There is a growing demand for detailed and accurate landslide maps and inventories around the globe, but particularly in hazard-prone regions such as the Himalayas. Most standard mapping methods require expert knowledge, supervision and fieldwork. In this study, we use optical data from the Rapid Eye satellite and topographic factors to analyze the potential of machine learning methods, i.e., artificial neural network (ANN), support vector machines (SVM) and random forest (RF), and different deep-learning convolution neural networks (CNNs) for landslide detection. We use two training zones and one test zone to independently evaluate the performance of different methods in the highly landslide-prone Rasuwa district in Nepal. Twenty different maps are created using ANN, SVM and RF and different CNN instantiations and are compared against the results of extensive fieldwork through a mean intersection-over-union (mIOU) and other common metrics. This accuracy assessment yields the best result of 78.26% mIOU for a small window size CNN, which uses spectral information only. The additional information from a 5 m digital elevation model helps to discriminate between human settlements and landslides but does not improve the overall classification accuracy. CNNs do not automatically outperform ANN, SVM and RF, although this is sometimes claimed. Rather, the performance of CNNs strongly depends on their design, i.e., layer depth, input window sizes and training strategies. Here, we conclude that the CNN method is still in its infancy as most researchers will either use predefined parameters in solutions like Google TensorFlow or will apply different settings in a trial-and-error manner. Nevertheless, deep-learning can improve landslide mapping in the future if the effects of the different designs are better understood, enough training samples exist, and the effects of augmentation strategies to artificially increase the number of existing samples are better understood.

/pdf/evaluation-of-different-machine-learning-methods-and-deep-5rqqhxf2cs.pdf

Evaluation of Different Machine Learning Methods and Deep-Learning Convolutional Neural Networks for Landslide Detection

Drone-surveillance for search and rescue in natural disaster

Spatial prediction of groundwater potential mapping based on convolutional neural network (CNN) and support vector regression (SVR)

Unmanned aerial vehicles (UAVs) play a primary role in a plethora of technical and scientific fields owing to their wide range of applications. In particular, the provision of emergency services during the occurrence of a crisis event is a vital application domain where such aerial robots can contribute, sending out valuable assistance to both distressed humans and rescue teams. Bearing in mind that time constraints constitute a crucial parameter in search and rescue (SAR) missions, the punctual and precise detection of humans in peril is of paramount importance. The paper in hand deals with real-time human detection onboard a fully autonomous rescue UAV. Using deep learning techniques, the implemented embedded system was capable of detecting open water swimmers. This allowed the UAV to provide assistance accurately in a fully unsupervised manner, thus enhancing first responder operational capabilities. The novelty of the proposed system is the combination of global navigation satellite system (GNSS) techniques and computer vision algorithms for both precise human detection and rescue apparatus release. Details about hardware configuration as well as the system’s performance evaluation are fully discussed.

Unsupervised Human Detection with an Embedded Vision System on a Fully Autonomous UAV for Search and Rescue Operations

Convolutional neural networks for object detection in aerial imagery for disaster response and recovery

There are numerous applications of unmanned aerial vehicles (UAVs) in the management of civil infrastructure assets. A few examples include routine bridge inspections, disaster management, power line surveillance and traffic surveying. As UAV applications become widespread, increased levels of autonomy and independent decision-making are necessary to improve the safety, efficiency, and accuracy of the devices. This paper details the procedure and parameters used for the training of convolutional neural networks (CNNs) on a set of aerial images for efficient and automated object recognition. Potential application areas in the transportation field are also highlighted. The accuracy and reliability of CNNs depend on the network’s training and the selection of operational parameters. This paper details the CNN training procedure and parameter selection. The object recognition results show that by selecting a proper set of parameters, a CNN can detect and classify objects with a high level of accuracy (97.5%) and computational efficiency. Furthermore, using a convolutional neural network implemented in the “YOLO” (“You Only Look Once”) platform, objects can be tracked, detected (“seen”), and classified (“comprehended”) from video feeds supplied by UAVs in real-time.

/pdf/object-recognition-in-aerial-images-using-convolutional-4t442bid2k.pdf

Object Recognition in Aerial Images Using Convolutional Neural Networks

https://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1014&context=publichealthresources

Bilateral claudication results in alterations in the gait biomechanics at the hip and ankle joints

Data mining is a discovery procedure to explore and visualize useful but less-than-obvious information or patterns in large collections of data. Given the amount and varying parameter types in a large data set such as that of the National Bridge Inventory (NBI), using traditional clustering techniques for discovery is impractical. As a consequence, the authors have applied a novel data discovery tool, called Two-step cluster analysis, to visualize associations between concrete bridge deck design parameters and bridge deck condition ratings. Two-step cluster analysis is a powerful knowledge discovery tool that can handle categorical and interval data simultaneously and is capable of reducing dimensions for large data sets. The analysis, of a total of 9,809 concrete highway bridge decks in the Northeast climatic region, found that bridges with cast-in-place decks that have a bituminous wearing surface, a preformed fabric membrane, and epoxy-coated reinforcement protection are strongly associated wit...

Data Mining of Bridge Concrete Deck Parameters in the National Bridge Inventory by Two-Step Cluster Analysis

A better understanding of the relationships among girder stresses, cross-frame design, and skew angle of two steel I-girder bridges, which have moderate and high levels of skew and differing cross-frame designs, was revealed via field testing under various load passes of a weighed load vehicle. Strains were measured at approximately 100 points on the two structures, and the majority of the instrumentation was located on the cross-frames. The data were analyzed in terms of vertical bending, lateral bending, and warping in the girders, and in the magnitudes and distribution of stresses in the cross-frames of the two structures. These metrics were compared between the two bridges and in relation to the assumptions typically used in contemporary bridge analysis. This comparison revealed the following: the magnitudes of lateral stresses were higher than current design guidance suggests in some cases when staggered cross-frames were used; significant magnitudes of warping stresses in the girder webs, wh...

Field Evaluation of Cross-Frame and Girder Live-Load Response in Skewed Steel I-Girder Bridges

Rigidity (muscle stiffness) is one of the most disabling symptoms in Parkinson's disease (PD). It is clinically defined as an increased resistance to passive movement of a joint. There is a fundamental gap between mechanistic and applied approaches to understanding this symptom. The objective of the current study was to apply a system identification and modeling approach to differentiating the contributions of neural (enhanced muscle reflex) and non-neural (altered mechanical properties of muscle fibers) factors to rigidity. Six patients participated in the study. The wrist joint torque and muscle activities of the wrist muscles were measured during externally induced movements. Each subject was tested in the Off- and On- medication states. System identification and modeling approach was applied to separate the neural from the non-neural component with respect to the overall stiffness. Results show that both factors are responsible for rigidity in PD. Neural-related reflex component is the predominant factor in overall rigidity. Medication therapy decreased the level of reflex component to overall rigidity.

Matija Radovic

Papers

Object Recognition in Aerial Images Using Convolutional Neural Networks

Bilateral claudication results in alterations in the gait biomechanics at the hip and ankle joints

Data Mining of Bridge Concrete Deck Parameters in the National Bridge Inventory by Two-Step Cluster Analysis

Field Evaluation of Cross-Frame and Girder Live-Load Response in Skewed Steel I-Girder Bridges

System Identification and Modeling Approach to Characterizing Rigidity in Parkinson's Disease: Neural and Non-Neural Contributions