Terrain Image Research Articles

Research on Terrain Recognition for Gait Selection of Hexapod Robot

In order to improve the stability of the hexapod robot walking in different terrains, this paper adopts the strategy corresponding to the terrain environment and the motion gait, that is, the robot selects the gait that keeps the robot walk safer and more stable according to the identified terrain. In this paper, using the self-established data set, Terrain6, for terrain classification, the feature extraction of terrain images for hexapod robot is realized firstly based on the Convolutional Neural Networks and the transfer learning. Secondly, according to the stacking fusion method, a terrain recognition model with higher precision is obtained by integrating three terrain classification models include the support vector machine, the naive Bayes and the random forest algorithm. Finally, the experiments show that the hexapod robot selects suitable gait based on the result of terrain recognition to cross complex environment, and the stable and efficient motion of the robot verifies the validity of the research results.

Motion Compensation for Radar Terrain Imaging Based on INS/GPS System

In order to obtain good quality radar terrain images using an aerial-based synthetic aperture radar, a motion compensation procedure must be applied. This procedure can use a precise navigation system in order to determine the aircraft’s position and velocity. A major challenge is to design a motion compensation procedure that can operate in real time, which is crucial to ensure convenient data for a human analyst. The article discusses a possibility of Inertial Measurement System (INS)/Global Positioning System (GPS) navigation system usage in such a radar imaging system. A Kalman filter algorithm designed for this system is described herein, and its modifications introduced by the authors allow the use of navigational data not aligned in time and captured with different frequencies. The presented navigation system was tested using measured data. Radar images obtained with the INS/GPS-based motion compensation system were compared to the INS-only results and images obtained without navigation corrections. The evaluation results presented in the paper show that the INS/GPS system allows for better reduction of geometric distortions in images compared to the INS-based approach, which makes it more suitable for typical applications.

Estimating the Structure of a Discontinuous Layer by Tomographic Methods

The method of constructing a mathematical model of the internal structure of the geological environment is considered for the case, where the function that describes this model has a discontinuity of the first kind. The model is proposed for use in mine seismic tomography. The results of the computing experiment show that even for small orders of Fourier sums, the Fourier coefficients found by means of information about the first times of the arrival of the seismic signal from the sources at the observation points are close to the Fourier coefficients found for the test function, which describes the given terrain image with tectonic damage. The described approaches can be used to improve the mathematical model of the distribution of the slowness of propagation of seismic waves in a given section of the geological environment.

Deep Residual Texture Network for Terrain Recognition

In terrain recognition, the terrain image not only has texture features but also contains spatial features. However, the traditional feature descriptors mainly focus on the texture features of terrain image, ignoring the spatial features of the image, and the convolutional neural network (CNN) can extract the spatial features of the image well due to the role of convolutional layer and the pooling layer. So how to extract these two features at the same time is a challenging problem. In this paper, we introduce a deep residual texture network (DrtNet) that builds a texture detail layer in the residual convolution network and becomes an end-to-end learning network. DrtNet, which simultaneously extracts the spatial geometric features and texture detail features of the terrain image, can successfully combine the traditional texture feature descriptor with the convolutional neural network. The experimental results show that the DrtNet achieves the accuracy of 97.85% on the SDU_Terrain16 dataset that was created by us and outperforms other traditional methods and current popular deep convolutional networks. In addition, DrtNet has achieved good results on two other material/texture datasets (GTOS and DTD).

An Instrument Anomaly in the Mars Exploration Rover Pancam 1,009‐nm Filter (R7): Characterization, Simulation, Correction, and Preliminary Verification

During pre‐flight calibration of the panoramic camera (Pancam) instrument on board the Mars Exploration Rovers MER A (Spirit) and MER B (Opportunity), a discrepancy was noted between 11‐band spectra extracted from Pancam images of the camera's radiometric calibration target and reflectance spectra obtained with a spectrometer. This discrepancy was observed in the longest‐wavelength filter of the camera (the longpass R7 filter with system λeff = 1,009 nm) and consisted of a reduction in contrast between bright and dark regions. Here we describe and characterize this effect. We propose that the effect arises because long‐wavelength photons close to the silicon band‐gap at 1,100 nm are allowed through the R7 filter, pass through the bulk charge‐coupled device, scatter from the backside, pass through the charge‐coupled device again, and are registered in a pixel other than the pixel through which they originally entered. Based on this hypothesis we develop a model capable of accurately simulating the effect, and correct for it. We present preliminary results from testing this correction on preflight, as well as in‐flight, images. The effect is small, but in some specific cases in small regions of high contrast, the effect is significant. In in‐flight images of Martian terrain we observed the signal in dark shadows to be artificially inflated by up to ∼ 33% and analysis of early‐mission calibration target images indicated that the reduced contrast due to the artifact is equivalent to >100 DN (full well = 4095 DN) for a hypothetical perfectly dark pixel.

The case against vast glaciation in Valles Marineris, Mars

The Valles Marineris of Mars form the largest system of interconnected canyons in the Solar System, where morphological, mineralogical, and structural evidence of widespread glaciation has been recently reported. However, neither precipitation models nor global water budgets can account for such a colossal fill, and the hypothesis has thus far remained unchallenged by additional scrutiny. Here, we present the first thorough case against a Valles Marineris glaciation by describing new evidence that precludes the existence of a glacier. Most crucially, we review High Resolution Imaging Science Experiment (HiRISE) images of chaos terrain in Candor Chasma — previously interpreted as remnant glacial ice — and identify layered, boulder-rich scarps bearing no similarity to scarps of massive ice in Promethei Terra or the North Polar Layered Deposits. We also find no significant differences in the structure, morphology, and composition of chaos terrain in Candor and the Chryse Planitia region, which suggests Candor's Chaos, like Chryse's, is fractured highland terrain. We also review several other key supports for the glacial hypothesis, including the coincidence of an apparent glacial trimline with the water-bearing mineral jarosite, and the apparent presence of glacial features including kettle holes and sandur plains. These too are found to have more plausible explanations with non-glacial origins including tectonic reorganisation, groundwater sapping, and chaotic fracturing. We conclude that the Valles Marineris glacial hypothesis is inconsistent with both observed morphology and our understanding of the ancient Martian climate, and we find no evidence to support its existence.

Algorithm of Unmanned Aerial Vehicle’s Positioning Using Sequence of Video Stream Frames

В статье описывается алгоритм, позволяющий осуществлять позиционирование беспилотных летательных аппаратов (БПЛА) по изображениям местности, получаемым с камеры, установленной на борту. Данный алгоритм может быть применен в системах, обеспечивающих навигацию БПЛА в условиях сильных помех сигнала систем глобального позиционирования (GPS). Алгоритм состоит из нескольких этапов: предварительной обработки, выделения опорных точек, сопоставления соседних кадров и определения смещения. На первом этапе каждое изображение подвергается предварительной обработке, заключающейся в уменьшении размера и переводе в режим градаций серого. На втором этапе на изображениях с помощью разности двух сглаживающих фильтров определяются сильно выраженные точки, модуль разности которых превышает определенный порог. Каждая опорная точка описывается типом точки и координатами. Данные точки необходимы для сопоставления соседних кадров. На третьем этапе опорные точки соседних кадров сопоставляются с помощью поиска пары одинакового типа в определенном радиусе. После сопоставления опорных точек происходит соединение всех кадров в одно целое изображение пройденного маршрута. Перед соединением текущий кадр подвергается перспективной трансформации. Для этого среди сопоставленных опорных точек выбираются четыре точки, образующие четырехугольник с максимальной площадью. По координатам этих точек на текущем и предыдущем кадрах вычисляются параметры матрицы трансформации, после чего кадры накладываются друг на друга, образуя изображение пройденного маршрута. По завершении обработки определяется смещение центра масс особых точек текущего кадра относительно предыдущего и вычисляются новые координаты. В заключение приведен пример сформированного изображения пройденного маршрута и проведена оценка эффективности данного алгоритма. Исследование алгоритма проведено на изображениях, полученных с помощью специального программного обеспечения, моделирующего видеопоток с бортовой камеры.

Mars-Relevant Field Experiences in Morocco: The Importance of Spatial Scales and Subsurface Exploration.

During field work at the Ibn Battuta Mars analogue sites, two research questions were analyzed: (1) How do we identify sampling sites using remote and local imaging and (2) what kind of information can be gained from shallow subsurface exploration? While remote images help in targeting field activities in general, the connection between observations at different spatial scales for different rocky desert terrain types is not well established; in this, focused comparison of remote in situ images of well-selected analogues would help a great deal. Dried up lake beds as discerned in remotely acquired data may not show signatures of past water activity, while shallow subsurface exploration could reveal the lacustrine period. Acquisition of several satellite images of the same terrain under different geometries would help to support the planning of such in situ work. The selection of appropriate sampling sites in fluvial settings could be improved by analyzing long, meter-high, open-air outcrops that formed during most recent fluvial episodes. Such settings are abundant on Earth and could be present on Mars but may be just below the resolution of available data. By using 20-30-cm-deep excavations, shallow subsurface exploration could reveal the last period of geological history that would have been unattainable by surface observation alone. Aggregates embedded in the original strata or from heavily pulverized samples could not be identified; only weakly fragmented samples viewed right after acquisition showed aggregates, and thus, the Close-Up Imager (CLUPI) on the ExoMover might provide information on cementation-related aggregation on the observing plate before crushing. The mechanical separation of different size grains (mainly clays and attached minerals) would also support the identification of individual components. To maximize context information during subsurface exploration, rover imaging should be accomplished before crushing; however, currently planned imaging may not be ideal for this.

Geological Analysis of Martian Rover‐Derived Digital Outcrop Models Using the 3‐D Visualization Tool, Planetary Robotics 3‐D Viewer—PRo3D

AbstractPanoramic camera systems on robots exploring the surface of Mars are used to collect images of terrain and rock outcrops which they encounter along their traverse. Image mosaics from these cameras are essential in mapping the surface geology and selecting locations for analysis by other instruments on the rover's payload. 2‐D images do not truly portray the depth of field of features within an image, nor their 3‐D geometry. This paper describes a new 3‐D visualization software tool for geological analysis of Martian rover‐derived Digital Outcrop Models created using photogrammetric processing of stereo‐images using the Planetary Robotics Vision Processing tool developed for 3‐D vision processing of ExoMars PanCam and Mars 2020 Mastcam‐Z data. Digital Outcrop Models are rendered in real time in the Planetary Robotics 3‐D Viewer PRo3D, allowing scientists to roam outcrops as in a terrestrial field campaign. Digitization of point, line, and polyline features is used for measuring the physical dimensions of geological features and communicating interpretations. Dip and strike of bedding and fractures is measured by digitizing a polyline along the contact or fracture trace, through which a best fit plane is plotted. The attitude of this plane is calculated in the software. Here we apply these tools to analysis of sedimentary rock outcrops and quantification of the geometry of fracture systems encountered by the science teams of NASA's Mars Exploration Rover Opportunity and Mars Science Laboratory rover Curiosity. We show the benefits PRo3D allows for visualization and collection of geological interpretations and analyses from rover‐derived stereo‐images.

Combined imaging matching method of side scan sonar images with prior position knowledge

Side scan sonar (SSS) image matching plays an important role in underwater applications, such as combining images to form wide range relief images, underwater simultaneous location and mapping, and construction of images of underwater terrain and objects. Because there are differences between the imaging mechanisms of SSS images and optical images, current matching algorithms for optical images do not always work well for SSS images. This study proposes a combined image matching approach for SSS images. First, the images are preprocessed to remove some environmental effects. Second, feature points that are stable for affine transformations are extracted from the SSS matching images based on speeded-up robust features with prior position knowledge. Third, geometric correction is made by the random sample consensus. After that, a similarity calculation is performed. The proposed approach speeds up the matching process and improves its accuracy by combining feature points matching and similarity calculation. It reduces the mismatching rate and the computation requirement by estimating the location uncertainty with prior knowledge and reducing the searching regions for image matching. Experiments show that the matching algorithm takes less time and has greater reliability and accuracy than traditional algorithms.

Drone images versus terrain images in advertisements: Images’ verticality effects and the mediating role of mental simulation on attitude towards the advertisement

Over the past decade, new techniques have been developed to entice and persuade advertisement viewers. One technique, which has grown in popularity, is the use of drones to capture images within an advertisement. While drones are particularly familiar to military use, new regulations have allowed businesses and amateurs to easily obtain and use drones to create images. With that in mind, this study aims to provide evidence that the verticality of images, i.e., those produced by the drone, can affect information processing of advertising, specifically, angles at which the images are being produced having a greater effect on the viewer than images produced at ground level. We propose a theoretical framework which suggests that mental simulation plays a mediating role between the advertisement watched and Attitude towards the Advertisement (Aad). By creating our own stimuli to eliminate outside discrepancies and focus on images, we were able to carry out an online experiment in which 67 university students participated. Results obtained show how drone images affect Aad as well as the mediating role of mental simulation. Implications for theory and practice are also discussed.

Procedural modeling of rivers from single image toward natural scene production

The rapid and flexible design of natural environments is an important yet challenging task in graphics simulation, virtual reality, and video game productions. This is particularly difficult for natural river modeling due to its complex topology, geometric diversity, and its natural interaction with the complicated terrain. In this paper, we introduce an integrated method for example-based procedural modeling to overcome such difficulties. First, we propose a compact parametric model to represent the certain river, which inherits typical features of natural rivers such as tributary, distributary, tortuosity, possible lakes adjacent to the river. Then, we demonstrate our method for generating 3D river scene solely based on the parametric model. However, choosing appropriate parameters is a tedious undertaking in practice. To further enhance our method's functionality, we rely upon a natural river image to extract meaningful parameters toward the rapid procedural production of the new river scene. Finally, we design a new method to compare two river scenes and iteratively optimize the river network by using the simulated annealing technique. Our method can produce natural river scenes from an example river network and single terrain image with little interaction, and the synthesized scene is visually consistent with the input example in terms of feature similarity. We also demonstrate that our procedural modeling approach is highly automatic toward rapid scene production through various graphics examples.

Application of Discrete Cross-Correlation Function for Observational-Comparative Navigation System

Abstract The article presents navigation system project operating on the principle scene matching area correlation (SMAC), using a digital camera, an MEMS e-compass sensor and an ultrasonic ranging module. Systems of this type are used as a component of advanced integrated navigation systems in view of its autonomy and capability of localizing aircrafts with high accuracy and precision. Steering and display of information are implemented using a computer application designed in Matlab programming environment. The object’s location is fixed, using discrete cross-correlation function through matching of the registered terrain image to digital orthophotomap. The article describes operations directly related to digital image processing, its implementation methods, a structural system design with explanations of each of the functional elements and presents devices used to build a complete integrated measurement unit model. It was used for the effectiveness measurement of determining the location of an object depending on the changes of angle and height of the flight as well as the luminance and noise level in a registered image. The measurements methodology was described which also includes an analysis of the results, an effectiveness evaluation and potential development directions of the designed system.

A Technique of Object Detection Using Information about Changes in the Terrain Image

In this paper, detection of objects in airborne systems of unmanned aerial vehicles using information about changes in the image of the terrain is considered. A proposed technique includes initial detection of objects by the level of local contrast over the field of the image misalignment function, calculation of the amplitude characteristics of objects, and their further filtering by a set of calculated values of characteristics.

Hybrid optical-digital system of texture recognition with liquid crystal input device

The results of experimental research of working hybrid optical-digital image recognition system based on features extracted from Fourier spectra are presented. To form the features an optical Fourier processor with liquid-crystal input device is used. Recognition is carried out with computer (PC) using a neural network algorithm. The possibility of applying the developed system for recognition of the terrain images obtained by aerial photography is shown.

Fully-automated estimation of snow depth in near real time with the use of unmanned aerial vehicles without utilizing ground control points

A new method for on-demand production of a numerical snow depth map, based on observations carried out in near real time by an unmanned aerial vehicle (UAV), is elaborated. The novelty of the approach resides in the neglection of artificial ground control points (GCPs). The method is based on processing oblique aerial geotagged images of snow-covered terrain taken with visible-light (RGB) camera installed on UAV. The compressed photographs (JPG format) are pre-processed with the use of parallel computing in order to resize the images. The structure-from-motion (SfM) procedure – herein based on the RunSFM solution – is utilized to generate a non-georeferenced dense point cloud. The subsequent automatic georeferencing procedure comprises two stages: (1) initial registration of the snow-covered dense point cloud using the Helmert transformation, the parameters of which are estimated from camera positions in local (image) and Universal Transverse Mercator (UTM) coordinate systems, (2) final registration of the snow-covered dense point cloud with the iterative closest point (ICP) algorithm applied in respect to a reference snow-free and highly accurate dense point cloud. The second phase is based on the automated ICP application to two subsets of these dense point clouds which include only tall land cover elements, mainly trees. This enables the ICP-based transformation of the source (with snow) dense point cloud into the reference one (snow-free). The two dense point clouds are subsequently interpolated to digital surface models (DSMs), the difference of which forms a numerical map of the estimated snow depth. The procedure is tested on the RGB images taken by the micro fixed-wing UAV in the Izerskie Mountains in southwestern Poland. In addition, a reproducibility test, which compares the above-mentioned procedure with its replica based on modified settings (NIR camera, lower flight altitude, bigger number of images, higher resolution, SfM reconstruction produced with Theia and OpenMVS), is also performed. The estimation is validated against spatially-distributed snow depth measurements. It is found that the method resolves snow depth with a correct order of magnitude. Mean absolute error (MAE) of snow depth estimation varied between 0.33 and 0.43m, whereas measured snow depths were between 0.24 and 1.06m (mean of 0.41m). Medium level of reproducibility is reported, with considerable similarity in statistics of snow depth estimation and only partial (local) agreement between spatial patterns of estimated and measured snow depth fields. In addition, the results show potential for using NIR images in producing realistic estimated snow depth surfaces.

Topographic correction method for steep mountain terrain images

Topographic correction method for steep mountain terrain images

Object Georeferencing in UAV-Based SAR Terrain Images

Abstract Synthetic aperture radars (SAR) allow to obtain high resolution terrain images comparable with the resolution of optical methods. Radar imaging is independent on the weather conditions and the daylight. The process of analysis of the SAR images consists primarily of identifying of interesting objects. The ability to determine their geographical coordinates can increase usability of the solution from a user point of view. The paper presents a georeferencing method of the radar terrain images. The presented images were obtained from the SAR system installed on board an Unmanned Aerial Vehicle (UAV). The system was developed within a project under acronym WATSAR realized by the Military University of Technology and WB Electronics S.A. The source of the navigation data was an INS/GNSS system integrated by the Kalman filter with a feed-backward correction loop. The paper presents the terrain images obtained during flight tests and results of selected objects georeferencing with an assessment of the accuracy of the method.

Determining shape of a seasonally shadowed asteroid using stellar occultation imaging

A key objective in exploration of small, asteroidal bodies is to determine global shape and volume. The accuracy to which volume can be determined limits determination of bulk density, an important measurement for understanding internal structure. A special case for a rendezvous mission that uses stereo imaging to determine shape is a body with high obliquity encountered near solstice: half of the body is in shadow, and imaging of illuminated terrain alone under-constrains global shape. In this paper we demonstrate the use of stellar occultation imaging to place an upper bound on volume of such a shadowed hemisphere. Thirty-three sets of images of the night side limb of Mercury, acquired by the Mercury Dual Imaging System (MDIS) wide-angle camera (WAC) on MESSENGER, were used to bound the radius of that planet's night side. The maximum radius determined from this limited image set agrees with the actual radius to within ~0.1%. We show, by simulation, expected performance of a campaign of such night side limb images to bound the shape of an irregular, high-obliquity asteroid encountered at solstice. We assumed a body the size and shape of Deimos imaged from a 40-km radius orbit by an imager having specifications of the MDIS/WAC but an updated detector sensitive to mv~10 stars, and a day-side stereo imaging campaign by a well-calibrated camera system. From an equatorial orbit, with one hemisphere in shadow, a campaign of ≥150 night side limb images determines volume of the shadowed hemisphere to ~4 to 6% accuracy. Increasing orbital inclination to improve sampling of high latitudes decreases residuals for the dark hemisphere by 2 to 3%, for the same number of images. A ~2 to 3% uncertainty in global volume – from stereo imaging of illuminated terrain and stellar occultation imaging of shadowed terrain – compares favorably to uncertainty of up to ±25% in the absence of direct measurements of the radius of the shadowed hemisphere.

A combination of k-means clustering and entropy filtering for band selection and classification in hyperspectral images

ABSTRACTHyperspectral images usually have large volumes of data comprising hundreds of spectral bands. Removal of redundant bands can both reduce computational time and improve classification performance. This work proposes and analyses a band-selection method based on the k-means clustering strategy combined with a classification approach using entropy filtering. Experimental results in different terrain images show that our method can significantly reduce the number of bands while maintaining an accurate classification.