Gradient Map Research Articles

New Results on Small and Dim Infrared Target Detection

Real-time small infrared (IR) target detection is critical to the performance of the situational awareness system in high-altitude aircraft. However, current IR target detection systems are generally hardware-unfriendly and have difficulty in achieving a robust performance in datasets with clouds occupying a large proportion of the image background. In this paper, we present new results by using an efficient method that extracts the candidate targets in the pre-processing stage and fuses the local scale, blob-based contrast map and gradient map in the detection stage. We also developed mid-wave infrared (MWIR) and long-wave infrared (LWIR) cameras for data collection experiments and algorithm evaluations. Experimental results using both publicly available datasets and image sequences acquired by our cameras clearly demonstrated that the proposed method achieves high detection accuracy with the mean AUC being at least 22.3% higher than comparable methods, and the computational cost beating the other methods by a large margin.

Histograms of oriented mosaic gradients for snapshot spectral image description

This paper presents a feature descriptor using Histogram of Oriented Mosaic Gradient (HOMG) that extracts spatial-spectral features directly from mosaic spectral images. Spectral imaging utilizes unique spectral signatures to distinguish objects of interest in the scene more discriminatively. Snapshot spectral cameras equipped with spectral filter arrays (SFAs) capture spectral videos in real time, making it possible to detect/track fast moving targets based on spectral imaging. How to effectively extract the spatial-spectral feature directly from the mosaic spectral images acquired by snapshot spectral cameras is a core issue for detection/tracking. So far, there is a lack of comprehensive and in-depth research on this issue. To this end, this paper proposed a new spatial-spectral feature extractor for mosaic spectral images. The proposed scheme finds two forms of SFA neighborhood (SFAN) to construct a feature extractor suitable for any SFA structure. Exploiting the spatial-spectral correlation in two SFANs, we design six mosaic spatial-spectral gradient operators to compute spatial-spectral gradient maps (SGMs). HOMG descriptors are constructed using the magnitude and orientation of SGMs. The effectiveness and generalizability of the proposed method have been verified with object tracking experiments. Compared to the state-of-the-art feature descriptors, HOMG ranked first on two datasets captured with snapshot spectral camera with different SFAs, achieving a gain of 3.9% and 5.9% in average success rate over the second-ranked feature.

Geothermal Reservoir Identification in Way Ratai Area Based on Gravity Data Analysis

Gravity research in the Way Ratai geothermal prospect area was conducted to determine geothermal reservoirs, heat sources, and the structure of the geothermal reservoir. The research carried out includes 3D inversion modeling of gravity data. The Bouguer anomaly in the study area has 50 mGal to 120 mGal with low anomalies located in the southeast (Ketang and Kelagian), Northeast (Gedong Air, Sungai Langka, Gunung Betung) areas, and in the Pesawaran mountain area. The high anomaly is in Merawan – Hanuberak – Padang Cermin, Sumbersari and Kaliawi. The horizontal gravity gradient map analysis shows a pattern of fault structure trending northwest-southeast and southwest-northeast, according to the main fault structure in the area. 3D inversion modeling obtains a density distribution between 1.8 g/cc to 3g/cc with a low-density distribution in the south, Mount Pesawaran/Ratai, Gunung Betung, and Sidoharum. The location of the manifestation is 9 km southeast of the Mount Ratai/Pesawaran summit. The existence of geothermal reservoirs is estimated to be in the Lubuk Badak and Miwung Hills areas which are located between the peaks of Mount Ratai/Pesawaran and geothermal manifestations. This is supported by the low-density distribution in the area and the resistivity map from audio-magnetotelluric data.

Resolving Chemical Gradients Around Seagrass Roots—A Review of Available Methods

Steep geochemical gradients surround roots and rhizomes of seagrass and protect the plants against the harsh conditions in anoxic sediment, while enabling nutrient uptake. Imbalance of these gradients, due to e.g., low plant performance and/or changing sediment biogeochemical conditions, can lead to plant stress and large-scale seagrass meadow die-off. Therefore, measuring and mapping the dynamic gradients around seagrass roots and rhizomes is needed to better understand plant responses to human impact and environmental changes. Historically, electrochemical microsensors enabled the first measurements of important chemical species like O2, pH or H2S with high sensitivity and spatial resolution giving important insights to the seagrass rhizosphere microenvironment; however, such measurements only provide information in one dimension at a time. In recent years, the use of reversible optical sensors (in the form of planar optodes or nanoparticles) and accumulative gel sampling methods like Diffusive Gradients in Thin films (DGT) have extended the array of analytes and allowed 2-D mapping of chemical gradients in the seagrass rhizosphere. Here, we review and discuss such microscale methods from a practical angle, discuss their application in seagrass research, and point toward novel experimental approaches to study the (bio)geochemistry around seagrass roots and rhizomes using a combination of available techniques, both in the lab andin situ.

A Splitting Result for Real Submanifolds of a Kähler Manifold

Let (Z,omega ) be a connected Kähler manifold with an holomorphic action of the complex reductive Lie group U^mathbb {C}, where U is a compact connected Lie group acting in a hamiltonian fashion. Let G be a closed compatible Lie group of U^mathbb {C} and let M be a G-invariant connected submanifold of Z. Let xin M. If G is a real form of U^mathbb {C}, we investigate conditions such that Gcdot x compact implies U^mathbb {C}cdot x is compact as well. The vice-versa is also investigated. We also characterize G-invariant real submanifolds such that the norm-square of the gradient map is constant. As an application, we prove a splitting result for real connected submanifolds of (Z,omega ) generalizing a result proved in Gori and Podestà (Ann Global Anal Geom 26: 315–318, 2004), see also Bedulli and Gori (Results Math 47: 194–198, 2005), Biliotti (Bull Belg Math Soc Simon Stevin 16: 107–116 2009).

Road Detection Based on Edge Feature with GAC Model in Aerial Image

Road detection from aerial images still is a challenging task since it is heavily influenced by spectral reflectance, shadows and occlusions. In order to increase the road detection accuracy, a proposed method for road detection by GAC model with edge feature extraction and segmentation is studied in this paper. First, edge feature can be extracted using the proposed gradient magnitude with Canny operator. Then, a reconstructed gradient map is applied in watershed transformation method, which is segmented for the next initial contour. Last, with the combination of edge feature and initial contour, the boundary stopping function is applied in the GAC model. The road boundary result can be accomplished finally. Experimental results show, by comparing with other methods in [Formula: see text]-measure system, that the proposed method can achieve satisfying results.

Reconciling the Infrared and Microwave Observations of the Lunar South Pole: A Study on Subsurface Temperature and Regolith Density

AbstractThe Diviner Lunar Radiometer Experiment on the Lunar Reconnaissance Orbiter and the Microwave Radiometer (MRM) onboard Chang'e‐2 (CE‐2) orbiter performed nearly coincident measurements of the lunar south polar region in 2010–2011. In this study, we reconcile infrared data from Diviner and microwave data from MRM to reveal thermal behavior of lunar regolith at very low temperatures. Assuming a uniform density structure and dielectric properties across the polar region, we retrieve a relative apparent thermal‐gradient map based on radiative transfer model. The result shows apparent thermal gradients in the permanently shadowed regions (PSRs) are much larger than that in non‐PSRs, which implies a systematic difference in thermal conductivity and/or density of lunar regolith in PSRs. We also model surface temperature and microwave brightness temperature over time in PSR and non‐PSR locations. The modeling results at these representative locations and the thermal gradient map confirm that the regolith in the PSRs is consistent with a much lower thermal conductivity and higher porosity than non‐PSRs.

Gradient Guided Pyramidal Convolution Residual Network with Interactive Connections for Pan-sharpening

ABSTRACT Convolutional neural networks (CNNs) have played a predominant role in the field of remote sensing over the last few years. As a significant branch of remote sensing image processing, pan-sharpening technique is to produce a high-resolution multi-spectral (HRMS) image based on a low-resolution multi-spectral (LRMS) image and a high-resolution (HR) panchromatic (PAN) image. Benefiting from the inherently powerful representation ability, deep-learning-based methods have also achieved promising and favourable performance in pan-sharpening community. However, these methods don’t take advantage of the gradient characteristic which contains abundant structure information to guide the pan-sharpening process, failing to achieve the desired spatial preservation. In this paper, we propose a gradient-guided pyramidal convolution residual network with interactive connections (GGPCRN) to relieve the above issue. Specifically, besides the indispensable reconstruction branch, an auxiliary gradient branch providing additional structure information is built to guide the recovery process. Moreover, we introduce pyramidal convolution containing a series of filters with varying depth and size into our network to capture different scales of details for better performance. To further enhance the guidance of gradient maps, two measures are taken. On the one hand, interactive connections are proposed to transfer the mutual effect between the reconstruction branch and gradient branch. On the other hand, we incorporate a mild gradient loss to force a second-order restraint on the pan-sharpened images, making the network concentrate more on structure preservation. Both reduced-resolution and full-resolution experiments suggest that our GGPCRN performs favourably against other methods in terms of quantitative evaluations and visual improvements.

B-AB01-03 HIGH-RESOLUTION REAL-TIME LEFT-VENTRICULAR ENDOCARDIAL ACTIVATION-REPOLARIZATION MAPPING IN AN IN-VIVO CANINE MODEL OF DRUG-INDUCED LONG-QT 1 SYNDROME AND TORSADES DE POINTES

Congenital loss of function or drug-induced inhibition of IKs causes repolarization lability, predisposing to QT prolongation (LQT1) and sympathetically-evoked torsades de pointes (TdP). High-resolution in-vivo mapping of endocardial activation-repolarization times (AT, RT), gradients and focal excitation conspiring to TdP in LQT1 is lacking. To study arrhythmogenic mechanisms of TdP in a canine model of drug-induced LQT1 using real-time and high-resolution electroanatomical mapping. IKs inhibition (HMR1556 0.025-0.05 mg/kg/min) was applied in three fentanyl/etomidate-anesthetized dogs; anesthesia to maintain near-normal autonomic responsiveness. The AcQMap3D noncontact imaging and mapping catheter (Acutus Medical, US) was positioned in the LV. Isoproterenol (1.25-2.5 mcg/kg) induced TdP. AT, RT (Wyatt), RTmax-min (ΔRT), local RT dispersion (per 5.5 mm) and TdP initiation/propagation were studied. IKs block prolonged QTc by 40% (386±22 ms) and mean RT by 44% (309±20 ms), as recorded with the AcQMap3D catheter (AT/RT errors 2.9±4.0/5.4±6.7 ms; cross-correlation 1.0±0.1; time difference 0.8±1.8 ms). ΔRT increased by 23% (85±7 ms) and max local RT dispersion by 19% (80±9 ms). Isoproterenol induced regional RT shortening (base>apex) and beat-to-beat instability, increasing ΔRT significantly by 122% (189±62 ms) and local RT dispersion by 91% (153±51 ms; Figure). Focal triggers of TdP emerged from regions bordering high RT dispersion, driving reentrant excitation. High-resolution mapping in an in-vivo LQT1/TdP model uniquely identifies spatiotemporal RT instabilities, focal excitation and TdP.

Determination of Biodegradation Zone in Central Sumatra Basin

It is commonly known that heavy oil is mostly formed through biodegradation process within reservoir or on the surface both by aerobic and/or anaerobic bacteria that can live under specfi c temperature level(s). In order to investigate heavy oil occurences in Central Sumatra Basin, eff orts have been spent to determine the depths that represent the maximum temperature. By integrating the maximum viable temperature of typical bacteria and temperature gradient data, the depth of heavy oil zone is determined. The work is a combination of establishment of geothermal gradient map and laboratory analysis on fi eld sampled oil for determining types and temperature characteristics of microorganism living in the samples. Heavy oil sampling is made on seepages in areas nearby Minas fi eld. Subsequent laboratory analysis reveals Burkholderia multivorans ATCC BAA-247 as the predominant bacteria having maximum viabl temperature of 60° C. Based on the established geothermal gradient map, this maximum temperature correspond to average depth of 1818 ft (555.5 m). This average depth is used as the lower depth for the biodegradation zone over which investigation over presence of heavy oil bearing reservoirs/traps is made.

Comparison of Terrestrial Gravity and EGM 2008 Data on Extracted Lineaments: A Case Study of the Adamawa Massif, North Cameroon

A structural study on the Adamawa Massif highlights the major structures (faults) of the shear zone as well as the igneous intrusion near Mayo Baleo. Through this study, a comparison on the use of gravity data from the EGM 2008 and terrestrial models was carried out on the basis of gridded gradient data obtained by applying the MAGMAP filtering algorithm on the Oasis Montaj software system. Using the gradients has an advantage in that they contain much information especially on geological structures. The resulting maxima have strong density contrasts, enabling them to be used as tools in explaining geological contacts. Gravimetric lineaments marked by the geometry of the peaks (elongated shape) and the igneous intrusion corresponding to the circular peak were detected on the horizontal gradient maps. Reconciliation with the geology, the slope map and satellite images justifies the presence of the Djohong-Ngaoundal-Dir-Tibati, Tibati-Banyo and Tignere-Banyo lineaments which indicate the underlying Foumban (extension), Tibati and Tchollire-Banyo faults respectively, comprising the shear system residing in the massif underlain by a gneiss-migmatite bedrock. The limitation on the use of gravity data from the EGM 2008 model is manifested through the Djohong lineament correlating with the large mylonitized feature trending NE-SW, defined as a steep relief (27° to 60°), separating the Mbere basin from the upper unit with about 358 m in height difference. This terrain is a small area of relief with morphological units adapted as in the case of the mountainous areas of the West to the use of gravity data from the EGM2008. The other lineaments that are part of the shear zone fault system are not elucidated because they correspond to deep mega structures that the EGM 2008 model resolution cannot attain at depth. The use of this model is suitable for studies of superficial structures which are associated with huge formations such as the volcanic dome of Mayo Baleo, an intrusion of young trachytes with steep slopes (27° to 60°) at an altitude of 2419 m.

Realistic medical image super-resolution with pyramidal feature multi-distillation networks for intelligent healthcare systems.

There are two key requirements for medical lesion image super-resolution reconstruction in intelligent healthcare systems: clarity and reality. Because only clear and real super-resolution medical images can effectively help doctors observe the lesions of the disease. The existing super-resolution methods based on pixel space optimization often lack high-frequency details which result in blurred detail features and unclear visual perception. Also, the super-resolution methods based on feature space optimization usually have artifacts or structural deformation in the generated image. This paper proposes a novel pyramidal feature multi-distillation network for super-resolution reconstruction of medical images in intelligent healthcare systems. Firstly, we design a multi-distillation block that combines pyramidal convolution and shallow residual block. Secondly, we construct a two-branch super-resolution network to optimize the visual perception quality of the super-resolution branch by fusing the information of the gradient map branch. Finally, we combine contextual loss and L1 loss in the gradient map branch to optimize the quality of visual perception and design the information entropy contrast-aware channel attention to give different weights to the feature map. Besides, we use an arbitrary scale upsampler to achieve super-resolution reconstruction at any scale factor. The experimental results show that the proposed super-resolution reconstruction method achieves superior performance compared to other methods in this work.

Exploratory analysis of machine learning methods in predicting subsurface temperature and geothermal gradient of Northeastern United States

Geothermal scientists have used bottom-hole temperature data from extensive oil and gas well datasets to generate heat flow and temperature-at-depth maps to locate potential geothermally active regions. Considering that there are some uncertainties and simplifying assumptions associated with the current state of physics-based models, in this study, the applicability of several machine learning models is evaluated for predicting temperature-at-depth and geothermal gradient parameters. Through our exploratory analysis, it is found that XGBoost and Random Forest result in the highest accuracy for subsurface temperature prediction. Furthermore, we apply our model to regions around the sites to provide 2D continuous temperature maps at three different depths using XGBoost model, which can be used to locate prospective geothermally active regions. We also validate the proposed XGBoost and DNN models using an extra dataset containing measured temperature data along the depth for 58 wells in the state of West Virginia. Accuracy measures show that machine learning models are highly comparable to the physics-based model and can even outperform the thermal conductivity model. Also, a geothermal gradient map is derived for the whole region by fitting linear regression to the XGBoost-predicted temperatures along the depth. Finally, through our analysis, the most favorable geological locations are suggested for potential future geothermal developments.

SmaQ: Smart Quantization for DNN Training by Exploiting Value Clustering

Advancements in modern deep learning have shown that deeper networks with larger datasets can achieve state of the art results in many different tasks. As networks become deeper, the memory requirement of neural network training proves to be the primary bottleneck of single-machine training. In this letter, we first study the characteristics of neural network weight, gradient, feature map, gradient map, and optimizer state distributions for some popular neural network architectures. Our investigation shows that the majority of the data structures used by neural networks can have their value distributions be approximated with normal distributions. We then introduce Smart Quantization (SmaQ), a quantization scheme that exploits this observed normal distribution to quantize the data structures. Our dynamic quantization method calculates the sampled mean and standard deviation of tensors and quantizes each tensor element to 6 or 8 bits based on the z-score of that value. Our scheme reduces the memory usage during training by up to 6.7x with minor losses in accuracy.

Flame analysis using a simple transmission digital holographic interferometer.

A digital holographic interferometer using a collimated beam in transmission mode to illuminate a flow coming from a diffusion flame is presented. The optical system proposes an indirect visualization of the flow to avoid saturation at the sensor. It can detect the intensity signal as a classical schlieren technique and the phase changes due to the presence of the flow. It is possible to retrieve a pseudo-3D flow's view and different gradient maps using the optical phase. According to the knife edge's position, these gradients could be observed in classical schlieren one at a time, but the proposed system could retrieve them all with a single image hologram. As proof of principle, a flame's flow is simultaneously observed with the optical system and a Z-type schlieren set up. A comparison of the visualized flows at different stages of the flame is presented and discussed. A temperature profile is obtained and validated with a thermocouple's point thermal measurements taking the resulting optical phase. Results from both optical techniques show a good agreement.

Small Fractures Caused by the 2019 Ridgecrest Earthquake Sequence: Insights From 3D Coseismic Displacement and Uniaxial Loading Rock Experiments

The moment magnitude (Mw) 6.4 and 7.1 Ridgecrest earthquake sequence that occurred on July 4 and 6, 2019, ruptured a conjugate fault system within the eastern California shear zone. In addition to the ∼50 km surface ruptures, the sequence activated a series of structures with lengths ranging from 1 to 10 km, which are well illuminated by phase gradient maps of Synthetic Aperture Radar (SAR) interferograms. The deformation patterns and mechanisms of these fractures have been well studied, yet the controlling factors of their spatial distribution are less discussed, which are important for understanding how the accumulated strain is releasedviadistributed faulting in the earthquake cycle. Here, we use multi-source SAR images to derive three-dimensional (3D) surface displacement along the main ruptures and the east–west strain across the detected small fractures caused by the 2019 Ridgecrest earthquake sequence. We find that the distribution of these fractures is related to the displacement pattern along the main rupture. Specifically, more fractures appeared in areas with larger slips normal to the main rupture as well as in the junction of the conjugated ruptures. We also conduct uniaxial loading rock experiments to evaluate the strain distribution before the samples were broken. Rock experiments show that rupturing of a conjugated fault system may produce local strain concentration along the main rupture, indicating the important role of the orthogonal faults in generating small fractures with different striking angles and deformation patterns. The 2019 Ridgecrest earthquake sequence exhibits complicated crust behaviors by rupturing an immature fault system, implying that the simple elastic rebound theory may be insufficient to model the coseismic deformation during the earthquake cycle, particularly in the zone with weak crust.

Mapping the human auditory cortex using spectrotemporal receptive fields generated with magnetoencephalography

We present a novel method to map the functional organization of the human auditory cortex noninvasively using magnetoencephalography (MEG). More specifically, this method estimates via reverse correlation the spectrotemporal receptive fields (STRF) in response to a temporally dense pure tone stimulus, from which important spectrotemporal characteristics of neuronal processing can be extracted and mapped back onto the cortex surface. We show that several neuronal populations can be found examining the spectrotemporal characteristics of their STRFs, and demonstrate how these can be used to generate tonotopic gradient maps. In doing so, we show that the spatial resolution of MEG is sufficient to reliably extract important information about the spatial organization of the auditory cortex, while enabling the analysis of complex temporal dynamics of auditory processing such as best temporal modulation rate and response latency given its excellent temporal resolution. Furthermore, because spectrotemporally dense auditory stimuli can be used with MEG, the time required to acquire the necessary data to generate tonotopic maps is significantly less for MEG than for other neuroimaging tools that acquire BOLD-like signals.

Contribution of Gravity and Hydrogeological Implication to the Study of the Subsoil in the East Region (Cameroon)

A geophysical study combining gravimetric and geo-electric methods was carried out in East Cameroon in order to characterize the soil and subsoil of this area. This work constitutes a preliminary study for the evaluation of areas suitable for hydraulic drilling. The aim of this work is to determine the different lineaments that would be found in this area as well as the different layers of land found in this part of Cameroon. The superposition of maxima map of the horizontal gradient reveals the presence of E-W, N-S, ENE-WSW, NNW-SSE, NNE-SSW and NE-SW oriented contacts. Borings curves obtained from the geoelectrical data allowed the geological section of the area to be drawn up, which is characterized by laterites, clays and fissured granites.

Tectonic controls on the salt diapir-related Mississippi Valley-type lead-zinc mineralization of Fej El Adoum ore deposit (Northern Tunisian Atlas): Constrains from detailed gravity and drill hole data

Relationship between salt domes and base metal mineralization is widely recognized across many salt-bearing basins (e.g., US Gulf Coast, Southern Europe and North Africa). In the northern Tunisian Atlas, the Mississippi Valley-type Fej El Adoum lead-zinc deposit represents one of these salt-related ore deposits. It consists of a combination of stratabound, veins, stockworks and cavity filling ore bodies, hosted by dolostone and limestone units, formed along the transition between Triassic evaporites and Cretaceous limestone.In the aim to better understand the concealed complex geological structures across the area, new-acquired detailed gravity data, constrained with field investigation and drill hole data were analyzed. Particularly, edge source detection techniques such as vertical gradient, total horizontal gradient and theta angle map were applied to outline structural contacts related to subsurface density variations. The main detected lineaments are in N–S, E-W and NW-SE directions. The N–S direction represents the deepest and most important lineament in direct contact with the evaporitic Triassic, while the transverse E-W direction is related to dextral strike-slip faults that subdivide the area into blocks. These major deeper lineaments have controlled the growth of the Triassic salt dome mainly during the Albo-Aptian as revealed by pinch outs of these series in direction of the crest of the structure. The detected lineaments seem to play a significant role in: (1) providing favorable pathways for the migration and ascent of mineralized fluids and (2) creating open space filling mainly at the interferences of N–S and E-W trending faults present at the contact between Triassic evaporitic unit and Cretaceous limestone. Our results suggest further reserves mainly in the E-W Ain Said occurrence and West of the Dar N'Hal closed mine.Our study also revealed the importance of structural study in the understanding of complex structures such as highly deformed salt dome cap rocks such as the Fej El Adoum diapiric structure.

Cultivar identification of sweet cherries based on texture parameters determined using image analysis

AbstractThe aim of this study was to develop discriminative models for distinguishing different cultivars of whole sweet cherries based on the texture parameters determined using image analysis. The whole fruit images of “Büttner's Red,” “Kordia,” and “Lapins” were acquired using a digital camera. The discriminative models were built for textures selected from individual color channels R, G, B, L, a, b, X, Y, and Z and color spaces CIE RGB (R—red, B—blue, G—green), CIE Lab (L*—lightness from black to white, a*—green and red, b*—blue and yellow), CIE XYZ (Y—lightness, X and Z—components of color information). The models were developed for texture sets without division and subsets of textures with division into those calculated based on the histogram, co‐occurrence matrix, run‐length matrix, autoregressive model, gradient map. The total accuracy reached 100% for the models built based on sets of textures without division selected from color channels R and X and color spaces RGB, Lab, and XYZ. In the case of divided sets of textures, the correctness of 100% was obtained for textures selected from histogram and co‐occurrence matrix for color space Lab. For color channels, the highest accuracy was equal to 97% for the model built based on the selected textures calculated based on the histogram for color channel L. The significance of this study is great for practical applications. The correct identification of sweet cherry cultivar using the discriminative models can be important for the selection of fruits with the desired properties for consumption and processing.Practical applicationsThe research involved the use of a procedure that allows the classification of sweet cherries in a non‐destructive, objective, and inexpensive manner. The obtained results were very satisfactory and allowed for the complete discrimination of sweet cherry cultivars based on texture parameters of the whole fruit. The developed models may be used in practice for cultivar identification of sweet cherries and detection of falsification or confirmation of authenticity. It will avoid the mixing of different cultivars with different properties intended for use and processing.