Anomalous Structures Research Articles

Cognitive visual anomaly detection with constrained latent representations for industrial inspection robot

With the fast growth of intelligent manufacturing industry, developing advanced industrial inspection robots is becoming a research and application hotspot in the fields of both computer vision and robotics. This kind of industrial inspection robots is expected to automatically detect anomalous structures (e.g., defects, damages, rejects, etc.) from the images of the manufactured products. Generally, the existing visual anomaly detection (VAD) methods mainly focus on modeling the complex and high-dimensional distribution of normal data, while neglecting the specific visual properties of abnormal data since their frequency of occurrence is much less than that of the normal data. In this paper, inspired by the human cognition on extracting abstractly visual properties and to distinguish the anomaly patterns from the observed data, we propose a novel cognitive VAD method for industrial inspection robot. Specifically, we introduce a constrained latent space to mimic the cognitive ability of humans, where the abstraction learned from the observed normal and anomaly data are represented. We build our method based on a convolutional generative adversarial network and a denoising auto-encoder, where the adversarial learning mechanism is adopted to establish the boundary between the normal and anomaly data. In the experiment, we evaluate our method on a real-world dataset where the images are captured for the manufactured products. The comprehensive results comparing with several recent VAD methods show that the proposed method is effective to detect the anomaly images of different categories with a high accuracy.

ICM-BTD: improved classification model for brain tumor diagnosis using discrete wavelet transform-based feature extraction and SVM classifier

In medical image processing, the detection, classification and segmentation of the tumor region from MRI scans accurately are very complicated, significant and time-consuming process. When there is a scenario occurs to handle with large amount of images for tumor diagnosis, there is need of an efficient and adaptive classification model to handle with the anomalous structures of human brains. The MRI brain images show the typical internal brain structure and hence help scholars and medical practitioners in accurate disease diagnosis. With that note, this paper develops a model called improved classification model for brain tumor diagnosis for appropriate classification of tumor images from input MRI images. Initially, filtering techniques are applied for preprocessing the acquired scan images and feature extraction is done with gray-level co-occurrence matrix and discrete wavelet transform equations, which produces more precise results. And, classification is done with the technique called support vector machine, in which the binary classifications are effectively done. The proposed model is evaluated under simulation, and the obtained results outperform the results of traditional brain tumor detection process based on precision, recall and processing time.

Mean plane of the Kuiper belt beyond 50 AU in the presence of Planet 9

Context. A recent observational census of Kuiper belt objects (KBOs) has unveiled anomalous orbital structures. This has led to the hypothesis that an additional ∼5 − 10 m⊕ planet exists. This planet, known as Planet 9, occupies an eccentric and inclined orbit at hundreds of astronomical units. However, the KBOs under consideration have the largest known semimajor axes at a > 250 AU; thus they are very difficult to detect. Aims. In the context of the proposed Planet 9, we aim to measure the mean plane of the Kuiper belt at a > 50 AU. In a comparison of the expected and observed mean planes, some constraints would be put on the mass and orbit of this undiscovered planet. Methods. We adopted and developed the theoretical approach of Volk & Malhotra (2017, AJ, 154, 62) to the relative angle δ between the expected mean plane of the Kuiper belt and the invariable plane determined by the eight known planets. Numerical simulations were constructed to validate our theoretical approach. Then similar to Volk & Malhotra (2017, AJ, 154, 62), we derived the angle δ for the real observed KBOs with 100 < a < 200 AU, and the measurement uncertainties were also estimated. Finally, for comparison, maps of the theoretically expected δ were created for different combinations of possible Planet 9 parameters. Results. The expected mean plane of the Kuiper belt nearly coincides with the said invariable plane interior to a = 90 AU. But these two planes deviate noticeably from each other at a > 100 AU owing to the presence of Planet 9 because the relative angle δ could be as large as ∼10°. Using the 1σ upper limit of δ < 5° deduced from real KBO samples as a constraint, we present the most probable parameters of Planet 9: for mass m9 = 10 m⊕, orbits with inclinations i9 = 30°, 20°, and 15° should have semimajor axes a9 > 530 AU, 450 AU, and 400 AU, respectively; for m9 = 5 m⊕, the orbit is i9 = 30° and a9 > 440 AU, or i9 < 20° and a9 > 400 AU. In this work, the minimum a9 increases with the eccentricity e9 (∈[0.2, 0.6]) but not significantly.

Microstructural refinement and anomalous eutectic structure induced by containerless solidification for high-entropy Fe–Co–Ni–Si–B alloys

The mechanical properties of high-entropy alloys (HEAs) strongly depend on their microstructural features. In this work, the phase formation and morphologies of Fe26.7Co26.7Ni26.7Si8.9B11.0 HEAs fabricated under different undercooling levels by the drop tube technique together with subsequent quenching were investigated in detail. It was found when the droplet size decreases (i.e. the increase of the undercooling level), the grain refinement is easily achieved without any phase transformation for the present hypoeutectic HEA which is composed of the fcc dendrites and eutectic structure. After quenching under large undercooling levels, the liquid should be deeply undercooled into an asymmetric couple zone, leading to the primary formation of eutectic structure and the subsequent precipitation of the fcc dendrites. The direct product of rapid solidification after deep undercooling and subsequent recalescence may induce the decoupled eutectic growth, resulting in a transition from the regular into anomalous eutectic structures. Furthermore, the volume fractions of the eutectic structures and the fcc dendrites increase and decrease with decreasing droplet size, respectively. As a result, the Vickers hardness gradually increases with decreasing droplet size, which exhibits a linear relationship with the volume fraction of eutectic matrix. The present study may provide a simple way to tailor microstructures and mechanical properties of HEAs.

Creating Computed Tomography Angiography Stereoscopic Models to Enhance the Learning of the Hepatobiliary Region

Current anatomical education continues to increase the use of three‐dimensional (3D) models rendered from radiographic imaging to assist student’s anatomical spatial awareness. Such orientation is valuable and clinically essential, especially when such knowledge is used in conjunction with diagnostic imaging modalities that can further guide treatment plans. Therefore, in programs that rely on cadaveric dissection, it is crucial to efficiently orient novice students to utilize imaging modalities early in their educational careers so that they can recognize any pathological or anomalous structures that other educational 3‐D resources may have limitations in displaying. Some 3‐D rendered models obtained from computed tomography angiography (CTA) have already been clinically derived to facilitate in orienting students to the intracranial neurovascular anatomy. However, there are a limited amount of CTA derived 3‐D models to aid in the study of other anatomical regions, in particular, the hepatobiliary region of the abdomen. In an effort to enhance the continued education of graduate and medical professional students, we have developed a 3D stereoscopic model rendered from CTA‐derived data that are consistent with the pancreatic hepatobiliary region with accompanying vasculature and associated gastrointestinal (GI) tract. Although 3‐D stereoscopic models usually use voxel‐based CTA data, our model uses semi‐auto combined rendering with complementary volume and surface rendering methods. Our ultimate goal is to utilize this virtual model presentation to help students clearly identify anatomical structures, both on cadaveric specimens and through imaging modalities. Future studies will consist of comparing students who are undertaking a gross anatomy course to students who are enrolled in the course but who also view the virtual 3D hepatobiliary model presentation to further support their anatomical education.

RCAnalyzer: visual analytics of rare categories in dynamic networks

A dynamic network refers to a graph structure whose nodes and/or links dynamically change over time. Existing visualization and analysis techniques focus mainly on summarizing and revealing the primary evolution patterns of the network structure. Little work focuses on detecting anomalous changing patterns in the dynamic network, the rare occurrence of which could damage the development of the entire structure. In this study, we introduce the first visual analysis system RCAnalyzer designed for detecting rare changes of sub-structures in a dynamic network. The proposed system employs a rare category detection algorithm to identify anomalous changing structures and visualize them in the context to help oracles examine the analysis results and label the data. In particular, a novel visualization is introduced, which represents the snapshots of a dynamic network in a series of connected triangular matrices. Hierarchical clustering and optimal tree cut are performed on each matrix to illustrate the detected rare change of nodes and links in the context of their surrounding structures. We evaluate our technique via a case study and a user study. The evaluation results verify the effectiveness of our system.

Multiple origins of anomalous eutectic microstructure in rapidly solidified Mg-Al alloy

Mg-33.3 wt.%Al eutectic alloys were fabricated by melt spinning at wheel speeds from 5 to 50 ms−1. SEM and S/TEM analyses showed the formation of anomalous eutectic structures at all wheel speeds. Four distinctive eutectic microstructures were observed, depending on the wheel speed, each with a unique formation mechanism. At low wheel speeds, solute trapping and high interfacial energy resulted in the remelting of primarily grown regular eutectic microstructures and was responsible for the formation of the anomalous eutectic Type A. A coupled-to-decoupled-growth transition occurred at intermediate wheel speeds giving rise to the anomalous eutectic Type B. The decoupled nucleation and growth of the eutectic phases followed by a coupled dendritic growth lead to the formation of flower-like microstructures at anomalous eutectic Type C. At sufficiently high wheel speed, no sign of regular eutectic was found and a unique cellular microstructure of ϒ phase along with inter- or intragranular α phase grains was formed, anomalous eutectic Type D, as a result of the decoupled nucleation and growth of the eutectic phases. The observed microstructures were examined thoroughly, and the correlated formation mechanisms were discussed on the basis of the SEM images and S/TEM and EDS analysis.

Magnon polarons in the spin Peltier effect

We report the observation of anomalous peak structures induced by hybridized magnon-phonon excitation (magnon polarons) in the magnetic field dependence of the spin Peltier effect (SPE) in a Lu$_{2}$Bi$_{1}$Fe$_{4}$Ga$_{1}$O$_{12}$ (BiGa:LuIG) with Pt contact. The SPE peaks coincide with magnetic fields tuned to the threshold of magnon-polaron formation, consistent with the previous observation in the spin Seebeck effect. The enhancement of SPE is attributed to the lifetime increase in spin current caused by magnon-phonon hybridization in BiGa:LuIG.

Quantifying Gerrymandering in North Carolina

ABSTRACTBy comparing a specific redistricting plan to an ensemble of plans, we evaluate whether the plan translates individual votes to election outcomes in an unbiased fashion. Explicitly, we evaluate if a given redistricting plan exhibits extreme statistical properties compared to an ensemble of nonpartisan plans satisfying all legal criteria. Thus, we capture how unbiased redistricting plans interpret individual votes via a state’s geo-political landscape. We generate the ensemble of plans through a Markov chain Monte Carlo algorithm coupled with simulated annealing based on a reference distribution that does not include partisan criteria. Using the ensemble and historical voting data, we create a null hypothesis for various election results, free from partisanship, accounting for the state’s geo-politics. We showcase our methods on two recent congressional districting plans of NC, along with a plan drawn by a bipartisan panel of retired judges. We find the enacted plans are extreme outliers whereas the bipartisan judges’ plan does not give rise to extreme partisan outcomes. Equally important, we illuminate anomalous structures in the plans of interest by developing graphical representations which help identify and understand instances of cracking and packing associated with gerrymandering. These methods were successfully used in recent court cases. Supplementary materials for this article are available online.

Post mortem coronary angiography in a preemie heart \u2013 a case report

BackgroundPostmortem coronary angiography has been used in forensic medicine for several decades but its use has never been documented in neonatal hearts. The objective of this case is to report the use of postmortem coronary angiography as a diagnostic modality for neonates suspected to have complex congenital heart anomalies.Case presentationA 36-week-old female infant required extracorporeal membranous oxygenation for persistent hypotension on day 1 of life. A congenital echocardiogram (ECHO) on day 3 of life revealed multiple anomalous vascular structures within the interventricular septum. The infant passed away on day 4 of life after the parents elected to withdraw support. A consent for autopsy was taken from the parents and a postmortem coronary angiography was performed. The coronary vessels were injected with Iodixanol contrast via a 24 G angiocath under fluoroscopy. The anomalous septal vessels were identified as dilated coronary artery and vein. No other anomalies were identified.ConclusionPostmortem coronary angiography complements other imaging procedures in understanding the nature of some complex congenital heart defects and in determining the cause of death in such neonates.

Diversion of the paleo‐Yellow River channel in the Qingtongxia area of Ningxia, China: Evidence from terraces and fluvial landforms

The Qingtongxia Grand Canyon (QGC) of the Yellow River is a region of intense tectonic deformation that is located in the southern Yinchuan Basin, at the junction of the western margin of the Ordos Plateau and the northeast arcuate structural belt of the Qinghai–Tibetan Plateau. The Yellow River makes a 90° turn as it traverses the Qingtongxia area, incising the hard Ordovician sandstones of Niushou Mountain, while leaving the relatively soft Quaternary sediments on the northern side of the channel undisturbed. Despite this apparent inconsistency with the expected pattern of river erosion, there has been no significant research to date on the formation of the QGC. Here, we utilize remote sensing, surficial geomorphology, and shallow coring to confirm the evolution of the Yellow River channel and formation of the QGC. Using Landsat Thermic Mapper and ASTER imagery, we identified a N–S‐oriented zone of high water‐content in the northern part of the QGC that is characterized at the surface by marsh and wetlands. Shallow cores reveal the zone is underlain by Yellow River gravel, while seismic profiling confirms anomalous seismic structures relative to the surrounding strata. Together, these data document a paleo‐Yellow River channel in the northern Qingtongxia area. We selected four representative profiles of the Yellow River terraces in the Qingtongxia and adjacent areas to acquire measurements on above water level and age. The uppermost terraces preserved at the entrance, central section, and exit of the QGC are labelled T6, T9, and T3, respectively. Our data indicate that the flow direction during the construction of T9 was different to that of the modern Yellow River and that canyon incision occurred during the development of T3, which we constrain to 65–85 ka. Ultimately, we (a) document a paleo‐Yellow River channel north of the QGC, which may have been affected by tectonic activity since the late Pleistocene and (b) show that the Yellow River has since been diverted towards the Yinchuan Basin via the QGC.

Comparison of the 2‐D and Quasi‐3‐D Geoelectric Models of the Ukrainian Eastern Carpathians and Their Link to the Tectonic Structure

AbstractWe present results of a study of the electrical conductivity distribution beneath the Ukrainian Eastern Carpathians where the Carpathian arc turns to the southeast and reaches the maximum curvature further in the Romanian territory. We have used the magnetotelluric and magnetovariational data in the period range of 1–6,000 s collected by geoelectric groups from several countries during deep induction experiments, which covered the studied area by a number of field stations clustered along profiles crossing the Carpathian arc. In this work, we compare the 2‐D and quasi‐3‐D thin sheet inverse models, which allow us to constrain both the geometry and electrical properties of the anomalous conductivity structures beneath the Carpathians. Comparison of the inverse results with the Carpathians tectonics indicates that the Carpathian conductivity anomaly within the Polish and Slovak territory is located approximately between the Magura nappe front and the Pieniny Klippen Belt. In Ukraine, where the Magura nappe narrows and disappears, the anomaly reaches the front of the Outer Carpathians. The resulting quasi‐3‐D model of the conductivity structure beneath the Carpathians is compared with other geophysical and geological data within the region. The relation of the position of the conductors to the heat flow models and to location of the earthquakes sources indicates possible geological mechanisms, such as saline pore fluids, thin films of grain boundary graphite, or partial melting, connected with recent activation zones.

Importance of later phases in seismic tomography

Seismic tomography is a powerful tool for mapping the three-dimensional structure of the Earth's interior. Tomographic images obtained in the past four decades have greatly improved our understanding of the Earth's heterogeneous structure and dynamics, which signify a revolution in Earth sciences. Most of the tomographic models are determined using the first P and S wave data generated by local earthquakes and/or teleseismic events. Five global seismic discontinuities exist in the Earth, including the Moho, the 410 km and 660 km discontinuities, the core-mantle boundary, and the inner-core boundary. In addition, local-scale seismic discontinuities are also revealed, particularly in subduction zones. These sharp discontinuities generate abundant reflected and converted seismic waves, the so-called later phases, which are identified in observed seismograms. In this article, I review the tomographic studies in the past three decades that made use of the later phase data. Because later phases have ray paths different from those of the first P and S waves, they illuminate the Earth's interior structure that is not well sampled by the first waves. Hence, the use of later phases in tomographic imaging has led to new discoveries of anomalous structures and geodynamic processes at different spatial scales, which shed new light on seismotectonics, magmatism and mantle dynamics. These practices indicate that the later phases are very important in seismic tomography, and so they should be collected from seismograms with a greater quantity and quality so as to obtain better tomographic images of the Earth's interior.

Morphological alterations caused by manual venom extraction on the main venom gland of Bothrops asper and Crotalus simus snakes (Serpentes: Viperidae): Long-term implications for antivenom production

The only scientifically validated treatment for snakebite envenomation is the administration of antivenoms. For their production, small quantities of snake venom are injected in animals to elicit a specific antibody response. Snakes are kept in captivity, and their venom is regularly extracted to assure antivenom access. It has already been reported that the pressure exerted upon the venom gland during this extraction can cause tissue damage and fibrosis, leading to a decrease in the venom yield. We described the histopathology of venom glands for B. asper and C. simus snakes used for antivenom production. Based on these reported tissue abnormalities, we quantify the tissue injury by a generated damage-SCORE and fibrosis. A variety of histopathological damages were found such as fibrosis, edema, necrosis, hemorrhage, and formation of anomalous structures, especially in C. simus, which is more prone to suffer severe damage. The level and severity of the damage depend on the frequency and the number of venom extractions. Furthermore, we design an experimental intensive venom extraction scheme with which we could confirm the causality of these effects. In addition to the histopathological damages, the LD50 and biochemical venom composition were also affected giving experimental evidence that the venom extraction not only causes tissue damage but also affects the composition stability and toxicity of the venom. In order to produce quality and effective antivenoms, an improvement of the management of snake collections could be established, such as rotation groups to assure the quality of the venom yielded.

Atypical Intracerebral Developmental Venous Anomalies in Sturge-Weber Syndrome: A Case Series and Review of Literature

BackgroundIntracranial vascular abnormalities in Sturge-Weber syndrome, including leptomeningeal angiomatosis, anomalous cortical venous structures, and transmedullary developmental venous anomalies, are well recognized. Prominent vascular flow voids on T2-weighted magnetic resonance imaging (MRI) are occasionally identified in patients with Sturge-Weber syndrome, raising concern of arteriovenous malformations, a congenital high-flow vascular malformation with a risk of bleeding. MethodsWe report four patients with prominent flow voids on conventional MRI that suggested high-flow lesions. ResultsDiagnostic evaluation was performed with cerebral angiography in one patient and with a combination of magnetic resonance angiography and magnetic resonance venography in three patients. In all four patients, the conventional MRI-identified lesions represented prominent developmental venous anomalies and not arteriovenous malformations. ConclusionsThis series highlights that developmental venous anomalies may appear in individuals with Sturge-Weber syndrome as unusually large and seemingly high-flow lesions on MRI. Noninvasive imaging with magnetic resonance angiography and magnetic resonance venography can be used in the management of such patients for further characterization of these vascular structures.

An integrated geophysical study east of the southern Chagos Laccadive Ridge Complex, Central Indian Ocean Basin: Implications for the Rodriguez Triple Junction dynamics in the Late Cretaceous

The Central Indian Ocean Basin, east of the southern Chagos Laccadive Ridge complex has evolved due to seafloor spreading in the N-S direction since the Late Cretaceous. The trace of the Rodriguez Triple Junction (RTJ) has been inferred in this basin, but its exact location and dynamics prior to C29o are not known. About 20,000 km of total magnetic intensity data along with satellite gravity mosaic were used to understand the dynamics of the RTJ in the Late Cretaceous.Synthetic seafloor spreading models endorse the presence of Late Cretaceous to early Palaeogene magnetic anomalies C34y to C24o in the study area. Half-spreading rates are initially low (32 mm/yr) between C34y to C33y, and increase up to 78 mm/yr by C28y. Thereafter a decreasing trend is seen (57 mm/yr). Towards the west in the study area, higher spreading rates (120 mm/yr) between C33y and C32ay, and distorted anomalies between chrons C32ay and C30y are observed. Detailed geophysical analysis suggests that the higher spreading rates are due to the presence of transferred crust, and the distorted anomalies are caused by the presence of fragmented tectonic blocks depicting microplate formation and rotation.Plate reconstruction models suggest that the RTJ at C34y time lay southwest of Comorin Ridge which coincided with the northern edge of the Conrad Rise. Until C33y, the RTJ left a linear trace parallel to the fracture zone trend on both the Indian and Antarctic plates, while it gradually moved eastwards on the African plate. The RTJ migrated westwards until C31y causing the transfer of a considerable amount of crust from the African plate to both the Indian and Antarctic plates. Thereafter, the RTJ rapidly reversed its migration direction towards the east until C29o causing the formation and rotation of a microplate on the Indian plate and lengthening of the Southwest Indian Ridge.The present study suggests that this episodic migration of the RTJ may be due to ridge propagation under the influence of the Marion hotspot which was in the vicinity during that period. This RTJ migration resulted in microplate formation, lithospheric transfer and thickened lithosphere similar to the inferences made at many triple junctions in the Pacific Ocean. Thus the origin of anomalous structures such as the Crozet Bank may be related to triple junction dynamics under the influence of Marion hotspot. Further, the counter clockwise rotation of the Southwest Indian Ridge between C33y and C29y may have a causal link with the formation and rotation of the microplate, which occurred during the same period. The present study thus facilitated the inference of the dynamics of the RTJ trace between C34y and C29o on the three plates.

Self-Organization, Structures, and Anomalous Transport in Turbulent Partially Magnetized Plasmas with Crossed Electric and Magnetic Fields.

Self-organization and anomalous transport in gradient-drift driven turbulence in partially magnetized plasmas with crossed electric and magnetic fields is demonstrated in two-dimensional fluid simulations. The development of large scale structures and flows is shown to occur as a result of the inverse energy cascade from short wavelength instabilities. The turbulence shows complex interaction of small scale modes with large scale zonal flow modes, vortices, and streamers resulting in strongly intermittent anomalous transport that significantly exceeds the classical collisional values. The turbulence driven secondary instabilities and large scale structures are shown to dominate the anomalous electron current. Such anomalous transport and structures are consistent with a number of experimental observations in laboratory plasmas.

The first evaluation of the dynamic hydration number of hydrated ions confined in mesoporous silica MCM-41

Ion hydration in narrow pores is important in engineering, biology, and chemistry. The structures of hydrated ions confined in nanospaces have been studied by X-ray diffraction and X-ray absorption spectroscopy, and these confined hydrated ions have been found to have anomalous structures. On the other hand, the dynamic behavior has been studied only by molecular dynamics (MD) simulations, which have revealed the unique dynamics of hydration in nanospaces. However, no experimental studies of the dynamics of confined hydrated ions have been carried out. Here, we focus on the dynamic hydration number (nDHN) used in bulk solutions to evaluate the dynamic behavior of electrolyte solutions confined in nanospaces. To evaluate nDHN in nanospaces, we derived the molality dependence of the 2H spin–lattice relaxation time of D2O confined in nanospaces by considering the contribution of water bound to the pore surface. Next, we applied this to electrolyte solutions confined in mesoporous silica MCM-41 and evaluated nDHN. The absolute value of the dynamic hydration number in mesopores is significantly larger than that in the bulk solution; thus, the confinement effect drastically enhances the dynamic hydration properties (positive and negative hydration). This result supports the compressed hydration structure of ions with positive hydration character reported by previous structural and simulation studies. Our new approach for the evaluation of nDHN in nanospaces can be expected to give new insights in the investigation of electrolyte solutions confined in nanospaces from the viewpoint of the experimental verification of dynamic features of the hydration structure.

Widespread seismic anisotropy in Earth’s lowermost mantle beneath the Atlantic and Siberia

Deep inside the Earth, just above the core-mantle boundary at around 2,700 km depth, large-scale mantle structures are assumed to play a key role for global geodynamic processes. While unusual hot regions are attributed to feed rising mantle plumes and volcanic hotspots, the accumulation of subducted lithospheric plates is associated with colder than average features. In both environments the appearance of dynamic-driven processes such as deformation and mantle flow can directly be inferred by the presence of seismic anisotropy. However, the geometries as well as the interactions of these massive anomalous structures with the surrounding mantle material are still under debate. Based on new seismic data from a dense and large-aperture recording network in Scandinavia we characterize the anisotropic signatures of two so far unexplored regions in the lowermost mantle by using observations of clearly discrepant SKS-SKKS shear wave splitting measurements. Thereby we can demonstrate that anisotropy is located along the northern edges of the Large Low Shear Velocity Province beneath Africa. Furthermore, we recover an anisotropic structure in a region of fast seismic velocity underneath Siberia which provides additional evidence for widespread deformation caused by a deeply subducted slab.

How to Calculate Bouguer Gravity Data in Planetary Studies

In terrestrial studies, Bouguer gravity data is routinely computed by adopting various numerical schemes, starting from the most basic concept of approximating the actual topography by an infinite Bouguer plate, through adding a planar terrain correction to account for a local/regional terrain geometry, to more advanced schemes that involve the computation of the topographic gravity correction by taking into consideration a gravitational contribution of the whole topography while adopting a spherical (or ellipsoidal) approximation. Moreover, the topographic density information has significantly improved the gravity forward modeling and interpretations, especially in polar regions (by accounting for a density contrast of polar glaciers) and in regions characterized by a complex geological structure. Whereas in geodetic studies (such as a gravimetric geoid modeling) only the gravitational contribution of topographic masses above the geoid is computed and subsequently removed from observed (free-air) gravity data, geophysical studies focusing on interpreting the Earth’s inner structure usually require the application of additional stripping gravity corrections that account for known anomalous density structures in order to reveal an unknown (and sought) density structure or density interface. In planetary studies, numerical schemes applied to compile Bouguer gravity maps might differ from terrestrial studies due to two reasons. While in terrestrial studies the topography is defined by physical heights above the geoid surface (i.e., the geoid-referenced topography), in planetary studies the topography is commonly described by geometric heights above the geometric reference surface (i.e., the geometric-referenced topography). Moreover, large parts of a planetary surface have negative heights. This obviously has implications on the computation of the topographic gravity correction and consequently Bouguer gravity data because in this case the application of this correction not only removes the gravitational contribution of a topographic mass surplus, but also compensates for a topographic mass deficit. In this study, we examine numerically possible options of computing the topographic gravity correction and consequently the Bouguer gravity data in planetary applications. In agreement with a theoretical definition of the Bouguer gravity correction, the Bouguer gravity maps compiled based on adopting the geoid-referenced topography are the most relevant. In this case, the application of the topographic gravity correction removes only the gravitational contribution of the topography. Alternative options based on using geometric heights, on the other hand, subtract an additional gravitational signal, spatially closely correlated with the geoidal undulations, that is often attributed to deep mantle density heterogeneities, mantle plumes or other phenomena that are not directly related to a topographic density distribution.