Figure - available from: Remote Sensing
This content is subject to copyright.
Transversal section realized on the line S1 with an indication of the size and depth of the buried objects.
Source publication
Detection and monitoring of underwater structures is one of the most challenging applicative scenarios for remote sensing diagnostic techniques, among which ground penetrating radar (GPR). With this aim, an imaging strategy belonging to the family of microwave tomographic approaches is proposed herein. This strategy allows the imaging of objects lo...
Similar publications
Technical infrastructures for water supply since antiquity have long been investigated by engineers, archaeologists, and historians, while hydraulic structures in rivers and brooks, in service before 1700, have hardly received any attention. The urgency for a better understanding of late medieval and early modern hydraulic activity in streams has r...
Citations
... However, influenced by the light refraction, absorption and scattering, underwater images suffer from color distortion, low contrast and blur, which result in unique degradation B Yinghui Zhang zhangyinghui@bnu.edu.cn B Fengxiang Ge ge@bnu.edu.cn 1 trait and cannot meet the expectations of underwater remote sensing techniques [2][3][4]. ...
... We describe the metrics applied in the experiment in Sect. 3 ...
Influenced by light scattering, absorption and water impurities, the quality of underwater image is so poor that significant attention on underwater image enhancement (UIE) has been attracted for producing high-quality visuality. However, both supervised and unsupervised learning-based methods are either more demanding in practical application or easily trap in mapping ambiguity. To solve these issues, we propose a novel unsupervised learning-based underwater image enhancement framework, termed as SFA-GAN. Specifically, we first propose a structure–frequency-aware regularization, which can alleviate both structure and texture loss from spatial and frequency domain, and force SFA-GAN to better enhance underwater images. Afterward, unlike cycle-based underwater image enhancement model with two-sided learning, SFA-GAN adopts one-sided UIE, which is more efficient and faster than the widely used cycle-consistency architectures in training. Extensive experiments on datasets, including: EUVP, UFO-120 and UIEB, demonstrate that the proposed SFA-GAN can achieve state-of-art results on some metrics and produce more clear underwater images without sacrificing model complexity.
... Significant applications with underwater image enhancement (UIE) have been employed in underwater robotic, biological research, surveillance of coral reefs for assistant analysis [1]. However, influenced by the light refraction, absorption and scattering, underwater images suffer from color distortion, low contrast and blur, which result in unique degradation trait and can not meet the expectations of underwater remote sensing techniques [2][3][4]. Thus, it is a challenging work to obtain high quality images captured from underwater. According to the characteristics of the under-water image, existed traditional UIE methods can be divided into two categories: physical model-based methods [5,6], non-physical modelbased methods [7,8]. ...
... Specifically, UCIQE is the evaluation method based on CIELAB space chromaticity, contrast and saturation measurement and can be calculated as: UCIQE = c 1 × σ c + c 2 × con l + c 3 × μ s (10) where is the standard deviation of chromaticity, is brightness contrast; is the average value of saturation, 1 =0.4680, 2 =0.2745, 3 Fig. 3 shows the visual comparison of UIE quality under paired and unpaired settings. We can observe that UGAN, FUnIE-GAN and Fre-GAN generate clear outlines guided by pair-wise enhancement, while suffering from fake back-grounds, as can be seen in Fig. 3(a). ...
Influenced by light scattering, absorption and water impurities, the quality of underwater image is so poor that it poses a great challenge to underwater target detection, marine biological research and marine exploration. Thus, significant attention on underwater image enhancement (UIE) has been attracted for producing high quality visuality as if the underwater image was taken in-air without any structure, texture and color loss. To solve this issue, previous work mainly focus on supervised-learning with large amount of paired data, which is more demanding in practical application. Recent Cycle-GAN based UIE break through the dependence on paired data but easily trap in mapping ambiguity. Essentially, two-sided cycle-consistency is a bijection and only focuses on the pixel level, which is too restrictive and can not accurately express underwater scene structure. Besides, high frequencies in reference images tend to be eschewed by generator, making it difficult to synthesize authentic textures and colors of underwater images. We therefore propose a novel unconstrained UIE framework, structure-frequency-aware generative adversarial network (SFA-GAN), which not only accurately preserves the structure of low quality underwater images, but also captures the high frequencies of the reference images under unconstrained settings. Extensive experiments on datasets EUVP, UFO-120 and UIEB demonstrate that the proposed SFA-GAN can achieves state-of-art results on some metrics and produce more clear underwater images without sacrificing model complexity.
... For high-level computer vision tasks [1], clear underwater images are also required. Furthermore, clear images help the development of undersea remote sensing techniques [2][3][4]. Raw images with low visual quality do not meet these expectations, where the clarity of raw images is degraded by both absorption and scattering [5][6][7]. The clarity of underwater images thus plays an essential role in scientific missions. ...
Underwater image restoration is of significant importance in unveiling the underwater world. Numerous techniques and algorithms have been developed in recent decades. However, due to fundamental difficulties associated with imaging/sensing, lighting, and refractive geometric distortions in capturing clear underwater images, no comprehensive evaluations have been conducted with regard to underwater image restoration. To address this gap, we constructed a large-scale real underwater image dataset, dubbed Heron Island Coral Reef Dataset (`HICRD’), for the purpose of benchmarking existing methods and supporting the development of new deep-learning based methods. We employed an accurate water parameter (diffuse attenuation coefficient) to generate the reference images. There are 2000 reference restored images and 6003 original underwater images in the unpaired training set. Furthermore, we present a novel method for underwater image restoration based on an unsupervised image-to-image translation framework. Our proposed method leveraged contrastive learning and generative adversarial networks to maximize the mutual information between raw and restored images. Extensive experiments with comparisons to recent approaches further demonstrate the superiority of our proposed method. Our code and dataset are both publicly available.
... For these aims, an archaeological site was reconstructed at the Hydrogeosite laboratory [36] with buried full-scale archaeological remains and different water content conditions were reproduced. The archaeological test site was characterized by remains simulating structures of the Lucanian and Roman times (walls, tombs, roads, etc.) covered by sediments [37][38][39][40]. The approach adopted is based on the cooperative use of GPR and electrical resistivity tomography (ERT), adopting 2D and 3D data acquisition strategies. ...
Geophysical techniques are widely applied in the archaeological field to highlight variations of the physical behaviour of the subsoil due to the presence of ancient and buried remains., Considerable efforts are required to understand the complexity of the relationship between archaeological features and their geophysical response where saturated conditions occur. In the case of lacustrine and wetland scenarios, geophysical contrasts or electromagnetic signal attenuation effects drastically reduce the capabilities of the geophysical methodologies for the detection of structures in such conditions. To identify the capability of the electrical and electromagnetic methods in different water-saturated scenarios, an experimental activity was performed at the Hydrogeosite CNR laboratory. The test allowed us to analyze the limits and potentialities of an innovative approach based on the combined use of the ground-penetrating radar and 2D and 3D electrical resistivity tomographies. Results showed the effectiveness of the ground-penetrating radar for detecting archaeological remains also in quasi-saturated and underwater scenarios despite the em signal attenuation phenomena; whilst the results obtained involving the resistivity tomographies offered a new perspective for the archaeological purposes due to the use of the loop–loop shaped array. Moreover, the radar signal attenuation, resolution and depth of investigation do not allow to fully characterize the archaeological site as in the case of the scenarios with a limited geophysical contrast (i.e., water-saturated and arid scenarios). The experimental tests show that these limits can be only partially mitigated through the integration of the geophysical methodologies and further efforts are necessary for improving the results obtainable with an integrated use of the adopted geophysical methodologies.
... techniques [2], [3], [4]. Raw images with low visual quality do not meet these expectations, where the clarity of raw images is degraded by both absorption and scattering [5], [6], [7]. ...
... For conventional underwater image enhancement methods, Histogram-prior [50] is based on a histogram distribution prior; it contains two steps, where the first step is an underwater image dehazing, and the second step is a contrast [48] and LReLU denotes LeakyReLU, we use negative slope = 0.2. For a 2D convolution with input channels = 3, output channels = 64, kernel size = 4, and stride = 2 is referred to as Conv (3,64,4,2). The output of discriminator is a 62 x 62 result metric. ...
Underwater image restoration is of significant importance in unveiling the underwater world. Numerous techniques and algorithms have been developed in the past decades. However, due to fundamental difficulties associated with imaging/sensing, lighting, and refractive geometric distortions, in capturing clear underwater images, no comprehensive evaluations have been conducted of underwater image restoration. To address this gap, we have constructed a large-scale real underwater image dataset, dubbed `HICRD' (Heron Island Coral Reef Dataset), for the purpose of benchmarking existing methods and supporting the development of new deep-learning based methods. We employ accurate water parameter (diffuse attenuation coefficient) in generating reference images. There are 2000 reference restored images and 6003 original underwater images in the unpaired training set. Further, we present a novel method for underwater image restoration based on unsupervised image-to-image translation framework. Our proposed method leveraged contrastive learning and generative adversarial networks to maximize the mutual information between raw and restored images. Extensive experiments with comparisons to recent approaches further demonstrate the superiority of our proposed method. Our code and dataset are publicly available at GitHub.
... Consequently, filtering procedures are necessary for reducing the effects of the undesired signals while improving the targets visibility. The choice of these procedures was driven by the specific surveyed scenario [75,92,94,95]. Figure 10 shows the processing chain adopted to improve the interpretability of the considered GPR data. Specifically, the data processing involves the following procedures: 1. ...
We report a geophysical study across an active normal fault in the Southern Apennines. The surveyed area is the "Il Lago" Plain (Pettoranello del Molise), at the foot of Mt. Patalecchia (Molise Apennines, Southern Italy), a small tectonic basin filled by Holocene deposits located at the NW termination of the major Quaternary Bojano basin structure. This basin, on the NE flank of the Matese Massif, was the epicentral area of the very strong 26 July, 1805, Sant'Anna earthquake (I0 = X MCS, Mw = 6.7). The "Il Lago" Plain is bordered by a portion of the right-stepping normal fault system bounding the whole Bojano Quaternary basin (28 km long). The seismic source responsible for the 1805 earthquake is regarded as one of the most hazardous structures of the Apennines; however , the position of its NW boundary of this seismic source is debated. Geological, geomorpholog-ical and macroseismic data show that some coseismic surface faulting also occurred in correspondence with the border fault of the "Il Lago" Plain. The study of the "Il Lago" Plain subsurface might help to constrain the NW segment boundary of the 1805 seismogenic source, suggesting that it is possibly a capable fault, source for moderate (Mw < 5.5) to strong earthquakes (Mw ≥ 5.5). Therefore, we constrained the geometry of the fault beneath the plain using low-frequency Ground Penetrating Radar (GPR) data supported by seismic tomography. Seismic tomography yielded preliminary information on the subsurface structures and the dielectric permittivity of the subsoil. A set of GPR parallel profiles allowed a quick and high-resolution characterization of the lateral extension of the fault, and of its geometry at depth. The result of our study demonstrates the optimal potential of combined seismic and deep GPR surveys for investigating the geometry of buried active normal faults. Moreover, our study could be used for identifying suitable sites for paleoseismic analyses, where record of earthquake surface faulting might be preserved in Holocene lacustrine sedimentary deposits. The present case demonstrates the possibility to detect with high accuracy the complexity of a fault-zone within a basin, inferred by GPR data, not only in its shallower part, but also down to about 100 m depth. Citation: Nappi, R.; Paoletti, V.; D'Antonio, D.; Soldovieri, F.; Capozzoli, L.; Ludeno, G.; Porfido, S.; Michetti, A.M. Joint Interpretation of Geophysical Results and Geological Observations for Detecting Buried Active Faults: The Case of the "Il Lago" Plain (Pettoranello del Molise, Italy). Remote Sens. 2021, 13, 1555.
... Such a model simplifies the calculation of the kernel of the linear integral equation and provides physical insight into the wave propagation and refraction phenomena. More recently, a further approximate RB model has been proposed by introducing the equivalent dielectric permittivity concept [32]. This latter model allows regarding the wave propagation in the two-layered scenario as occurring into a medium with an "equivalent" and spatially varying dielectric permittivity. ...
... An approximate RB model has been recently proposed to avoid the computational burden of the SD and RB models [32]. This model introduces an equivalent permittivity (EP), such as an equivalent wavenumber, which allows regarding the propagation in the two-layered scenario as occurring into a medium with an equivalent and spatially varying dielectric permittivity. ...
Ground-penetrating radars operating at a stand-off distance from the probed domain are attractive diagnostic tools in several contexts. Despite notable advantages offered in terms of noninvasiveness and flexibility, the usage of stand-off configurations makes the imaging process more difficult. Indeed, an accurate reconstruction of the subsurface/hidden scene requires accounting for the presence of the air-soil interface into the scattering model. This article deals with a comparison of three different linear models for describing the scattering phenomenon under a stand-off configuration, where the presence of the layered medium is explicitly accounted for in the scattering model. The first model is rigorous and based on the spectral domain representation of the incident field and Green's function for the half-space scenario. The second model exploits a ray representation of the wave propagation from the radar to the target and vice versa by accounting for the refraction phenomenon at the interface. The third model is an approximated one, where the layered nature of the scenario is taken into account by means of an equivalent wavenumber, which allows a straightforward evaluation of the incident field and Green's function. The resolution performance of each inversion approach is assessed by means of a numerical analysis. Moreover, data referred to an experiment carried out in controlled conditions are processed and reconstructions are provided to support the comparative analysis among the three imaging approaches.
... Forward modeling of GPR is used to simulate and analyze the electromagnetic response of an underground target by a numerical method, in order to obtain the reflected wave-shape of an underground target on the surface and to understand the propagation laws of the electromagnetic wave in an underground structure [6][7][8]. Moreover, it can deepen the understanding of a GPRmeasured profile, and thus improve the processing and interpretation accuracy of the GPR detection data [9][10][11][12][13]. In addition, forward modeling is still the foundation for determining the inversion of underground target structural parameters based on GPR-measured signals. ...
Accurate forward modeling is of great significance for improving the accuracy and speed of inversion. For forward modeling of large sizes and fine structures, numerical accuracy and computational efficiency are not high, due to the stability conditions and the dense grid number. In this paper, the symplectic partitioned Runge-Kutta (SPRK) method, surface conformal technique, and graphics processor unit (GPU) acceleration technique are combined to establish a precise and efficient numerical model of electromagnetic wave propagation in complex geoelectric structures, with the goal of realizing a refined and efficient calculation of the electromagnetic response of an arbitrarily shaped underground target. The results show that the accuracy and efficiency of ground-penetrating radar (GPR) forward modeling are greatly improved when using our algorithm. This provides a theoretical basis for accurately interpreting GPR detection data and accurate and efficient forward modeling for the next step of inversion imaging.
... GPR allows for the discovery of archaeological features in lacustrine areas; the main problem is related to attenuation, caused by the presence of clay or a high water content, which can significantly reduce the ability of the system to investigate the subsoil at greater depths [14][15]. ...
This paper describes the results obtained using an archaeogeophysical-based approach for discovering new Roman structures belonging to the ancient settlement of Forentum, currently identifiable by a well-preserved sanctuary from the third century BC. The investigated area has been affected by invasive anthropic activities that have partially damaged the Roman structures. Extensive geophysical measurements, including detailed ground-penetrating radar investigations supported by magnetometric data, have allowed for the identification of an impressive complex of structures composed of various buildings. Magnetometric and electromagnetic anomalies suggest the existence of an “urban” dimension close to the Gravetta Sanctuary, totally unearthed and unknown until now, organized into regular patterns in a similar way to the most famous site in the vicinity of Bantia, or the famous Apulian archaeological sites of Ordona and Arpi.
... The use of GPR in archaeology dated back to the 1970s, and more recently (from 1990) it has been quite systematically adopted to detect and map subsurface archaeological artifacts, features and patterning, changes in material properties and voids. Indeed, its success in the archaeological prospection field is showed by the great number of researches where the method is used in order to detect buried structures in urban and rural areas and discover archaeological features in a great number of scenarios, such as to identify ancient settlements (Trinks et al. 2014), locate unearthed burials and ceremonial offering Pipan et al. 2001), reconstruct the history of ancient buildings (Goodman and Piro 2009;Masini et al. 2017), image structures and infrastructure (Malfitana et al. 2015;Florit et al. 2018), and identify sub-water structures (Qin et al. 2018;Ludeno et al. 2018). ...
Even if, in recent decades, the use of remote sensing technologies (from satellite, aerial and ground) for archaeology is stepping into its golden age, in Southern America geophysics for preventive archaeology is more recent and less used than in Europe, Central America and Middle East. In this paper, we provide a brief overview and show the preliminary results obtained from the investigations conducted in Chachabamba (Peru). The archaeological area is located on a strategic terrace overlooking three Inca roads, which served the most important ceremonial centres (including Machu Picchu) of the Urubamba Valley also known as the Sacred Valley. In particular, Chachabamba investigations were conducted with two principal aims: (1) to give new impetus to archaeological research with targeted investigations aimed at improving and completing the site’s knowledge framework; (2) to experiment and validate an archaeogeophysical approach to be reapplied in other sites of the Urubamba valley, including Machu Picchu, having similar characteristics as those found in Chachabamba.
