scholarly journals Deep Convolutional Neural Network for Flood Extent Mapping Using Unmanned Aerial Vehicles Data

Sensors ◽  
2019 ◽  
Vol 19 (7) ◽  
pp. 1486 ◽  
Author(s):  
Asmamaw Gebrehiwot ◽  
Leila Hashemi-Beni ◽  
Gary Thompson ◽  
Parisa Kordjamshidi ◽  
Thomas Langan
Keyword(s):  
Unmanned Aerial Vehicles ◽  
Image Classification ◽  
Urban Areas ◽  
Human Life ◽  
Confusion Matrix ◽  
Support Vector ◽  
Response Planning ◽  
Aerial Vehicles ◽  
Hazardous Area ◽  
Flooded Areas

Flooding is one of the leading threats of natural disasters to human life and property, especially in densely populated urban areas. Rapid and precise extraction of the flooded areas is key to supporting emergency-response planning and providing damage assessment in both spatial and temporal measurements. Unmanned Aerial Vehicles (UAV) technology has recently been recognized as an efficient photogrammetry data acquisition platform to quickly deliver high-resolution imagery because of its cost-effectiveness, ability to fly at lower altitudes, and ability to enter a hazardous area. Different image classification methods including SVM (Support Vector Machine) have been used for flood extent mapping. In recent years, there has been a significant improvement in remote sensing image classification using Convolutional Neural Networks (CNNs). CNNs have demonstrated excellent performance on various tasks including image classification, feature extraction, and segmentation. CNNs can learn features automatically from large datasets through the organization of multi-layers of neurons and have the ability to implement nonlinear decision functions. This study investigates the potential of CNN approaches to extract flooded areas from UAV imagery. A VGG-based fully convolutional network (FCN-16s) was used in this research. The model was fine-tuned and a k-fold cross-validation was applied to estimate the performance of the model on the new UAV imagery dataset. This approach allowed FCN-16s to be trained on the datasets that contained only one hundred training samples, and resulted in a highly accurate classification. Confusion matrix was calculated to estimate the accuracy of the proposed method. The image segmentation results obtained from FCN-16s were compared from the results obtained from FCN-8s, FCN-32s and SVMs. Experimental results showed that the FCNs could extract flooded areas precisely from UAV images compared to the traditional classifiers such as SVMs. The classification accuracy achieved by FCN-16s, FCN-8s, FCN-32s, and SVM for the water class was 97.52%, 97.8%, 94.20% and 89%, respectively.

scholarly journals Landslide Mapping Using Multiscale LiDAR Digital Elevation Models

2020 ◽  
Vol 26 (4) ◽  
pp. 405-425
Author(s):  
Javed Miandad ◽  
Margaret M. Darrow ◽  
Michael D. Hendricks ◽  
Ronald P. Daanen
Keyword(s):  
Image Classification ◽  
Vegetation Index ◽  
Confusion Matrix ◽  
Object Oriented ◽  
Support Vector ◽  
Classification Approach ◽  
Resolution Image ◽  
Kappa Value ◽  
The Individual ◽  
Segmentation Image

ABSTRACT This study presents a new methodology to identify landslide and landslide-susceptible locations in Interior Alaska using only geomorphic properties from light detection and ranging (LiDAR) derivatives (i.e., slope, profile curvature, and roughness) and the normalized difference vegetation index (NDVI), focusing on the effect of different resolutions of LiDAR images. We developed a semi-automated object-oriented image classification approach in ArcGIS 10.5 and prepared a landslide inventory from visual observation of hillshade images. The multistage work flow included combining derivatives from 1-, 2.5-, and 5-m-resolution LiDAR, image segmentation, image classification using a support vector machine classifier, and image generalization to clean false positives. We assessed classification accuracy by generating confusion matrix tables. Analysis of the results indicated that LiDAR image scale played an important role in the classification, and the use of NDVI generated better results. Overall, the LiDAR 5-m-resolution image with NDVI generated the best results with a kappa value of 0.55 and an overall accuracy of 83 percent. The LiDAR 1-m-resolution image with NDVI generated the highest producer accuracy of 73 percent in identifying landslide locations. We produced a combined overlay map by summing the individual classified maps that was able to delineate landslide objects better than the individual maps. The combined classified map from 1-, 2.5-, and 5-m-resolution LiDAR with NDVI generated producer accuracies of 60, 80, and 86 percent and user accuracies of 39, 51, and 98 percent for landslide, landslide-susceptible, and stable locations, respectively, with an overall accuracy of 84 percent and a kappa value of 0.58. This semi-automated object-oriented image classification approach demonstrated potential as a viable tool with further refinement and/or in combination with additional data sources.

scholarly journals A COMPARATIVE ASSESSMENT OF EFFICACY OF SUPER RESOLVED AIRBORNE HYPERSPECTRAL OUTPUTS IN URBAN MATERIAL AND LAND COVER INFORMATION EXTRACTION

2018 ◽  
Vol XLII-5 ◽  
pp. 653-658 ◽  
Author(s):  
K. Mishra ◽  
A. Siddiqui ◽  
V. Kumar
Keyword(s):  
Land Cover ◽  
Spatial Resolution ◽  
Urban Areas ◽  
Near Infrared ◽  
Infrared Imaging ◽  
Spatial Information ◽  
Confusion Matrix ◽  
Spectral Characteristics ◽  
Support Vector ◽  
Spectral Angle Mapper

<p><strong>Abstract.</strong> Urban areas despite being heterogeneous in nature are characterized as mixed pixels in medium to coarse resolution imagery which renders their mapping as highly inaccurate. A detailed classification of urban areas therefore needs both high spatial and spectral resolution marking the essentiality of different satellite data. Hyperspectral sensors with more than 200 contiguous bands over a narrow bandwidth of 1&amp;ndash;10<span class="thinspace"></span>nm can distinguish identical land use classes. However, such sensors possess low spatial resolution. As the exchange of rich spectral and spatial information is difficult at hardware level resolution enhancement techniques like super resolution (SR) hold the key. SR preserves the spectral characteristics and enables feature visualization at a higher spatial scale. Two SR algorithms: Anchored Neighbourhood Regression (ANR) and Sparse Regression and Natural Prior (SRP) have been executed on an airborne hyperspectral scene of Advanced Visible/Near Infrared Imaging Spectrometer-Next Generation (AVIRIS-NG) for the mixed environment centred on Kankaria Lake in the city of Ahmedabad thereby bringing down the spatial resolution from 8.1<span class="thinspace"></span>m to 4.05<span class="thinspace"></span>m. The generated super resolved outputs have been then used to map ten urban material and land cover classes identified in the study area using supervised Spectral Angle Mapper (SAM) and Support Vector Machine (SVM) classification methods. Visual comparison and accuracy assessment on the basis of confusion matrix and Pearson’s Kappa coefficient revealed that SRP super-resolved output classified using radial basis function (RBF) kernel based SVM is the best outcome thereby highlighting the superiority of SR over simple scaling up and resampling approaches.</p>

scholarly journals Choice of unmanned aerial vehicles for identification of mosquito breeding sites

2020 ◽  
Vol 15 (1) ◽  
Author(s):  
Franciely Velozo Aragão ◽  
Fernanda Cavicchioli Zola ◽  
Luis Henrique Nogueira Marinho ◽  
Daiane Maria De Genaro Chiroli ◽  
Aldo Braghini Junior ◽  
...  
Keyword(s):  
Unmanned Aerial Vehicles ◽  
Urban Areas ◽  
High Mobility ◽  
World Health ◽  
Breeding Sites ◽  
Manual Search ◽  
Aerial Vehicles ◽  
The World ◽  
Mosquito Breeding ◽  
Health Organization

The disordered urban growth that may favour the emergence of the Aedes aegypti mosquito in cities is a problem of increasing magnitude in middle- and high-income countries in the tropical part of the world. Currently, the World Health Organization (WHO) considers the control and elimination of Ae. aegypti a world-wide high priority as it is the main vector of many rapidly spreading viral diseases, dengue in particular. A major difficulty in controlling the proliferation of this vector is associated with identification of the breeding sites. The use of Unmanned Aerial Vehicles (UAVs) can be an efficient alternative to manual search because of high mobility and the ability to overcome physical obstacles, particularly in urban areas where it can offer close-up images of potential breeding sites that are difficult to reach. The objective of this study was to find a way to select the most suitable UAV for the identification of Ae. aegypti habitats by providing images of potential mosquito breeding sites. This can be accomplished by a Multiple-Criteria Decision Method (MCDM) based on an Analytical Hierarchy Process (AHP) for the evaluation of weights of the criteria used for characterizing UAVs. The alternatives were analyzed and ranked using the Fuzzy Set Theory (FST) merged with the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS). The methodology is explained and discussed with respect to identification and selection of the most appropriate UAV for aerial mapping of Aedes breeding sites.

scholarly journals Environment Classification for Unmanned Aerial Vehicle Using Convolutional Neural Networks

2020 ◽  
Vol 10 (14) ◽  
pp. 4991
Author(s):  
Carlos Villaseñor ◽  
Alberto A. Gallegos ◽  
Javier Gomez-Avila ◽  
Gehová López-González ◽  
Jorge D. Rios ◽  
...  
Keyword(s):  
Unmanned Aerial Vehicles ◽  
Computational Cost ◽  
Support Vector ◽  
Normal Amount ◽  
Human Expert ◽  
Aerial Vehicles ◽  
Training Samples ◽  
Aerial Vehicle ◽  
Water Accumulation ◽  
Low Computational Cost

Environment classification is one of the most critical tasks for Unmanned Aerial Vehicles (UAV). Since water accumulation may destabilize UAV, clouds must be detected and avoided. In a previous work presented by the authors, Superpixel Segmentation (SPS) descriptors with low computational cost are used to classify ground, sky, and clouds. In this paper, an enhanced approach to classify the environment in those three classes is presented. The proposed scheme consists of a Convolutional Neural Network (CNN) trained with a dataset generated by both, an human expert and a Support Vector Machine (SVM) to capture context and precise localization. The advantage of using this approach is that the CNN classifies each pixel, instead of a cluster like in SPS, which improves the resolution of the classification, also, is less tedious for the human expert to generate a few training samples instead of the normal amount that it is required. This proposal is implemented for images obtained from video and photographic cameras mounted on a UAV facing in the same direction of the vehicle flight. Experimental results and comparison with other approaches are shown to demonstrate the effectiveness of the algorithm.

scholarly journals A Contribution to Considerations of the Role of Embedded Systems

2014 ◽  
Vol 5 (1) ◽  
pp. 47-56 ◽  
Author(s):  
Josip Stepanić ◽  
Josip Kasać ◽  
Marjana Merkač
Keyword(s):  
Embedded Systems ◽  
Unmanned Aerial Vehicles ◽  
Human Life ◽  
Human Resources Management ◽  
Point Of View ◽  
Systems Science ◽  
General Level ◽  
Specific Level ◽  
Aerial Vehicles ◽  
Gradual Process

Abstract Background: Embedded systems are a ubiquitous part of modern civilisation. Trends point to further intensification of their use. In this article we discuss long-term implications of that process, from the point of view of systems science. Objectives: On a general level, we relate embedded systems to a general class of objects and argue about their role in human life. On a somewhat more specific level, we consider in more details the development of unmanned aerial vehicles. Methods/Approach: In order to achieve the set objectives, we conducted inductive theoretical considerations and presented the results in this section. Results: The hierarchy of notions relating human civilization to environment is established, and embedded systems are positioned within it. Conclusions: Broadening and intensification of the use of embedded systems is a gradual process, heavily intertwined with societal changes. The case study of the development of the unmanned aerial vehicles reveals the potentials of the concept of embedded systems, also in the area of human resources management

scholarly journals Evaluation of Aboveground Nitrogen Content of Winter Wheat Using Digital Imagery of Unmanned Aerial Vehicles

Sensors ◽  
2019 ◽  
Vol 19 (20) ◽  
pp. 4416 ◽  
Author(s):  
Baohua Yang ◽  
Mengxuan Wang ◽  
Zhengxia Sha ◽  
Bing Wang ◽  
Jianlin Chen ◽  
...  
Keyword(s):  
Winter Wheat ◽  
Unmanned Aerial Vehicles ◽  
Nitrogen Content ◽  
Spatial Information ◽  
Vegetation Indices ◽  
Scale Transformation ◽  
Support Vector ◽  
Spatial Feature ◽  
Aerial Vehicles ◽  
Wavelet Features

Nitrogen (N) content is an important basis for the precise management of wheat fields. The application of unmanned aerial vehicles (UAVs) in agriculture provides an easier and faster way to monitor nitrogen content. Previous studies have shown that the features acquired from UAVs yield favorable results in monitoring wheat growth. However, since most of them are based on different vegetation indices, it is difficult to meet the requirements of accurate image interpretation. Moreover, resampling also easily ignores the structural features of the image information itself. Therefore, a spectral-spatial feature is proposed combining vegetation indices (VIs) and wavelet features (WFs), especially the acquisition of wavelet features from the UAV image, which was transformed from the spatial domain to the frequency domain with a wavelet transformation. In this way, the complete spatial information of different scales can be obtained to realize good frequency localization, scale transformation, and directional change. The different models based on different features were compared, including partial least squares regression (PLSR), support vector regression (SVR), and particle swarm optimization-SVR (PSO-SVR). The results showed that the accuracy of the model based on the spectral-spatial feature by combining VIs and WFs was higher than that of VIs or WF indices alone. The performance of PSO-SVR was the best (R2 = 0.9025, root mean square error (RMSE) = 0.3287) among the three regression algorithms regardless of the use of all the original features or the combination features. Our results implied that our proposed method could improve the estimation accuracy of aboveground nitrogen content of winter wheat from UAVs with consumer digital cameras, which have greater application potential in predicting other growth parameters.

scholarly journals Short-Range Transportation Using Unmanned Aerial Vehicles (UAVs) during Disasters in Japan

Drones ◽  
2020 ◽  
Vol 4 (4) ◽  
pp. 68
Author(s):  
Koki Yakushiji ◽  
Hiroshi Fujita ◽  
Mikio Murata ◽  
Naoki Hiroi ◽  
Yuuichi Hamabe ◽  
...  
Keyword(s):  
Unmanned Aerial Vehicles ◽  
Urban Areas ◽  
Disaster Response ◽  
Blood Product ◽  
The United States ◽  
Transport Systems ◽  
Medical Assistance ◽  
Medical Supplies ◽  
Aerial Vehicles ◽  
Life Saving

Larger types of small unmanned aerial vehicles (UAVs) are beginning to be used in the United States and Europe for commercial transportation. Additionally, some blood product transport systems have been commercialized in Rwanda and other countries and used in pandemic operations for the coronavirus disease 2019 (COVID-19) in infected areas. Conversely, implementing goods transportation for commercial purposes in Japan has been difficult, especially in urban areas, due to national legislation. This study examined UAV-assisted transportation in Japan, a natural disaster hotspot, with a focus on the potential uses of UAVs in situations where traffic blockages make ground transportation impossible. UAVs were used to transport 17 kg of medical supplies belonging to a disaster medical assistance team (DMAT), along with 100 emergency meals. We also transported insulin under controlled-temperature conditions, as well as many other emergency supplies. Using UAVs to transport emergency supplies could be an effective approach when dealing with disasters. This paper summarizes the effectiveness of this approach for medical care and disaster response activities. We present a method for using drones to bridge the gap between medical and firefighting personnel, such as DMAT personnel, who are engaged in life-saving activities at the time of a disaster, and those who are unable to transport necessary goods by land using terrestrial vehicles due to traffic interruptions.

scholarly journals Presenting an Effective Method to Detect and Track the Broken Path in VANET Using UAVs

2020 ◽  
Vol 2020 ◽  
pp. 1-12
Author(s):  
Zohreh Bakhtiari ◽  
Rozita Jamili Oskouei ◽  
Mona Soleymani ◽  
Akhtar Hussain Jalbani
Keyword(s):  
Cloud Computing ◽  
Unmanned Aerial Vehicles ◽  
Urban Areas ◽  
Vehicular Ad Hoc Networks ◽  
Ad Hoc ◽  
High Mobility ◽  
Geographic Routing ◽  
Negative Influence ◽  
Aerial Vehicles ◽  
And Performance

The routing process in vehicular ad hoc networks (VANETs) is a challenging task in urban areas which is due to the high mobility of vehicles, repetitive defects of the communication path, and the various barriers that may affect the reliability of data transmission and routing. Accordingly, the connectivity in vehicular communications has received the researchers’ attention, so different geographic routing protocols have been proposed in this respect. Unmanned aerial vehicles (UAVs) are useful for overcoming routing constraints. Cloud computing has also been defined as a new infrastructure for VANET which is made up of a significant number of computing nodes including stable data centers as well as a set of mobile computing devices embedded on vehicles. The aim of this research is to simulate a VANET in an urban area using cloud computing infrastructure and applying unmanned aerial vehicles (UAV) so that the negative influence of barriers in packet delivery and routing is avoided. To evaluate, the proposed method is compared with the basic protocol ClouDiV. Ns-2 simulation results show that the proposed method outperforms with different densities and variable times in terms of efficiency and performance.

scholarly journals Low-Density Polyethylene (LDPE) Food Packaging Defect Classification using Local Binary Pattern (LBP)

2021 ◽  
Vol 2129 (1) ◽  
pp. 012052
Author(s):  
N Sabri ◽  
H N Abdull Hamed ◽  
M A Isa ◽  
N S Ghazali ◽  
Z Ibrahim
Keyword(s):  
Image Classification ◽  
Food Packaging ◽  
Confusion Matrix ◽  
Primary Phase ◽  
Small Sample ◽  
Human Vision ◽  
Classification Model ◽  
Low Density Polyethylene ◽  
Support Vector ◽  
Low Density

Abstract The motivation of this research is to automate the current food packaging inspection process by implementing the non-destructive approach. The current practices require human intervention where human vision tends to overlook the faulty on the package resulting in accuracy dilemma. Human also may be exhausted due to repeated activities. This paper provides the primary phase for effective automation of image classification solution implemented using Weka software. An evaluation of the performance of the Support Vector Machine (SVM), K-nearest Neighbour (KNN) and Random Forest (RF) classification models for Low-Density Polyethylene (LDPE) food packaging defect image classification using a small sample of dataset and Linear Binary Pattern (LBP) as feature extraction algorithm is investigated. Four criteria have been used to evaluate the performance of each classification model which is accuracy, sensitivity, specificity and precision obtained from the confusion matrix table. The results indicate that SVM performs better than RF and KNN with 95% accuracy, 95% sensitivity, 72% specificity and 95% precision in classifying LDPE food packaging defect images.

scholarly journals Using Unmanned Aerial Vehicles (UAVs) to analyze the urban environment

2019 ◽  
Vol 3 (2) ◽  
pp. 20 ◽  
Author(s):  
Efthimios Bakogiannis ◽  
Charalampos Kyriakidis ◽  
Vasileios Zafeiris
Keyword(s):  
Unmanned Aerial Vehicles ◽  
Urban Areas ◽  
3D Models ◽  
Two Dimensions ◽  
Quality Data ◽  
Building Structures ◽  
Geospatial Information ◽  
Aerial Vehicles ◽  
Time Required ◽  
The Cost

Over the last decades, the evolution of technology has helped us to facilitate various types of works in areas related to land and property management as well as spatial planning. The exploitation of new tools and methods has prompted the international interest in the recording and modeling of geospatial information in more than two dimensions depicted in traditional projects, until then. This has contributed to address a series of issues related to intense urbanization, as well as challenges in identifying complex ownership and building structures. A relatively recent such method is the mapping of buildings and wider spatial units by using Unmanned Aerial Vehicles (UAVs) that has contributed to the production of 3D models by using the appropriate software. This technology finds resonant in recent years in Greece. However, it has not been applied to the mapping of large spatial units such as urban areas. This research paper performs a wide area mapping using UAV. Its purpose is to investigate to what extent the UAVs can do it successfully. For this reason, a brief evaluation is attempted, taking into account the accuracy of the data as well as the cost and time required in relation to traditional techniques. The result justifies the specific technique that appears to produce good quality metering and quality data while helping to save resources.

Sign in / Sign up

Export Citation Format

Share Document