395 Hits in 6.6 sec

Flood Mapping with Convolutional Neural Networks Using Spatio-Contextual Pixel Information

Chandrama Sarker, Luis Mejias, Frederic Maire, Alan Woodley
2019 Remote Sensing  
In this study, we propose a fully convolutional neural networks (F-CNNs) classification model to map the flooding extent from Landsat satellite images.  ...  A new set of different Landsat images covering flooded areas across Australia were used to evaluate the classification performance of the model.  ...  The High Performance Computing (HPC) lab of Queensland University of Technology has been utilized to run the experimental work on training data preparation and F-CNNs model training.  ... 
doi:10.3390/rs11192331 fatcat:hshwng3jhnfo3kzydvshwopyzy


V. A. Knyaz, V. A. Knyaz, P. V. Moshkantsev
2019 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
They proved the effectivity of the developed techniques for collecting and augmenting synchronized multimodal imagery dataset for convolutional neural networks model training and evaluating.  ...  So the purpose of the performed study was to develop a technique for joint calibration of color and long wave infra-red cameras which are to be used for collecting multimodal dataset needed for the tasks  ...  The results of joint color and long wave infra-red cameras calibration were used for processing of multimodal image dataset collected for the developing and evaluating convolutional neural network models  ... 
doi:10.5194/isprs-archives-xlii-2-w18-79-2019 doaj:b57be756d7c94a0a91297b0fc906a62e fatcat:zqdpcvufsbeidkmzavmzra4kmy

Detecting Gas Vapor Leaks Using Uncalibrated Sensors [article]

Diaa Badawi, Tuba Ayhan, Sule Ozev, Chengmo Yang, Alex Orailoglu, A. Enis Çetin
2019 arXiv   pre-print
Finally, we use conventional convolutional neural networks as a baseline method and compare their performance with the two aforementioned deep neural network algorithms in order to evaluate their effectiveness  ...  Additive neural networks (termed AddNet) are based on a multiplication-devoid operator and consequently exhibit energy-efficiency compared to regular neural networks.  ...  VOC gas absorbs the infra-red light appearing as a white cloud in the black-hot mode infra-red image as shown in Fig. 2 .  ... 
arXiv:1908.07619v1 fatcat:u6vk3zwjuvhvfnqdcj7kaz5aqm

Deep Perceptual Mapping for Thermal to Visible Face Recognition [article]

M. Saquib Sarfraz, Rainer Stiefelhagen
2015 arXiv   pre-print
Our approach captures the highly non-linear relationship be- tween the two modalities by using a deep neural network.  ...  We show substantive performance improvement on a difficult thermal-visible face dataset.  ...  As opposed to the visible spectrum wavelength (0.35µm to 0.74µm), the infra-red spectrum lies in four main ranges. Near infra-red 'NIR' (0.74µm-1µm), short-wave infra-red c 2015.  ... 
arXiv:1507.02879v1 fatcat:ikrpe46z4bhk7cy7zrskrqdso4

Optical Convolutional Neural Networks – Combining Silicon Photonics and Fourier Optics for Computer Vision [article]

Edward Cottle, Florent Michel, Joseph Wilson, Nick New, Iman Kundu
2020 arXiv   pre-print
The Convolutional Neural Network (CNN) is a state-of-the-art architecture for a wide range of deep learning problems, the quintessential example of which is computer vision.  ...  CNNs principally employ the convolution operation, which can be accelerated using the Fourier transform.  ...  was performed using an InGaaS Short-Wavelength Infra-Red (SWIR) camera.  ... 
arXiv:2103.09044v1 fatcat:zk5i4fol2rgndhzjq7o5wn6jfe

Forest Classification Method Based on Convolutional Neural Networks and Sentinel-2 Satellite Imagery

Eka Miranda, Achmad Benny Mutiara, Wahyu Catur Wibowo
2019 International Journal of Fuzzy Logic and Intelligent Systems  
The objective of this study is to develop a classification method based on convolutional neural network (CNN) and Sentinel-2 satellite imagery including the spectral feature, spectral index and spatial  ...  Overall accuracy was used to evaluate the performance of the classification result.  ...  He is a lecturer at Faculty of Computer Science, University of Indonesia. Email:  ... 
doi:10.5391/ijfis.2019.19.4.272 fatcat:ol7klapbyrgdvdy4bbwpj6mtbq

Deep Perceptual Mapping for Cross-Modal Face Recognition

M. Saquib Sarfraz, Rainer Stiefelhagen
2016 International Journal of Computer Vision  
Our approach captures the highly non-linear relationship between the two modalities by using a deep neural network.  ...  Due to a very large modality gap, thermal-to-visible face recognition is one of the most challenging face matching problem.  ...  Near infra-red 'NIR' (0.74µm-1µm), short-wave infra-red 'SWIR' (1-3µm), mid-wave infra-red 'MWIR' (3-5µm) and long-wave infra-red .  ... 
doi:10.1007/s11263-016-0933-2 fatcat:nyrnxf4czzf2nj2fxhvyccvsnq

Machine Learning and Big Scientific Data [article]

Tony Hey, Keith Butler, Sam Jackson, Jeyarajan Thiyagalingam
2019 arXiv   pre-print
After a brief review of some initial applications of machine learning at the Rutherford Appleton Laboratory, we focus on challenges and opportunities for AI in advancing materials science.  ...  Finally, we discuss the importance of developing some realistic machine learning benchmarks using Big Scientific Data coming from a number of different scientific domains.  ...  The spatial resolution is 0.5km in the VNIR and short-wave infra-red (SWIR) channels and 1km in the thermal IR channels. In all experiments the nadir view of channels S1-S9 are used as inputs.  ... 
arXiv:1910.07631v1 fatcat:fpwolsvxmbc7dci3lzwv3r3ba4

Spotting insects from satellites: modeling the presence of Culicoides imicola through Deep CNNs [article]

Stefano Vincenzi, Angelo Porrello, Pietro Buzzega, Annamaria Conte, Carla Ippoliti, Luca Candeloro, Alessio Di Lorenzo, Andrea Capobianco Dondona, Simone Calderara
2019 arXiv   pre-print
In this respect, we frame our task as a binary classification problem, underpinning Convolutional Neural Networks (CNNs) as being able to learn useful representation from multi-band images.  ...  Additionally, we provide a multi-instance variant, aimed at extracting temporal patterns from a short sequence of spectral images.  ...  Differently, the authors of [13] described a threedimensional Convolutional Neural Network (3D-CNN), inferring crops categories (i.e.  ... 
arXiv:1911.10024v1 fatcat:j6yrrv37mfh2fadxccoxpzh3q4

On the performance of fusion based planet-scope and Sentinel-2 data for crop classification using inception inspired deep convolutional neural network

Nasru Minallah, Mohsin Tariq, Najam Aziz, Waleed Khan, Atiq ur Rehman, Samir Brahim Belhaouari, Robertas Damasevicius
2020 PLoS ONE  
In this study, a temporal convolutional neural network (TempCNNs) model is implemented for crop classification, while considering remotely sensed imagery of the selected pilot region with specific focus  ...  on the Tobacco crop.  ...  Pooling layers are not used in our architectures, as it's mostly used in computer vision tasks where a whole image or parts of image are provided to the neural network as 2D signals with channels on the  ... 
doi:10.1371/journal.pone.0239746 pmid:32986785 fatcat:2dpv7rajpjgcnhomgjhtbj3u5y

An Automated Snow Mapper Powered by Machine Learning

Haojie Wang, Limin Zhang, Lin Wang, Jian He, Hongyu Luo
2021 Remote Sensing  
A multispectral image from Sentinel-2B, a digital elevation model, and machine learning algorithms such as random forest and convolutional neural network, were utilized.  ...  It is built in a Python environment based on object-based analysis. AutoSMILE was first applied in a mountainous area of 1002 km2 in Bome County, eastern Tibetan Plateau.  ...  , Red Edge 3, Narrow Near Infra-Red, Short Wave Infrared 1 and Short Wave Infrared 2 bands) and another three with 60 m spatial resolution (Coastal Aerosol, Water Vapor and Cirrus bands).  ... 
doi:10.3390/rs13234826 fatcat:xap7v7et6nbyvkwd2hqt4ikgme

Monitoring the Impact of Large Transport Infrastructure on Land Use and Environment Using Deep Learning and Satellite Imagery

Marko Pavlovic, Slobodan Ilic, Nenad Antonic, Dubravko Culibrk
2022 Remote Sensing  
Bands stretch from aerosols, over Near Infrared to the Short Wave Infra-Red (SWIR) spectral range with the spatial resolution ranging from 10 m to 60 m [17] .  ...  They try to tackle this problem by providing an overview of the evolution of DL with a focus on image segmentation and object detection in convolutional neural networks (CNN).  ... 
doi:10.3390/rs14102494 dblp:journals/remotesensing/PavlovicIAC22 fatcat:64i527enj5gzvnfizxxtnt534i

Investigating the Impact of Using IR Bands on Early Fire Smoke Detection from Landsat Imagery with a Lightweight CNN Model

Liang Zhao, Jixue Liu, Stefan Peters, Jiuyong Li, Simon Oliver, Norman Mueller
2022 Remote Sensing  
Adding both Short-Wave Infra-Red (SWIR) bands can further improve the model performance compared with adding only one SWIR band.  ...  To prepare for potential on-board-of-small-satellite detection, we designed a lightweight Convolutional Neural Network (CNN) model named "Variant Input Bands for Smoke Detection (VIB_SD)", which achieved  ...  The images contain six spectral bands, including the RGB bands, the Near-Infra-Red (NIR) band, and two Short-Wave Infra-Red (SWIR) bands (i.e., SWIR_1 and SWIR_2), all possessing a 30-m spatial resolution  ... 
doi:10.3390/rs14133047 fatcat:a7qs5uhvqjdgpjupk5uywv67xq

NDVI Classification using Supervised Learning Method

To keep the Land quality at its best possible, the study on Land cover images, which are acquired from satellites based on time series, spatial and colour, are required to understand how the Land can be  ...  Hence, a novel approach with proper model, Machine Learning algorithm and greater accuracy is always acceptable  ...  The BAND 1: Green, BAND II: Red, BAND III: Near Infra Red, BAND IV: Short Wave Infra Red.This image obtained is pre-processed by false coloring technique and thereby enhancing the view of image using histogram  ... 
doi:10.35940/ijitee.b7083.019320 fatcat:b6tannvskjdo7i6lc42k64mf4q

Inter-band Retrieval and Classification Using the Multi-labeled Sentinel-2 BigEarthNet Archive

Ushasi Chaudhuri, Subhadip Dey, Mihai Datcu, Biplab Banerjee, Avik Bhattacharya
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
This work exploits the contextual information capturing ability of deep neural networks (DNNs), particularly investigating multispectral band properties from Sentinel-2 image patches.  ...  Availability of an enormous volume of data demands handling large-scale, diverse data, which have been made possible with neural network-based architectures.  ...  There are six 20 m bands, i.e., band number 5 (vegetation red edge), 6 (vegetation red edge), 7 (vegetation red edge), 8A (vegetation red edge), 11 (short wave infrared), and 12 (short wave infrared),  ... 
doi:10.1109/jstars.2021.3112209 fatcat:jx3wuv2k3nb6the5ofuojgdbaq
« Previous Showing results 1 — 15 out of 395 results