1,721,320 research outputs found
Landslide detection by deep learning of non-nadiral and crowdsourced optical images
Full text linkThe recent development of mobile surveying platforms and crowdsourced geoinformation has produced a huge amount of non-validated data that are now available for research and application. In the field of risk analysis, with particular reference to landslide hazard, images generated by autonomous platforms (such as UAVs, ground-based acquisition systems, satellite sensors) and pictures obtained from web data mining are easily gathered and contribute to the fast surge in the amount of non-organized information that may engulf data storage facilities. Therefore, the high potential impact of such methods is severely reduced by the need of a massive amount of human intelligence tasks (HITs), which is necessary to filter and classify the data, whatever the final purpose. In this work, we present a new set of convolutional neural networks (CNNs) specifically designed for the automated recognition of landslides and mass movements in non-standard pictures that can be used in automated image classification, in supporting UAV autonomous guidance and in the filtering of data-mined information. Computer vision can be of great help in fostering the autonomous capability of intelligent systems to complement, or completely substitute, HITs. Image and object recognition are at the forefront of this research field. The deep learning procedure has been accomplished by applying transfer learning to some of the top-performer CNNs available in the literature. Results show that the deep learning machines, calibrated on a relevant dataset of validated images of landforms, may supply reliable predictions with computational time and resource requirements compatible with most of the UAV platforms and web data mining applications in landslide hazard studies. Average accuracy achieved by the proposed methods ranges between 87 and 90% and is consistently higher than that obtained by general-purpose state-of-the-art image recognition convolutional neural networks. The method can be applied to early warning, vulnerability assessment, residual risk estimation, model parameterisation and landslide mapping. Specific advantages will be the reduction of the present limitations in the intelligent guidance of landslide mapping drones, the classification of fake news, the validation of post-disaster information and the correct interpretation of an impending change in the environment
Knee Surgery using Computer Assisted Surgery and Robotics
No full textThis book discusses the full range of current applications of computer-assisted surgery and robotics in the field of knee surgery, and also considers potential future applications. The impact of computer-assisted surgery on a wide range of surgical procedures is clearly explained. Procedures considered include total knee arthroplasty, unicompartmental knee arthroplasty, cruciate ligament reconstruction, patellofemoral arthroplasty, and revision surgery. In each case, technical aspects are thoroughly addressed in a readily understandable manner. Knee Surgery Using Computer-Assisted Surgery and Robotics will be an ideal guide to this exciting field for both novice and more experienced surgeons who treat knee injuries and disorders
La deformazione gravitativa profonda del Monte Penna (La Verna): primi risultati dell'applicazione di un sistema di monitoraggio mediante GPS
No full text
Fusion of GNSS and Satellite Radar Interferometry: Determination of 3D Fine-Scale Map of Present-Day Surface Displacements in Italy as Expressions of Geodynamic Processes
Full text linkWe present a detailed map of ground movement in Italy derived from the combination of the Global Navigation Satellite System (GNSS) and Satellite Synthetic Aperture Radar (SAR) interferometry. These techniques are two of the most used space geodetic techniques to study Earth surface deformation. The above techniques provide displacements with respect to different components of the ground point position; GNSSs use the geocentric International Terrestrial Reference System 1989 (ITRS89), whereas the satellite SAR interferometry components are identified by the Lines of Sight (LOSs) between a satellite and ground points. Moreover, SAR interferometry is a differential technique, and for that reason, displacements have no absolute reference datum. We performed datum alignment of InSAR products using precise velocity fields derived from GNSS permanent stations. The result is a coherent ground velocity field with detailed boundaries of velocity patterns that provide new information about the complex geodynamics involved on the Italian peninsula and about local movements
Landslide damming hazard susceptibility maps: a new GIS-based procedure for risk management
Full text linkA complete landslide dam hazard management incorporates two assessment phases: the damming probability and the breach hazard. A prompt evaluation of the dam stability is crucial during the emergency to mitigate its consequences, but a reliable risk assessment can be realized only after the event has occurred, when the available time is very short. Therefore, it is necessary to develop tools able to help in mapping the spatial probability of damming over large areas for land-use planning, in order to better constrain consequence analysis and risk scenarios for setting up mitigation measures. In this work, a semi-automated GIS-based mapping methodology, based on a statistical correlation of morphometric parameters described by a morphological index, is proposed to spatially assess the likelihood of a river obstruction by landslide damming through two main mechanisms: the reactivation of existing landslides and the formation of new landslides. The two mapping methods (damming predisposition and damming probability) were used on a test area, the Arno River basin in Italy. The Eastern part of the basin resulted as the most susceptible to damming events in the whole basin. These are the highest mountain ridges in the basin (about 1600 m a.s.l.), characterized by calcareous, arenaceous, and marl lithology. The results are confirmed by the high concentration of the known historical landslide dams in the area according to existing inventories
Closed reduction of acute volar dislocation of the distal radioulnar joint
No full textIsolated acute distal radioulnar joint (DRUJ) dislocation is a rare injury. In this report we describe a case of acute traumatic volar dislocation of the ulnar head in a 70-year-old man after an accidental fal
Mako Robotic Arm-Assisted Unicompartmental Knee Arthroplasty
No full textThe Mako [Mako Surgical Corp. (Stryker), Fort Lauderdale, FL, USA] robotic arm-assisted (RA) unicompartmental knee arthroplasty (UKA) enables the surgeon to perform resurfacing partial knee replacement adapting implant placement to the patient’s anatomy before bone preparation with higher accuracy, reproducibility and survivorship than conventional UKA. This robotic system allows three UKA procedures: medial, lateral, and patello-femoral partial knee replacement. A three-dimensional reconstruction is processed from a lower limb CT scan of the patient and is used for preoperative and intraoperative planning. Using a navigation-based system established on infrared optical arrays, with stable sensors fixed to the patient’s femur and tibia, the surgeon is able to perform surgery with the assistance of a robotic arm that provides neurosensory haptic feedback for bone preparation. This system is based on the concept of patient-specific knee arthroplasty, based on the patient’s anatomy, allowing the surgeon to avoid overcorrection and components’ malalignment during implant placement. The sequence of five phases is the base of every Mako RA-UKA: system setting, preoperative planning and surgical setting, system registration, intraoperative planning and soft-tissue balancing, and finally haptically controlled bone preparation
Improving Landslide Detection on SAR Data Through Deep Learning
No full textIn this letter, we use deep learning convolutional neural networks (CNNs) to compare the landslide mapping and classification performances of optical images (from Sentinel-2) and synthetic aperture radar (SAR) images (from Sentinel-1). The training, validation, and test zones used to independently evaluate the performance of the CNN on different datasets are located in the eastern Iburi subprefecture in Hokkaido, where, at 03.08 local time (JST) on September 6, 2018, an Mw 6.6 earthquake triggered about 8000 coseismic landslides. We analyzed the conditions before and after the earthquake exploiting multipolarization SAR as well as optical data by means of a CNN implemented in TensorFlow that points out the locations where the landslide class is predicted as more likely. As expected, the CNN runs on optical images proved itself excellent for the landslide detection task, achieving an overall accuracy of 98.96%, while CNNs based on the combination of ground range detected (GRD) SAR data reached overall accuracies beyond 95%. Our findings show that the integrated use of SAR data may also allow for rapid detection even during storms and under dense cloud cover and provides comparable accuracy to classical optical change detection in landslide recognition and detection
- …
