Maria Vakalopoulou

Building detection in very high resolution multispectral data with deep learning features

The automated man-made object detection and building extraction from single satellite images is, still, one of the most challenging tasks for various urban planning and monitoring engineering applications. To this end, in this paper we propose an automated building detection framework from very high resolution remote sensing data based on deep convolutional neural networks. The core of the developed method is based on a supervised classification procedure employing a very large training dataset. An MRF model is then responsible for obtaining the optimal labels regarding the detection of scene buildings. The experimental results and the performed quantitative validation indicate the quite promising potentials of the developed approach.

Automatic Descriptor-Based Co-Registration of Frame Hyperspectral Data

Frame hyperspectral sensors, in contrast to push-broom or line-scanning ones, produce hyperspectral datasets with, in general, better geometry but with unregistered spectral bands. Being acquired at different instances and due to platform motion and movements (UAVs, aircrafts, etc.), every spectral band is displaced and acquired with a different geometry. The automatic and accurate registration of hyperspectral datasets from frame sensors remains a challenge. Powerful local feature descriptors when computed over the spectrum fail to extract enough correspondences and successfully complete the registration procedure. To this end, we propose a generic and automated framework which decomposes the problem and enables the efficient computation of a sufficient amount of accurate correspondences over the given spectrum, without using any ancillary data (e.g., from GPS/IMU). First, the spectral bands are divided in spectral groups according to their wavelength. The spectral borders of each group are not strict and their formulation allows certain overlaps. The spectral variance and proximity determine the applicability of every spectral band to act as a reference during the registration procedure. The proposed decomposition allows the descriptor and the robust estimation process to deliver numerous inliers. The search space of possible solutions has been effectively narrowed by sorting and selecting the optimal spectral bands which under an unsupervised manner can quickly recover hypercube’s geometry. The developed approach has been qualitatively and quantitatively evaluated with six different datasets obtained by frame sensors onboard aerial platforms and UAVs. Experimental results appear promising.

Simultaneous registration and change detection in multitemporal, very high resolution remote sensing data

In order to exploit the currently continuous streams of massive, multi-temporal, high-resolution remote sensing datasets there is an emerging need to address efficiently the image registration and change detection challenges. To this end, in this paper we propose a modular, scalable, metric free single shot change detection/registration method. The approach exploits a decomposed interconnected graphical model formulation where registration similarity constraints are relaxed in the presence of change detection. The deformation space is discretized, while efficient linear programming and duality principles are used to optimize a joint solution space where local consistency is imposed on the deformation and the detection space. Promising results on large scale experiments demonstrate the extreme potentials of our method.

Graph-Based Registration, Change Detection, and Classification in Very High Resolution Multitemporal Remote Sensing Data

In this paper, we propose a modular, scalable, metric-free, single-shot change detection/registration method. The developed framework exploits the relation between the registration and change detection problems, while under a fruitful synergy, the coupling energy term constrains adequately both tasks. In particular, through a decomposed interconnected graphical model, the registration similarity constraints are relaxed in the presence of change detection. Moreover, the deformation space is discretized, while efficient linear programming and duality principles are used to optimize a joint solution space where local consistency is imposed on the deformation and the detection space as well. The proposed formulation is able to operate in a fully unsupervised manner addressing binary change detection problems, i.e., change or no-change with respect to different similarity metrics. Furthermore, the framework has been formulated to address automatically the detection of from-to change trajectories under a supervised setting. Promising results on large scale experiments demonstrate the extreme potentials of our method.

Simultaneous registration, segmentation and change detection from multisensor, multitemporal satellite image pairs

In this paper, a novel generic framework has been designed, developed and validated for addressing simultaneously the tasks of image registration, segmentation and change detection from multisensor, multiresolution, multitemporal satellite image pairs. Our approach models the inter-dependencies of variables through a higher order graph. The proposed formulation is modular with respect to the nature of images (various similarity metrics can be considered), the nature of deformations (arbitrary interpolation strategies), and the nature of segmentation likelihoods (various classification approaches can be employed). Inference of the proposed formulation is achieved through its mapping to an overparametrized pairwise graph which is then optimized using linear programming. Experimental results and the performed quantitative evaluation indicate the high potentials of the developed method.

Multitemporal Very High Resolution From Space: Outcome of the 2016 IEEE GRSS Data Fusion Contest

In this paper, the scientific outcomes of the 2016 Data Fusion Contest organized by the Image Analysis and Data Fusion Technical Committee of the IEEE Geoscience and Remote Sensing Society are discussed. The 2016 Contest was an open topic competition based on a multitemporal and multimodal dataset, which included a temporal pair of very high resolution panchromatic and multispectral Deimos-2 images and a video captured by the Iris camera on-board the International Space Station. The problems addressed and the techniques proposed by the participants to the Contest spanned across a rather broad range of topics, and mixed ideas and methodologies from the remote sensing, video processing, and computer vision. In particular, the winning team developed a deep learning method to jointly address spatial scene labeling and temporal activity modeling using the available image and video data. The second place team proposed a random field model to simultaneously perform coregistration of multitemporal data, semantic segmentation, and change detection. The methodological key ideas of both these approaches and the main results of the corresponding experimental validation are discussed in this paper.

Detailed Land Cover Mapping from Multitemporal Landsat-8 Data of Different Cloud Cover

Detailed, accurate and frequent land cover mapping is a prerequisite for several important geospatial applications and the fulfilment of current sustainable development goals. This paper introduces a methodology for the classification of annual high-resolution satellite data into several detailed land cover classes. In particular, a nomenclature with 27 different classes was introduced based on CORINE Land Cover (CLC) Level-3 categories and further analysing various crop types. Without employing cloud masks and/or interpolation procedures, we formed experimental datasets of Landsat-8 (L8) images with gradually increased cloud cover in order to assess the influence of cloud presence on the reference data and the resulting classification accuracy. The performance of shallow kernel-based and deep patch-based machine learning classification frameworks was evaluated. Quantitatively, the resulting overall accuracy rates differed within a range of less than 3%; however, maps produced based on Support Vector Machines (SVM) were more accurate across class boundaries and the respective framework was less computationally expensive compared to the applied patch-based deep Convolutional Neural Network (CNN). Further experimental results and analysis indicated that employing all multitemporal images with up to 30% cloud cover delivered relatively higher overall accuracy rates as well as the highest per-class accuracy rates. Moreover, by selecting 70% of the top-ranked features after applying a feature selection strategy, slightly higher accuracy rates were achieved. A detailed discussion of the quantitative and qualitative evaluation outcomes further elaborates on the performance of all considered classes and highlights different aspects of their spectral behaviour and separability.

Patch-based deep learning architectures for sparse annotated very high resolution datasets

In this paper, we compare the performance of different deep-learning architectures under a patch-based framework for the semantic labeling of sparse annotated urban scenes from very high resolution images. In particular, the simple convolutional network ConvNet, the AlexNet and the VGG models have been trained and tested on the publicly available, multispectral, very high resolution Summer Zurich v1.0 dataset. Experiments with patches of different dimensions have been performed and compared, indicating the optimal size for the semantic segmentation of very high resolution satellite data. The overall validation and assessment indicated the robustness of the high level features that are computed with the employed deep architectures for the semantic labeling of urban scenes.

Per city-block, density estimation at build-up areas from aerial RGB imagery with deep learning

Estimating the density of the ‘urban fabric’ land cover classes is of major importance for various urban and regional planning activities. However, the generation of such maps is still challenging requiring significant time and labor costs for the per city-block analysis of very high resolution remote sensing data. In this paper, we propose a supervised classification approach based on deep learning towards the accurate density estimation of build-up areas. In particular, for the training procedure we exploit information both from maps (open street, google, etc) and from very high resolution RGB google image mosaics. A patch-based, deep learning model was trained against five land cover classes. During the prediction phase the per city-block classification procedure delivered the locations and percentages of impervious, soil and green regions. Experimental results and validation at two European cities i.e., Athens and Bilbao, indicated overall accuracy rates of 95%. Results, also, highly match with the corresponding layers from the Copernicus Urban Atlas product.

Integrating edge/boundary priors with classification scores for building detection in very high resolution data

Automatic and accurate detection of man-made objects, such as buildings, is one of the main problems that the remote sensing community has been focusing on for the last decades. In this paper, we propose a Conditional Random Field (CRF) formulation which is using edge/boundary localization priors towards accurate building detection. These edge priors have been integrated/fused with the classification scores from a deep learning Convolutional Neural Network (CNN) architecture under a single energy formulation. The validation of the developed methodology had been performed on the recently published SpaceNet dataset. Experimental results and quantitative evaluation, based on different accuracy statistics, indicate the great potential of the proposed approach.