scholarly journals An Efficient Multi-Sensor Remote Sensing Image Clustering in Urban Areas via Boosted Convolutional Autoencoder (BCAE)

2021 ◽  
Vol 13 (13) ◽  
pp. 2501
Author(s):  
Maryam Rahimzad ◽  
Saeid Homayouni ◽  
Amin Alizadeh Naeini ◽  
Saeed Nadi
Keyword(s):  
Remote Sensing ◽  
Urban Areas ◽  
Vegetation Index ◽  
Feature Learning ◽  
Optical Data ◽  
Image Clustering ◽  
Surface Model ◽  
Discrimination Power ◽  
Urban Image ◽  
Convolutional Autoencoder

High-resolution urban image clustering has remained a challenging task. This is mainly because its performance strongly depends on the discrimination power of features. Recently, several studies focused on unsupervised learning methods by autoencoders to learn and extract more efficient features for clustering purposes. This paper proposes a Boosted Convolutional AutoEncoder (BCAE) method based on feature learning for efficient urban image clustering. The proposed method was applied to multi-sensor remote-sensing images through a multistep workflow. The optical data were first preprocessed by applying a Minimum Noise Fraction (MNF) transformation. Then, these MNF features, in addition to the normalized Digital Surface Model (nDSM) and vegetation indexes such as Normalized Difference Vegetation Index (NDVI) and Excess Green (ExG(2)), were used as the inputs of the BCAE model. Next, our proposed convolutional autoencoder was trained to automatically encode upgraded features and boost the hand-crafted features for producing more clustering-friendly ones. Then, we employed the Mini Batch K-Means algorithm to cluster deep features. Finally, the comparative feature sets were manually designed in three modes to prove the efficiency of the proposed method in extracting compelling features. Experiments on three datasets show the efficiency of BCAE for feature learning. According to the experimental results, by applying the proposed method, the ultimate features become more suitable for clustering, and spatial correlation among the pixels in the feature learning process is also considered.

Damage assessment mapping of rural environments; integration of SAR and Optical data

2021 ◽  
Author(s):  
Shiran Havivi ◽  
Stanley R. Rotman ◽  
Dan G. Blumberg ◽  
Shimrit Maman
Keyword(s):  
Natural Disaster ◽  
Rural Areas ◽  
Urban Areas ◽  
Damage Assessment ◽  
Vegetation Index ◽  
Urban Environments ◽  
Optical Data ◽  
Damage Level ◽  
Rural Environments ◽  
Rural And Urban

<p>The damage caused by a natural disaster in rural areas differs in nature, extent, landscape and in structure, from the damage in urban environments. Previous and current studies focus mainly on mapping damaged structures in urban areas after catastrophe events such as an earthquake or tsunami. Yet, research focusing on the damage level or its distribution in rural areas is absent. In order to apply an emergency response and for effective disaster management, it is necessary to understand and characterize the nature of the damage in each different environment. </p><p>Havivi et al. (2018), published a damage assessment algorithm that makes use of SAR images combined with optical data, for rapid mapping and compiling a damage assessment map following a natural disaster. The affected areas are analyzed using interferometric SAR (InSAR) coherence. To overcome the loss of coherence caused by changes in vegetation, optical images are used to produce a mask by computing the Normalized Difference Vegetation Index (NDVI) and removing the vegetated area from the scene. Due to the differences in geomorphological settings and landuse\landcover between rural and urban settlements, the above algorithm is modified and adjusted by inserting the Modified Normalized Difference Water Index (MNDWI) to better suit rural environments and their unique response after a disaster. MNDWI is used for detection, identification and extraction of waterbodies (such as irrigation canals, streams, rivers, lakes, etc.), allowing their removal which causes lack of coherence at the post stage of the event. Furthermore, it is used as an indicator for highlighting prone regions that might be severely affected pre disaster event. Thresholds are determined for the co-event coherence map (≤ 0.5), the NDVI (≥ 0.4) and the MNDWI (≥ 0), and the three layers are combined into one. Based on the combined map, a damage assessment map is generated. </p><p>As a case study, this algorithm was applied to the areas affected by multi-hazard event, following the Sulawesi earthquake and subsequent tsunami in Palu, Indonesia, which occurred on September 28th, 2018. High-resolution COSMO-SkyMed images pre and post the event, alongside a Sentinel-2 image pre- event are used as inputs. The output damage assessment map provides a quantitative assessment and spatial distribution of the damage in both the rural and urban environments. The results highlight the applicability of the algorithm for a variety of disaster events and sensors. In addition, the results enhance the contribution of the water component to the analysis pre and post the event in rural areas. Thus, while in urban regions the spatial extent of the damage will occur in its proximity to the coastline or the fault, rural regions, even in significant distance will experience extensive damage due secondary hazards as liquefaction processes.     </p>

scholarly journals Integrated Approach for Building Extraction from InSAR and Optical Image using Object Oriented Analysis Technique

2017 ◽  
Vol 13 ◽  
pp. 16-23
Author(s):  
Bikash Kumar Karna ◽  
Ashutosh Bhardawaj
Keyword(s):  
Urban Areas ◽  
City Planning ◽  
Object Oriented ◽  
Building Blocks ◽  
Optical Data ◽  
Fuzzy Membership Function ◽  
Surface Model ◽  
Building Extraction ◽  
Object Oriented Analysis

Building extraction in built-up area is of great interest for visualization, simulation and monitoring urban landscape which is used for town/city planning as well as regional planning. Building extraction in urban areas based on merely a single high resolution optical data is often hard to conduct and to improve quality of building detection with consistency, completeness and correctness. Optical images are one of the major sources of individual building extraction from orthoimage but most of these do not produce anticipated result especially to building’s shape and outlines in dense urban environment. Extraction of objects from InSAR images is a complicated phenomenon for interpretability due to side looking geometry and effects of layover, foreshortening, shadowing and multi bounce scattering. In this study, buildings and building blocks are extracted from fusion of optical and InSAR data using object oriented analysis (OOA) technique. The improvement of building footprint has done with rectangular fit for building hypothesis and building height from normalized digital surface model (nDSM) based on fuzzy membership function. The results of building extraction has found reasonably good and accurate in planned urban layouts. The quality of building extraction has highly dependent on settlement density, contrast and other image characteristics.Nepalese Journal on Geoinformatics -13, 2014, Page: 16-23

scholarly journals RANDOM-FOREST-ENSEMBLE-BASED CLASSIFICATION OF HIGH-RESOLUTION REMOTE SENSING IMAGES AND NDSM OVER URBAN AREAS

2017 ◽  
Vol XLII-2/W7 ◽  
pp. 887-892
Author(s):  
X. F. Sun ◽  
X. G. Lin
Keyword(s):  
Remote Sensing ◽  
High Resolution ◽  
Random Forest ◽  
Urban Areas ◽  
Vegetation Index ◽  
Conditional Random Field ◽  
Remote Sensing Data ◽  
Intermediate Step ◽  
Sensing Data ◽  
3D Geometry

As an intermediate step between raw remote sensing data and digital urban maps, remote sensing data classification has been a challenging and long-standing research problem in the community of remote sensing. In this work, an effective classification method is proposed for classifying high-resolution remote sensing data over urban areas. Starting from high resolution multi-spectral images and 3D geometry data, our method proceeds in three main stages: feature extraction, classification, and classified result refinement. First, we extract color, vegetation index and texture features from the multi-spectral image and compute the height, elevation texture and differential morphological profile (DMP) features from the 3D geometry data. Then in the classification stage, multiple random forest (RF) classifiers are trained separately, then combined to form a RF ensemble to estimate each sample’s category probabilities. Finally the probabilities along with the feature importance indicator outputted by RF ensemble are used to construct a fully connected conditional random field (FCCRF) graph model, by which the classification results are refined through mean-field based statistical inference. Experiments on the ISPRS Semantic Labeling Contest dataset show that our proposed 3-stage method achieves 86.9% overall accuracy on the test data.

scholarly journals A Hybrid Autoencoder Network for Unsupervised Image Clustering

Algorithms ◽  
2019 ◽  
Vol 12 (6) ◽  
pp. 122 ◽  
Author(s):  
Pei-Yin Chen ◽  
Jih-Jeng Huang
Keyword(s):  
Feature Learning ◽  
Image Data ◽  
Image Clustering ◽  
Great Success ◽  
Distance Metrics ◽  
Clustering Methods ◽  
Unsupervised Feature Learning ◽  
Agglomerative Clustering ◽  
Convolutional Autoencoder ◽  
Important Field

Image clustering involves the process of mapping an archive image into a cluster such that the set of clusters has the same information. It is an important field of machine learning and computer vision. While traditional clustering methods, such as k-means or the agglomerative clustering method, have been widely used for the task of clustering, it is difficult for them to handle image data due to having no predefined distance metrics and high dimensionality. Recently, deep unsupervised feature learning methods, such as the autoencoder (AE), have been employed for image clustering with great success. However, each model has its specialty and advantages for image clustering. Hence, we combine three AE-based models—the convolutional autoencoder (CAE), adversarial autoencoder (AAE), and stacked autoencoder (SAE)—to form a hybrid autoencoder (BAE) model for image clustering. The MNIST and CIFAR-10 datasets are used to test the result of the proposed models and compare the results with others. The results of the clustering criteria indicate that the proposed models outperform others in the numerical experiment.

scholarly journals The NDVI-CV Method for Mapping Evergreen Trees in Complex Urban Areas Using Reconstructed Landsat 8 Time-Series Data

Forests ◽  
2019 ◽  
Vol 10 (2) ◽  
pp. 139 ◽  
Author(s):  
Yingying Yang ◽  
Taixia Wu ◽  
Shudong Wang ◽  
Jing Li ◽  
Farhan Muhanmmad
Keyword(s):  
Remote Sensing ◽  
Time Series ◽  
Urban Areas ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Time Series Data ◽  
Series Data ◽  
Distribution Data ◽  
Landsat 8 ◽  
Oxygen Balance

Evergreen trees play a significant role in urban ecological services, such as air purification, carbon and oxygen balance, and temperature and moisture regulation. Remote sensing represents an essential technology for obtaining spatiotemporal distribution data for evergreen trees in cities. However, highly developed subtropical cities, such as Nanjing, China, have serious land fragmentation problems, which greatly increase the difficulty of extracting evergreen trees information and reduce the extraction precision of remote-sensing methods. This paper introduces a normalized difference vegetation index coefficient of variation (NDVI-CV) method to extract evergreen trees from remote-sensing data by combining the annual minimum normalized difference vegetation index (NDVIann-min) with the CV of a Landsat 8 time-series NDVI. To obtain an intra-annual, high-resolution time-series dataset, Landsat 8 cloud-free and partially cloud-free images over a three-year period were collected and reconstructed for the study area. Considering that the characteristic growth of evergreen trees remained nearly unchanged during the phenology cycle, NDVIann-min is the optimal phenological node to separate this information from that of other vegetation types. Furthermore, the CV of time-series NDVI considers all of the phenologically critical phases; therefore, the NDVI-CV method had higher extraction accuracy. As such, the approach presented herein represents a more practical and promising method based on reasonable NDVIann-min and CV thresholds to obtain spatial distribution data for evergreen trees. The experimental verification results indicated a comparable performance since the extraction accuracy of the model was over 85%, which met the classification accuracy requirements. In a cross-validation comparison with other evergreen trees’ extraction methods, the NDVI-CV method showed higher sensitivity and stability.

scholarly journals Sensitivity of a Ratio Vegetation Index Derived from Hyperspectral Remote Sensing to the Brown Planthopper Stress on Rice Plants

Sensors ◽  
2019 ◽  
Vol 19 (2) ◽  
pp. 375 ◽  
Author(s):  
Ye Tan ◽  
Jia-Yi Sun ◽  
Bing Zhang ◽  
Meng Chen ◽  
Yu Liu ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Spectral Index ◽  
Vegetation Index ◽  
Brown Planthopper ◽  
Surface Model ◽  
Rice Plants ◽  
Quadratic Curve ◽  
End Products ◽  
Quadratic Surface ◽  
Potential Applications

Remote sensing end-products related to vegetation have potential applications in monitoring the health of crops. The sensitivity of a spectral index to crop stress determines its application prospect. Our aim in this study was to explore the sensitivity of a ratio vegetation index (RVI) to identify the damage caused by brown planthoppers (BPHs) on rice plants, and to evaluate the potential application of hyperspectral end-products to monitor population size of BPH. Different numbers of the second-instar nymphs were released onto potted rice at the tillering stage. The plants were exposed to BPH for two, four, six, and eight days, and reflectance from the damaged rice was measured using a hyperspectral spectroradiometer. Measurements were done again two, four, and six days after exposure (recover days), and then the spectral index RVI746/670 was compared among rice plants infested with different numbers of BPH. The relationships between RVI746/670, the number of BPH and exposure day were simulated by linear and curve models. BPH damage resulted in a decreased spectral index RVI746/670 of rice plants. RVI746/670 well indicated the damage of rice plants caused by six–eight BPH nymphs per plant in six–eight days, but the index could not identify the damage of these nymphs in two days. The RVI746/670 showed a two–four-day delay to indicate a slight BPH damage. The spectral index RVI746/670 could indicate the physiologic compensation of plants for the feeding of BPH and the post-effect of BPH damage on rice. The RVI746/670 of rice showed a quadratic curve relation with the number of BPH nymphs and a quadratic or linear relation with the exposure day. The recover day had no significant effects on RVI746/670. The RVI746/670 (Y) could be simulated by a quadratic surface model based on the number of BPH (N) and exposure day (T): Y = 3.09427 + 0.59111T + 0.44296N − 0.03683T2 − 0.03035N2 − 0.08188NT (R2 = 0.5228, p < 0.01). In summary, the spectral index RVI746/670 of rice is sensitive to damage caused by BPH.

scholarly journals Investigating Sentinel 2 Multispectral Imagery Efficiency in Describing Spectral Response of Vineyards Covered with Plastic Sheets

Agronomy ◽  
2020 ◽  
Vol 10 (12) ◽  
pp. 1909
Author(s):  
Enrico Borgogno-Mondino ◽  
Laura de Palma ◽  
Vittorino Novello
Keyword(s):  
Remote Sensing ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Spectral Response ◽  
Table Grape ◽  
Optical Data ◽  
Strong Impact ◽  
Spectral Indices ◽  
Intrinsic Nature ◽  
Sentinel 2

The protection of vineyards with overhead plastic covers is a technique largely applied in table grape growing. As with other crops, remote sensing of vegetation spectral reflectance is a useful tool for improving management even for table grape viticulture. The remote sensing of the spectral signals emitted by vegetation of covered vineyards is currently an open field of investigation, given the intrinsic nature of plastic sheets that can have a strong impact on the reflection from the underlying vegetation. Baring these premises in mind, the aim of the present work was to run preliminary tests on table grape vineyards covered with polyethylene sheets, using Copernicus Sentinel 2 (Level 2A product) free optical data, and compare their spectral response with that of similar uncovered vineyards to assess if a reliable spectral signal is detectable through the plastic cover. Vine phenology, air temperature and shoot growth, were monitored during the 2016 growing cycle. Twenty-four Copernicus Sentinel 2 (S2, Level 2A product) images were used to investigate if, in spite of plastic sheets, vine phenology can be similarly described with and without plastic covers. For this purpose, time series of S2 at-the-ground reflectance calibrated bands and correspondent normalized difference vegetation index (NDVI), modified soil-adjusted vegetation index, version two (MSAVI2) and normalized difference water index (NDWI) spectral indices were obtained and analyzed, comparing the responses of two covered vineyards with different plastic sheets in respect of two uncovered ones. Results demonstrated that no significant limitation (for both bands and spectral indices) was introduced by plastic sheets while monitoring spectral behavior of covered vineyards.

scholarly journals Improving the Accuracy of Multiple Algorithms for Crop Classification by Integrating Sentinel-1 Observations with Sentinel-2 Data

2021 ◽  
Vol 13 (2) ◽  
pp. 243
Author(s):  
Amal Chakhar ◽  
David Hernández-López ◽  
Rocío Ballesteros ◽  
Miguel A. Moreno
Keyword(s):  
Remote Sensing ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Temporal Dynamics ◽  
Remote Sensing Data ◽  
Radar Data ◽  
Optical Data ◽  
Support Vector ◽  
Sensing Data ◽  
Crop Classification

The availability of an unprecedented amount of open remote sensing data, such as Sentinel-1 and -2 data within the Copernicus program, has boosted the idea of combining the use of optical and radar data to improve the accuracy of agricultural applications such as crop classification. Sentinel-1’s Synthetic Aperture Radar (SAR) provides co- and cross-polarized backscatter, which offers the opportunity to monitor agricultural crops using radar at high spatial and temporal resolution. In this study, we assessed the potential of integrating Sentinel-1 information (VV and VH backscatter and their ratio VH/VV with Sentinel-2A data (NDVI) to perform crop classification and to define which are the most important input data that provide the most accurate classification results. Further, we examined the temporal dynamics of remote sensing data for cereal, horticultural, and industrial crops, perennials, deciduous trees, and legumes. To select the best SAR input feature, we tried two approaches, one based on classification with only SAR features and one based on integrating SAR with optical data. In total, nine scenarios were tested. Furthermore, we evaluated the performance of 22 nonparametric classifiers on which most of these algorithms had not been tested before with SAR data. The results revealed that the best performing scenario was the one integrating VH and VV with normalized difference vegetation index (NDVI) and cubic support vector machine (SVM) (the kernel function of the classifier is cubic) as the classifier with the highest accuracy among all those tested.

scholarly journals Target Classification of Similar Spatial Characteristics in Complex Urban Areas by Using Multispectral LiDAR

2022 ◽  
Vol 14 (1) ◽  
pp. 238
Author(s):  
Binhan Luo ◽  
Jian Yang ◽  
Shalei Song ◽  
Shuo Shi ◽  
Wei Gong ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Urban Areas ◽  
Point Cloud ◽  
Single Channel ◽  
Urban Land ◽  
Spectral Information ◽  
Surface Model ◽  
Three Dimension ◽  
Channel Information

With the rapid modernization, many remote-sensing sensors were developed for classifying urban land and environmental monitoring. Multispectral LiDAR, which serves as a new technology, has exhibited potential in remote-sensing monitoring due to the synchronous acquisition of three-dimension point cloud and spectral information. This study confirmed the potential of multispectral LiDAR for complex urban land cover classification through three comparative methods. Firstly, the Optech Titan LiDAR point cloud was pre-processed and ground filtered. Then, three methods were analyzed: (1) Channel 1, based on Titan data to simulate the classification of a single-band LiDAR; (2) three-channel information and the digital surface model (DSM); and (3) three-channel information and DSM combined with the calculated three normalized difference vegetation indices (NDVIs) for urban land classification. A decision tree was subsequently used in classification based on the combination of intensity information, elevation information, and spectral information. The overall classification accuracies of the point cloud using the single-channel classification and the multispectral LiDAR were 64.66% and 93.82%, respectively. The results show that multispectral LiDAR has excellent potential for classifying land use in complex urban areas due to the availability of spectral information and that the addition of elevation information to the classification process could boost classification accuracy.

scholarly journals Mapping daily evapotranspiration at field to continental scales using geostationary and polar orbiting satellite imagery

2011 ◽  
Vol 15 (1) ◽  
pp. 223-239 ◽  
Author(s):  
M. C. Anderson ◽  
W. P. Kustas ◽  
J. M. Norman ◽  
C. R. Hain ◽  
J. R. Mecikalski ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Land Surface ◽  
Vegetation Index ◽  
Normalized Difference Vegetation Index ◽  
Spatial Scales ◽  
Balance Model ◽  
Surface Model ◽  
Surface Moisture ◽  
The Us ◽  
Polar Orbiting Satellite

Abstract. Thermal infrared (TIR) remote sensing of land-surface temperature (LST) provides valuable information about the sub-surface moisture status required for estimating evapotranspiration (ET) and detecting the onset and severity of drought. While empirical indices measuring anomalies in LST and vegetation amount (e.g., as quantified by the Normalized Difference Vegetation Index; NDVI) have demonstrated utility in monitoring ET and drought conditions over large areas, they may provide ambiguous results when other factors (e.g., air temperature, advection) are affecting plant functioning. A more physically based interpretation of LST and NDVI and their relationship to sub-surface moisture conditions can be obtained with a surface energy balance model driven by TIR remote sensing. The Atmosphere-Land Exchange Inverse (ALEXI) model is a multi-sensor TIR approach to ET mapping, coupling a two-source (soil + canopy) land-surface model with an atmospheric boundary layer model in time-differencing mode to routinely and robustly map daily fluxes at continental scales and 5 to 10-km resolution using thermal band imagery and insolation estimates from geostationary satellites. A related algorithm (DisALEXI) spatially disaggregates ALEXI fluxes down to finer spatial scales using moderate resolution TIR imagery from polar orbiting satellites. An overview of this modeling approach is presented, along with strategies for fusing information from multiple satellite platforms and wavebands to map daily ET down to resolutions on the order of 10 m. The ALEXI/DisALEXI model has potential for global applications by integrating data from multiple geostationary meteorological satellite systems, such as the US Geostationary Operational Environmental Satellites, the European Meteosat satellites, the Chinese Fen-yung 2B series, and the Japanese Geostationary Meteorological Satellites. Work is underway to further evaluate multi-scale ALEXI implementations over the US, Europe, Africa and other continents with geostationary satellite coverage.

Sign in / Sign up

Export Citation Format

Share Document