scholarly journals Generating High Resolution Climate Change Projections through Single Image Super-Resolution: An Abridged Version

2018 ◽  
Author(s):  
Thomas Vandal ◽  
Evan Kodra ◽  
Sangram Ganguly ◽  
Andrew Michaelis ◽  
Ramakrishna Nemani ◽  
...  
Keyword(s):  
Climate Change ◽  
High Resolution ◽  
Statistical Downscaling ◽  
Power Plants ◽  
Super Resolution ◽  
Critical Systems ◽  
Resolution Mapping ◽  
Spatio Temporal ◽  
Single Image Super Resolution ◽  
Processing Techniques

The impacts of climate change are felt by most critical systems, such as infrastructure, ecological systems, and power-plants. However, contemporary Earth System Models (ESM) are run at spatial resolutions too coarse for assessing effects this localized. Local scale projections can be obtained using statistical downscaling, a technique which uses historical climate observations to learn a low-resolution to high-resolution mapping. The spatio-temporal nature of the climate system motivates the adaptation of super-resolution image processing techniques to statistical downscaling. In our work, we present DeepSD, a generalized stacked super resolution convolutional neural network (SRCNN) framework with multi-scale input channels for statistical downscaling of climate variables. A comparison of DeepSD to four state-of-the-art methods downscaling daily precipitation from 1 degree (~100km) to 1/8 degrees (~12.5km) over the Continental United States. Furthermore, a framework using the NASA Earth Exchange (NEX) platform is discussed for downscaling more than 20 ESM models with multiple emission scenarios.

scholarly journals TWIST-GAN: Towards Wavelet Transform and Transferred GAN for Spatio-Temporal Single Image Super Resolution

2021 ◽  
Vol 12 (6) ◽  
pp. 1-20
Author(s):  
Fayaz Ali Dharejo ◽  
Farah Deeba ◽  
Yuanchun Zhou ◽  
Bhagwan Das ◽  
Munsif Ali Jatoi ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Super Resolution ◽  
Generative Adversarial Networks ◽  
Single Image ◽  
Generative Adversarial Network ◽  
Adversarial Network ◽  
Adversarial Networks ◽  
Image Super Resolution ◽  
Spatio Temporal ◽  
Single Image Super Resolution

Single Image Super-resolution (SISR) produces high-resolution images with fine spatial resolutions from a remotely sensed image with low spatial resolution. Recently, deep learning and generative adversarial networks (GANs) have made breakthroughs for the challenging task of single image super-resolution (SISR) . However, the generated image still suffers from undesirable artifacts such as the absence of texture-feature representation and high-frequency information. We propose a frequency domain-based spatio-temporal remote sensing single image super-resolution technique to reconstruct the HR image combined with generative adversarial networks (GANs) on various frequency bands (TWIST-GAN). We have introduced a new method incorporating Wavelet Transform (WT) characteristics and transferred generative adversarial network. The LR image has been split into various frequency bands by using the WT, whereas the transfer generative adversarial network predicts high-frequency components via a proposed architecture. Finally, the inverse transfer of wavelets produces a reconstructed image with super-resolution. The model is first trained on an external DIV2 K dataset and validated with the UC Merced Landsat remote sensing dataset and Set14 with each image size of 256 × 256. Following that, transferred GANs are used to process spatio-temporal remote sensing images in order to minimize computation cost differences and improve texture information. The findings are compared qualitatively and qualitatively with the current state-of-art approaches. In addition, we saved about 43% of the GPU memory during training and accelerated the execution of our simplified version by eliminating batch normalization layers.

scholarly journals Spatial Enhancement of AWiFS along Wider Swath using NSCT

2018 ◽  
Vol 7 (3.12) ◽  
pp. 474
Author(s):  
K S. R. Radhika ◽  
C V. Rao ◽  
V Kamakshi Prasad
Keyword(s):  
Temporal Resolution ◽  
Accuracy Assessment ◽  
Super Resolution ◽  
Computation Time ◽  
Processing Technique ◽  
Sensor Data ◽  
Liss Iii ◽  
Novel Processing ◽  
Single Image Super Resolution ◽  
Processing Techniques

Image acquisition in a wider swath, cannot assess the best spatial resolution (SR) and temporal resolution (TR) simultaneously, due to inherent limitations of space borne sensors. But any of the information extraction from remote sensed (RS) images demands the above characteristics. As this is not possible onboard, suitable ground processing techniques need to be evolved to realise the requirements through advanced image processing techniques. The proposed work deals with processing of two onboard sensor data viz., Resourcesat-1 (RS1): LISS-III, which has medium swath combined with AWiFS, which has wider swath data to provide high spatial and temporal resolution at the same instant. LISS-III at 23m and 24 days, AWiFS at 56m and 5 days spatial and temporal revisits acquire the data at different swaths. In the process of acquisition at the same time, the 140km swath of LISS-III coincides at the exact centre line 740km swath of AWiFS. If the non-overlapping area of AWiFS has same features of earth’s surface as of LISS-III overlapping area, it then provides a way to increase the SR of AWiFS to SR of LISS-III in the same non-overlapping area. Using this knowledge, a novel processing technique Fast One Pair Learning and Prediction (FOPLP) is developed in which time is optimized against the existing methods. FOPLP improves the SR of LISS-III in non-overlapping area using technique Single Image Super Resolution (SISR) with Non Sub sampled Contourlet Transforms (NSCT) method and is applied on different sets of images. The proposed technique resulting into an image having TR of 5 days, 740km swath at SR of 23m. Results have shown the strength of the proposed method in terms of computation time and prediction accuracy assessment.  

High-resolution mapping of floodplain vegetation changes in large tropical rivers

2021 ◽  
Author(s):  
Luca Salerno ◽  
Álvaro Moreno-Martínez ◽  
Emma Izquierdo-Verdiguier ◽  
Nicholas Clinton ◽  
Annunziato Siviglia ◽  
...  
Keyword(s):  
Climate Change ◽  
High Resolution ◽  
Vegetation Dynamics ◽  
Tree Mortality ◽  
Vegetation Index ◽  
Riparian Forests ◽  
Tropical Rivers ◽  
Floodplain Vegetation ◽  
High Resolution Mapping ◽  
Resolution Mapping

<p>Tropical floodplain forests are among the most complex ecosystem on earth, featured by vegetation adapted to survive in seasonal flood environments. Although their ability to resist the periodic water level oscillations, recent studies have shown that riparian forests are extremely sensitive to long-term hydrological changes caused by both anthropogenic and natural disturbances. During the recent decades fragmentation and regulation of rivers induced severe alterations of natural “flood pulse” and sediment supply along the whole watercourse, causing massive tree mortality and compromising seeds spreading. The hydroclimatic anomalies of El Nino/Southern Oscillation (ENSO) and climate change impact on riparian environments, aggravating forest stress and vulnerability to fires, in cases of prolonged drought, while inducing tree mortality for anoxia, when a multi-year uninterrupted flood occurred.</p><p>In order to develop future solutions to mitigate the consequences of these disturbances and to enable a sustainable and effective management of riparian forests in the aquatic-terrestrial transitional zone (ATTZ), large-scale monitoring of these areas is necessary. Mapping and monitoring of floodplain vegetation are extremely important not only to assess vegetation status but also because vegetation represents an indicator for early signs of any physical or chemical environmental degradation. Remote sensing offers practical and efficient techniques to estimate biochemical and biophysical parameters and analyse their evolution over time even for very remote and poor accessible areas such as tropical floodplains. Nevertheless, as the main vegetation dynamics are in the narrow area at the interface terrestrial and aquatic systems, a high spatial and temporal resolution of the data is needed for their analysis. Furthermore, the extreme cloudiness of tropical regions contaminates the land surface observation causing gap in the data.</p><p>In the present study, we combine Landsat (30m spatial resolution and 16 day revisit cycle) and the MODIS missions, both from Terra and Aqua platforms (500m spatial resolution and daily revisit cycle), using HISTARFM algorithm, to reduce noise and produce monthly gap-free high-resolution (30 m) observations over land and the associated estimation of uncertainties. Subsequently, high resolution maps of normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI) were derived from the high-resolution gap free reflectance data. Furthermore, estimation of NDVI and EVI uncertainties was calculated through an error propagation analyses from uncertainties of reflectance estimates.</p><p>The framework we developed has been used to derive high resolution mapping of floodplain vegetation in the large tropical rivers that during the last decades experimented a hydrological regime transition. In a first-phase, vegetation dynamic analysis focused of the tropical large rivers in Amazonia and preliminary results of the temporal series will be presented.</p><p>The coupling of hydro-geomorphological and vegetation data enables the monitoring of riparian vegetation dynamics and a better understanding of the impact that the human footprint and climate change have on them.</p>

A High-Resolution Climate Model for the U.S. Pacific Northwest: Mesoscale Feedbacks and Local Responses to Climate Change*

2008 ◽  
Vol 21 (21) ◽  
pp. 5708-5726 ◽  
Author(s):  
Eric P. Salathé ◽  
Richard Steed ◽  
Clifford F. Mass ◽  
Patrick H. Zahn
Keyword(s):  
Climate Change ◽  
High Resolution ◽  
Snow Cover ◽  
Pacific Northwest ◽  
Statistical Downscaling ◽  
Large Scale ◽  
Climate Model ◽  
Mesoscale Model ◽  
Global Model ◽  
Temperature And Precipitation

Abstract Simulations of future climate scenarios produced with a high-resolution climate model show markedly different trends in temperature and precipitation over the Pacific Northwest than in the global model in which it is nested, apparently because of mesoscale processes not being resolved at coarse resolution. Present-day (1990–99) and future (2020–29, 2045–54, and 2090–99) conditions are simulated at high resolution (15-km grid spacing) using the fifth-generation Pennsylvania State University–NCAR Mesoscale Model (MM5) system and forced by ECHAM5 global simulations. Simulations use the Intergovernmental Panel on Climate Change (IPCC) Special Report on Emissions Scenarios (SRES) A2 emissions scenario, which assumes a rapid increase in greenhouse gas concentrations. The mesoscale simulations produce regional alterations in snow cover, cloudiness, and circulation patterns associated with interactions between the large-scale climate change and the regional topography and land–water contrasts. These changes substantially alter the temperature and precipitation trends over the region relative to the global model result or statistical downscaling. Warming is significantly amplified through snow–albedo feedback in regions where snow cover is lost. Increased onshore flow in the spring reduces the daytime warming along the coast. Precipitation increases in autumn are amplified over topography because of changes in the large-scale circulation and its interaction with the terrain. The robustness of the modeling results is established through comparisons with the observed and simulated seasonal variability and with statistical downscaling results.

Video super resolution using convolutional neural network and image fusion techniques

2021 ◽  
Vol 24 (4) ◽  
pp. 279-287
Author(s):  
Vikas Kumar ◽  
Tanupriya Choudhury ◽  
Suresh Chandra Satapathy ◽  
Ravi Tomar ◽  
Archit Aggarwal
Keyword(s):  
Neural Network ◽  
High Resolution ◽  
Convolutional Neural Network ◽  
Image Fusion ◽  
Reference Model ◽  
Super Resolution ◽  
Motion Blur ◽  
Single Image ◽  
Image Super Resolution ◽  
Single Image Super Resolution

Recently, huge progress has been achieved in the field of single image super resolution which augments the resolution of images. The idea behind super resolution is to convert low-resolution images into high-resolution images. SRCNN (Single Resolution Convolutional Neural Network) was a huge improvement over the existing methods of single-image super resolution. However, video super-resolution, despite being an active field of research, is yet to benefit from deep learning. Using still images and videos downloaded from various sources, we explore the possibility of using SRCNN along with image fusion techniques (minima, maxima, average, PCA, DWT) to improve over existing video super resolution methods. Video Super-Resolution has inherent difficulties such as unexpected motion, blur and noise. We propose Video Super Resolution – Image Fusion (VSR-IF) architecture which utilizes information from multiple frames to produce a single high- resolution frame for a video. We use SRCNN as a reference model to obtain high resolution adjacent frames and use a concatenation layer to group those frames into a single frame. Since, our method is data-driven and requires only minimal initial training, it is faster than other video super resolution methods. After testing our program, we find that our technique shows a significant improvement over SCRNN and other single image and frame super resolution techniques.

Image Super-Resolution via Saliency Sparse Representation

2014 ◽  
Vol 568-570 ◽  
pp. 659-662
Author(s):  
Xue Jun Zhang ◽  
Bing Liang Hu
Keyword(s):  
High Resolution ◽  
Sparse Representation ◽  
Saliency Detection ◽  
Computational Cost ◽  
Super Resolution ◽  
Compact Representation ◽  
High Resolution Images ◽  
Image Super Resolution ◽  
Single Image Super Resolution ◽  
Marked Area

The paper proposes a new approach to single-image super resolution (SR), which is based on sparse representation. Previous researchers just focus on the global intensive patch, without local intensive patch. The performance of dictionary trained by the local saliency intensive patch is more significant. Motivated by this, we joined the saliency detection to detect marked area in the image. We proposed a sparse representation for saliency patch of the low-resolution input, and used the coefficients of this representation to generate the high-resolution output. Compared to precious approaches which simply sample a large amount of image patch pairs, the saliency dictionary pair is a more compact representation of the patch pairs, reducing the computational cost substantially. Through the experiment, we demonstrate that our algorithm generates high-resolution images that are competitive or even superior in quality to images produced by other similar SR methods.

Risk assessment of dengue virus amplification in Europe based on spatio-temporal high resolution climate change projections

Erdkunde ◽  
2011 ◽  
Vol 65 (2) ◽  
pp. 137-150 ◽  
Author(s):  
Stephanie Margarete Thomas ◽  
Dominik Fischer ◽  
Stefanie Fleischmann ◽  
Torsten Bittner ◽  
Carl Beierkuhnlein
Keyword(s):  
Climate Change ◽  
Risk Assessment ◽  
High Resolution ◽  
Dengue Virus ◽  
Climate Change Projections ◽  
Spatio Temporal
Sign in / Sign up

Export Citation Format

Share Document