Search

Land use is an important variable in remote sensing which describes the functions carried out on a piece of land in order to obtain benefits and is especially useful to the personnel working in the fields of urban management and planning. The land use information is maintained by national mapping agencies in geo-spatial databases. Commonly, land use data is stored in the form of polygon objects; the label of the object indicates land use. The main goal of classification of land use objects is to update an existing database in an automatic process. Recently, Convolutional Neural Networks (CNN) have been widely used to tackle this task utilizing high resolution aerial images (and derived data such as digital surface model). One big challenge classifying polygons is to deal with the large variation in their geometrical extent. For this challenge, we adopt the method of Yang et al. (2019) to decompose polygons into regular patches of fixed size. The decomposition leads to two sets of polygons: small and large, where the former suffers from a lower identification rate. In this paper, we propose CNN methods which incorporate dense connectivity and integrate it with intermediate information via global average pooling to improve land use classification, mainly focusing on small polygons. We present different network variants by incorporating intermediate information via global average pooling from different stages of the network. We test our methods on two sites; our experiments show that the dense connectivity and integration of intermediate information has a positive effect not only on the classification accuracy on the whole but also on the identification of small polygons. © 2020 International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences - ISPRS Archives.

Pedestrian tracking is a significant problem in autonomous driving. The majority of studies carries out tracking in the image domain, which is not sufficient for many realistic applications like path planning, collision avoidance, and autonomous navigation. In this study, we address pedestrian tracking using stereo images and tracking-by-detection. Our framework comes in three primary phases: (1) people are detected in image space by the mask R-CNN detector and their positions in 3D-space are computed using stereo information; (2) corresponding detections are assigned to each other across consecutive frames based on visual characteristics and 3D geometry; and (3) the current positions of pedestrians are corrected using their previous states using an extended Kalman filter. We use our tracking-to-confirm-detection method, in which detections are treated differently depending on their confidence metrics. To obtain a high recall value while keeping a low number of false positives. While existing methods consider all target trajectories have equal accuracy, we estimate a confidence value for each trajectory at every epoch. Thus, depending on their confidence values, the targets can have different contributions to the whole tracking system. The performance of our approach is evaluated using the Kitti benchmark dataset. It shows promising results comparable to those of other state-of-the-art methods.

Interactive motion planing and collaborative positioning will play a key role in future autonomous driving applications. For this purpose, the precise reconstruction and pose estimation of other traffic participants, especially of other vehicles, is a fundamental task and will be tackled in this paper based on street level stereo images obtained from a moving vehicle. We learn a shape prior, consisting of vehicle geometry and appearance features, and we fit a vehicle model to initially detected vehicles. This is achieved by minimising an energy function, jointly incorporating 3D and 2D information to infer the model’s optimal and precise pose parameters. For evaluation we use the object detection and orientation benchmark of the KITTI dataset (Geiger et al., 2012). We can show a significant benefit of each of the individual energy terms of the overall objective function. We achieve good results with up to 94.8% correct and precise pose estimations with an average absolute error smaller than 3° for the orientation and 33 cm for position.

In this paper we present a novel approach for image orientation by combining relative rotations and tie points. First, we choose an initial image pair with enough correspondences and large triangulation angle, and we then iteratively add clusters of new images. The rotation of these newly added images is estimated from relative rotations by single rotation averaging. In the next step, a linear equation system is set up for each new image to solve the translation parameters with triangulated tie points which can be viewed in that new image, followed by a resection for refinement. Finally, we optimize the cluster of reconstructed images by local bundle adjustment. We show results of our approach on different benchmark datasets. Furthermore, we orient several larger datasets incl. unordered image datasets to demonstrate the robustness and performance of our approach.

In this paper we deal with the problem of measuring the similarity between training and tests datasets in the context of transfer learning (TL) for image classification. TL tries to transfer knowledge from a source domain, where labelled training samples are abundant but the data may follow a different distribution, to a target domain, where labelled training samples are scarce or even unavailable, assuming that the domains are related. Thus, the requirements w.r.t. the availability of labelled training samples in the target domain are reduced. In particular, if no labelled target data are available, it is inherently difficult to find a robust measure of relatedness between the source and target domains. This is of crucial importance for the performance of TL, because the knowledge transfer between unrelated data may lead to negative transfer, i.e. to a decrease of classification performance after transfer. We address the problem of measuring the relatedness between source and target datasets and investigate three different strategies to predict and, consequently, to avoid negative transfer in this paper. The first strategy is based on circular validation. The second strategy relies on the Maximum Mean Discrepancy (MMD) similarity metric, whereas the third one is an extension of MMD which incorporates the knowledge about the class labels in the source domain. Our method is evaluated using two different benchmark datasets. The experiments highlight the strengths and weaknesses of the investigated methods. We also show that it is possible to reduce the amount of negative transfer using these strategies for a TL method and to generate a consistent performance improvement over the whole dataset.

In this paper we investigate the potential of automatic supervised classification for urban hydrological applications. In particular, we contribute to runoff simulations using hydrodynamic urban drainage models. In order to assess whether the capacity of the sewers is sufficient to avoid surcharge within certain return periods, precipitation is transformed into runoff. The transformation of precipitation into runoff requires knowledge about the proportion of drainage-effective areas and their spatial distribution in the catchment area. Common simulation methods use the coefficient of imperviousness as an important parameter to estimate the overland flow, which subsequently contributes to the pipe flow. The coefficient of imperviousness is the percentage of area covered by impervious surfaces such as roofs or road surfaces. It is still common practice to assign the coefficient of imperviousness for each particular land parcel manually by visual interpretation of aerial images. Based on classification results of these imagery we contribute to an objective automatic determination of the coefficient of imperviousness. In this context we compare two classification techniques: Random Forests (RF) and Conditional Random Fields (CRF). Experimental results performed on an urban test area show good results and confirm that the automated derivation of the coefficient of imperviousness, apart from being more objective and, thus, reproducible, delivers more accurate results than the interactive estimation. We achieve an overall accuracy of about 85 % for both classifiers. The root mean square error of the differences of the coefficient of imperviousness compared to the reference is 4.4 % for the CRF-based classification, and 3.8 % for the RF-based classification.

Self-driving cars and robots that run autonomously over long periods of time need high-precision and up-to-date models of the changing environment. The main challenge for creating long term maps of dynamic environments is to identify changes and adapt the map continuously. Changes can occur abruptly, gradually, or even periodically. In this work, we investigate how dense mapping data of several epochs can be used to identify the temporal behavior of the environment. This approach anticipates possible future scenarios where a large fleet of vehicles is equipped with sensors which continuously capture the environment. This data is then being sent to a cloud based infrastructure, which aligns all datasets geometrically and subsequently runs scene analysis on it, among these being the analysis for temporal changes of the environment. Our experiments are based on a LiDAR mobile mapping dataset which consists of 150 scan strips (a total of about 1 billion points), which were obtained in multiple epochs. Parts of the scene are covered by up to 28 scan strips. The time difference between the first and last epoch is about one year. In order to process the data, the scan strips are aligned using an overall bundle adjustment, which estimates the surface (about one billion surface element unknowns) as well as 270,000 unknowns for the adjustment of the exterior orientation parameters. After this, the surface misalignment is usually below one centimeter. In the next step, we perform a segmentation of the point clouds using a region growing algorithm. The segmented objects and the aligned data are then used to compute an occupancy grid which is filled by tracing each individual LiDAR ray from the scan head to every point of a segment. As a result, we can assess the behavior of each segment in the scene and remove voxels from temporal objects from the global occupancy grid.

A novel calibration process of RICOH-THETA, full-view fisheye camera, is proposed which has numerous applications as a low cost sensor in different disciplines such as photogrammetry, robotic and machine vision and so on. Ricoh Company developed this camera in 2014 that consists of two lenses and is able to capture the whole surrounding environment in one shot. In this research, each lens is calibrated separately and interior/relative orientation parameters (IOPs and ROPs) of the camera are determined on the basis of designed calibration network on the central and side images captured by the aforementioned lenses. Accordingly, designed calibration network is considered as a free distortion grid and applied to the measured control points in the image space as correction terms by means of bilinear interpolation. By performing corresponding corrections, image coordinates are transformed to the unit sphere as an intermediate space between object space and image space in the form of spherical coordinates. Afterwards, IOPs and EOPs of each lens are determined separately through statistical bundle adjustment procedure based on collinearity condition equations. Subsequently, ROPs of two lenses is computed from both EOPs. Our experiments show that by applying 3*3 free distortion grid, image measurements residuals diminish from 1.5 to 0.25 degrees on aforementioned unit sphere.

With rapidly increasing deployment of surveillance cameras, the reliable methods for automatically analyzing the surveillance video and recognizing special events are demanded by different practical applications. This paper proposes a novel effective framework for security event analysis in surveillance videos. First, convolutional neural network (CNN) framework is used to detect objects of interest in the given videos. Second, the owners of the objects are recognized and monitored in real-time as well. If anyone moves any object, this person will be verified whether he/she is its owner. If not, this event will be further analyzed and distinguished between two different scenes: moving the object away or stealing it. To validate the proposed approach, a new video dataset consisting of various scenarios is constructed for more complex tasks. For comparison purpose, the experiments are also carried out on the benchmark databases related to the task on abandoned luggage detection. The experimental results show that the proposed approach outperforms the state-of-the-art methods and effective in recognizing complex security events.

Shabkola is a village located in Mazandaran province of northern Iran that suffers from the mass movement happening in the upstream. Deforestation and changes to land use are the main reasons for the soil instability in this region, which together with steep slope, relatively high precipitation rate and natural erosion has led to such a condition. The area of mass movement is approximately 90 hectares which is a big threat for people living in the region. In this study, we have utilized two different geodetic techniques including InSAR time-series analysis and GPS measurements to assess slope stability in Shabkola. The SAR dataset includes 19 ALOS/PALSAR images spanning from July 2007 to February 2011 while GPS observations are collected in 5 campaigns from September 2011 to May 2014. Displacement as much as approximately 11.7 m in slope direction was detected by GPS observations for the 2011-2014 time period. Most of the slope geometry is in north-south direction, for which the sensitivity of InSAR for displacement detection is low. However, ALOS PALSAR data analysis revealed a previously unknown landslide, covered by dense vegetation in the northern part of main Shabkola landslide, showing line-of-sight velocity of approximately 2cm/year in the time period 2007-2011.

Supervised classification of remotely sensed images is a classical method for change detection. The task requires training data in the form of image data with known class labels, whose manually generation is time-consuming. If the labels are acquired from the outdated map, the classifier must cope with errors in the training data. These errors, referred to as label noise, typically occur in clusters in object space, because they are caused by land cover changes over time. In this paper we adapt a label noise tolerant training technique for classification, so that the fact that changes affect larger clusters of pixels is considered. We also integrate the existing map into an iterative classification procedure to act as a prior in regions which are likely to contain changes. Our experiments are based on three test areas, using real images with simulated existing databases. Our results show that this method helps to distinguish between real changes over time and false detections caused by misclassification and thus improves the accuracy of the classification results.

ISPRS Hannover Workshop on High-Resolution Earth Imaging for Geospatial Information -- JUN 14-17, 2011 -- Hannover, GERMANY, WOS: 000358235900060, In recent years a new generation of high-resolution SAR satellites became operational like the Canadian Radarsat-2, the Italian Cosmo/Skymed, and the German TerraSAR-X systems. The spatial resolution of such devices achieves the meter domain or even below. Key products derived from remote sensing imagery are Digital Elevation Models (DEM). Based on SAR data various techniques can be applied for such purpose, for example, Radargrammetry (i.e., SAR Stereo) and SAR Interferometry (InSAR). In the framework of the ISPRS Working Group VII/2 "SAR Interferometry" a long term scientific project is conducted that aims at the validation of DEM derived from data of modern SAR satellite sensors. In this paper, we present DEM results yield for the city of Barcelona which were generated by means of SAR Interferometry., Int Soc Photogrammetry & Remote Sensing

We propose a new approach for the automatic detection of network structures in raster data. The model for the network structure is represented by a graph whose nodes and edges correspond to junction-points and to connecting line segments, respectively; nodes and edges are further described by certain parameters. We embed this model in the probabilistic framework of marked point processes and determine the most probable configuration of objects by stochastic sampling. That is, different graph configurations are constructed randomly by modifying the graph entity parameters, by adding and removing nodes and edges to/ from the current graph configuration. Each configuration is then evaluated based on the probabilities of the changes and an energy function describing the conformity with a predefined model. By using the Reversible Jump Markov Chain Monte Carlo sampler, a global optimum of the energy function is determined. We apply our method to the detection of river and tidal channel networks in digital terrain models. In comparison to our previous work, we introduce constraints concerning the flow direction of water into the energy function. Our goal is to analyse the influence of different parameter settings on the results of network detection in both, synthetic and real data. Our results show the general potential of our method for the detection of river networks in different types of terrain.

This paper proposes a novel approach for linear feature detection. The contribution is twofold: a novel model for spatial point processes and a new method for linear feature detection. It describes a linear feature as a string of points, represents all features in an image as a configuration of a spatial point process, and formulates feature detection as finding the optimal configuration of a spatial point process. Further, a prior term is proposed to favor straight linear configurations, and a data term is constructed to superpose the points on linear features. The proposed approach extracts straight linear features in a global framework. The paper reports ongoing work. As demonstrated in preliminary experiments, globally optimal linear features can be detected. National Natural Science Foundation of China/41071263 National Natural Science Foundation of China/41571335 Zhejiang Provincial Natural Science Foundation of China/LY13D010003 Key Laboratory for National Geographic Census and Monitoring National Administration of Surveying, Mapping and Geoinformation/2014NGCM

We propose a novel hierarchical approach for the classification of airborne 3D lidar points. Spatial and semantic context is incorporated via a two-layer Conditional Random Field (CRF). The first layer operates on a point level and utilises higher order cliques. Segments are generated from the labelling obtained in this way. They are the entities of the second layer, which incorporates larger scale context. The classification result of the segments is introduced as an energy term for the next iteration of the point-based layer. This framework iterates and mutually propagates context to improve the classification results. Potentially wrong decisions can be revised at later stages. The output is a labelled point cloud as well as segments roughly corresponding to object instances. Moreover, we present two new contextual features for the segment classification: the distance and the orientation of a segment with respect to the closest road. It is shown that the classification benefits from these features. In our experiments the hierarchical framework improve the overall accuracies by 2.3% on a point-based level and by 3.0% on a segment-based level, respectively, compared to a purely point-based classification.

Online multi-person tracking in image sequences is commonly guided by recursive filters, whose predictive models define the expected positions of future states. When a predictive model deviates too much from the true motion of a pedestrian, which is often the case in crowded scenes due to unpredicted accelerations, the data association is prone to fail. In this paper we propose a novel predictive model on the basis of Gaussian Process Regression. The model takes into account the motion of every tracked pedestrian in the scene and the prediction is executed with respect to the velocities of all interrelated persons. As shown by the experiments, the model is capable of yielding more plausible predictions even in the presence of mutual occlusions or missing measurements. The approach is evaluated on a publicly available benchmark and outperforms other state-of-the-art trackers.