requires methods capable of learning high-level features while dealing with large volume of data. Towards such goal, Convolutional Networks can learn specific and adaptable features based on the data. However, these networks are not capable of processing a whole remote sensing image, given its huge size. To overcome such limitation, the image is processed using fixed size patches. The definition of the input patch size is usually performed empirically (evaluating several sizes) or imposed (by network constraint). Both strategies suffer from drawbacks and could not lead to the best patch size. To alleviate this problem, several works exploited multi-context information by combining networks or layers. This process increases the number of parameters resulting in a more difficult model to train. In this work, we propose a novel technique to perform semantic segmentation of remote sensing images that exploits a multi-context paradigm without increasing the number of parameters while defining, in training time, the best patch size. The main idea is to train a dilated network with distinct patch sizes, allowing it to capture multi-context characteristics from heterogeneous contexts. While processing these varying patches, the network provides a score for each patch size, helping in the definition of the best size for the current scenario. A systematic evaluation of the proposed algorithm is conducted using four high-resolution remote sensing datasets with very distinct properties. Our results show that the proposed algorithm provides improvements in pixelwise classification accuracy when compared to state-of-the-art methods.

In traditional semantic segmentation, knowing about all existing classes is essential to yield effective results with the majority of existing approaches. However, these methods trained in a Closed Set of classes fail when new classes are found in the test phase, not being able to recognize that an unseen class has been fed. This means that they are not suitable for Open Set scenarios, which are very common in real-world computer vision and remote sensing applications. In this paper, we discuss the limitations of Closed Set segmentation and propose two fully convolutional approaches to effectively address Open Set semantic segmentation: OpenFCN and OpenPCS. OpenFCN is based on the well-known OpenMax algorithm, configuring a new application of this approach in segmentation settings. Open-PCS is a fully novel approach based on feature-space from DNN activations that serve as features for computing PCA and multi-variate gaussian likelihood in a lower dimensional space. In addition to OpenPCS and aiming to reduce the RAM memory requirements of the methodology, we also propose a slight variation of the method (OpenIPCS) that uses an iteractive version of PCA able to be trained in small batches. Experiments were conducted on the well-known ISPRS Vaihingen/Potsdam and the 2018 IEEE GRSS Data Fusion Challenge datasets. OpenFCN showed little-to-no improvement when compared to the simpler and much more time efficient SoftMax thresholding, while being some orders of magnitude slower. OpenPCS achieved promising results in almost all experiments by overcoming both OpenFCN and SoftMax thresholding. OpenPCS is also a reasonable compromise between the runtime performances of the extremely fast SoftMax thresholding and the extremely

Land cover classification is a task that requires methods capable of learning high-level features while dealing with high volume of data. Overcoming these challenges, Convolutional Networks (ConvNets) can learn specific and adaptable features depending on the data while, at the same time, learn classifiers. In this work, we propose a novel technique to automatically perform pixel-wise land cover classification. To the best of our knowledge, there is no other work in the literature that perform pixel-wise semantic segmentation based on data-driven feature descriptors for high-resolution remote sensing images. The main idea is to exploit the power of ConvNet feature representations to learn how to semantically segment remote sensing images. First, our method learns each label in a pixel-wise manner by taking into account the spatial context of each pixel. In a predicting phase, the probability of a pixel belonging to a class is also estimated according to its spatial context and the learned patterns. We conducted a systematic evaluation of the proposed algorithm using two remote sensing datasets with very distinct properties. Our results show that the proposed algorithm provides improvements when compared to traditional and stateof-the-art methods that ranges from 5 to 15% in terms of accuracy.

Flooding is the world's most costly type of natural disaster in terms of both economic losses and human causalities. A first and essential procedure towards flood monitoring is based on identifying the area most vulnerable to flooding, which gives authorities relevant regions to focus. In this work, we propose several methods to perform flooding identification in highresolution remote sensing images using deep learning. Specifically, some proposed techniques are based upon unique networks, such as dilated and deconvolutional ones, while other was conceived to exploit diversity of distinct networks in order to extract the maximum performance of each classifier. Evaluation of the proposed methods were conducted in a high-resolution remote sensing dataset. Results show that the proposed algorithms outperformed state-of-the-art baselines, providing improvements ranging from 1 to 4% in terms of the Jaccard Index.

It is undeniable that aerial/satellite images can provide useful information for a large variety of tasks. But, since these images are always looking from above, some applications can benefit from complementary information provided by other perspective views of the scene, such as ground-level images. Despite a large number of public repositories for both georeferenced photographs and aerial images, there is a lack of benchmark datasets that allow the development of approaches that exploit the benefits and complementarity of aerial/ground imagery. In this paper, we present two new publicly available datasets named AiRound and CV-BrCT. The first one contains triplets of images from the same geographic coordinate with different perspectives of view extracted from various places around the world. Each triplet is composed of an aerial RGB image, a ground-level perspective image, and a Sentinel-2 sample. The second dataset contains pairs of aerial and street-level images extracted from southeast Brazil. We design an extensive set of experiments concerning multi-view scene classification, using early and late fusion. Such experiments were conducted to show that image classification can be enhanced using multi-view data.

No abstract