🌊Exciting News! #HighlyCitedPaper🌊 The paper on "Prediction of Significant Wave Height in Offshore China Based on the Machine Learning Method" has been highly cited! Thank you to all the researchers and practitioners for your interest. #ShandongUniversityofScienceandTechnology 👉Read the paper here: https://lnkd.in/g_PUjBjz #WavePrediction #MachineLearning #Oceanography
Journal of Marine Science and Engineering’s Post
More Relevant Posts
-
Our latest article, "Unlocking the Use of Raw Multispectral Earth Observation Imagery for Onboard Artificial Intelligence," has been now published in the IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. This research addresses the current challenges in applying AI onboard Earth Observation satellites for time-critical applications like natural disaster response. We present a novel methodology for automating dataset creation from raw Sentinel-2 data, enabling detection of events such as thermal hotspots. This approach enhances the availability of raw datasets, fostering the development of energy-efficient onboard AI processing systems. Key Contributions: 👉🏻1. Novel Methodology: Developed an automated pipeline for creating datasets from raw multispectral imagery. 👉🏻 2. THRawS Dataset: Introduced THRawS, the first dataset of Sentinel-2 raw data containing thermal hotspots, including wildfires and volcanic eruptions. 👉🏻 3. Open-Source Tools: Released PyRawS, an open-source Python package for processing Sentinel-2 raw data. 👉🏻 4. Enhanced Onboard AI: Facilitated research into lightweight pre-processing techniques for efficient onboard AI applications. 🔗 Check out the full article here: [IEEE Xplore](https://lnkd.in/dPKi9jQz) Gabriele Meoni Nicolas Longépé Federico Serva #AI #EarthObservation #RemoteSensing #SatelliteData #Research #Innovation #IEEE #Sentinel #Copernicus
-
-
1 month and over 10,000 downloads in #QGIS later, we're releasing our most significant update to our AI Digitization plugin! Now, the plugin will digitize differently based on how zoomed in or out you are, so large features and smaller, more detailed features get a better balance of speed and accuracy (shown in the gif below). This dramatically improves performance on aerial imagery, computer generated maps, and large features on high resolution maps 👀 And, coming soon, is a completely new, more powerful AI model trained to preform better on aerial imagery, geology maps, and black and white maps, all while getting distracted less and making longer completions... #geospatial #gis #geoai
-
-
When you download our plugin in QGIS you'll automatically be on 1.10, or if you already installed it there's an option to update in the manage plugins window Give the new zoom a try and let us know what you think!
1 month and over 10,000 downloads in #QGIS later, we're releasing our most significant update to our AI Digitization plugin! Now, the plugin will digitize differently based on how zoomed in or out you are, so large features and smaller, more detailed features get a better balance of speed and accuracy (shown in the gif below). This dramatically improves performance on aerial imagery, computer generated maps, and large features on high resolution maps 👀 And, coming soon, is a completely new, more powerful AI model trained to preform better on aerial imagery, geology maps, and black and white maps, all while getting distracted less and making longer completions... #geospatial #gis #geoai
-
-
Amazing GIS tool for your polylines and polygons! I have done similar picking roads during the development of PIE-Label, I wonder if this auto-plot plugin works for unseparated or overlapping shape files, and satellite radar or optical images with tree shadows and various ground features such as muddy, concrete or asphalt roads. #QGIS #mapping #shapefile #ai
1 month and over 10,000 downloads in #QGIS later, we're releasing our most significant update to our AI Digitization plugin! Now, the plugin will digitize differently based on how zoomed in or out you are, so large features and smaller, more detailed features get a better balance of speed and accuracy (shown in the gif below). This dramatically improves performance on aerial imagery, computer generated maps, and large features on high resolution maps 👀 And, coming soon, is a completely new, more powerful AI model trained to preform better on aerial imagery, geology maps, and black and white maps, all while getting distracted less and making longer completions... #geospatial #gis #geoai
-
-
Excellent!. Digitizing has been a real pain for some, especially tracing it manually on a raster map. Loving this autotracing tool already and will definitely give it a try!
1 month and over 10,000 downloads in #QGIS later, we're releasing our most significant update to our AI Digitization plugin! Now, the plugin will digitize differently based on how zoomed in or out you are, so large features and smaller, more detailed features get a better balance of speed and accuracy (shown in the gif below). This dramatically improves performance on aerial imagery, computer generated maps, and large features on high resolution maps 👀 And, coming soon, is a completely new, more powerful AI model trained to preform better on aerial imagery, geology maps, and black and white maps, all while getting distracted less and making longer completions... #geospatial #gis #geoai
-
-
QGIS is getting AI digitizing capabilities!
1 month and over 10,000 downloads in #QGIS later, we're releasing our most significant update to our AI Digitization plugin! Now, the plugin will digitize differently based on how zoomed in or out you are, so large features and smaller, more detailed features get a better balance of speed and accuracy (shown in the gif below). This dramatically improves performance on aerial imagery, computer generated maps, and large features on high resolution maps 👀 And, coming soon, is a completely new, more powerful AI model trained to preform better on aerial imagery, geology maps, and black and white maps, all while getting distracted less and making longer completions... #geospatial #gis #geoai
-
-
Advisor | Geospatial | GeoAI | CMO | Problem Solver | Former Esri executive | Earth Champion | Speaker
Check out the latest update to the #QGIS digitizing plugin. There are still thousands of maps needing to be digitized! Let’s go! #GIS #opensource
1 month and over 10,000 downloads in #QGIS later, we're releasing our most significant update to our AI Digitization plugin! Now, the plugin will digitize differently based on how zoomed in or out you are, so large features and smaller, more detailed features get a better balance of speed and accuracy (shown in the gif below). This dramatically improves performance on aerial imagery, computer generated maps, and large features on high resolution maps 👀 And, coming soon, is a completely new, more powerful AI model trained to preform better on aerial imagery, geology maps, and black and white maps, all while getting distracted less and making longer completions... #geospatial #gis #geoai
-
-
Enhancing Flood Mapping Accuracy with SAR Imagery and Deep Learning: The WVResU-Net Approach
Postdoctoral Fellow (Remote Sensing) || Deep Learning || Python || Earth Observation Expert || GeoAI || GIS || Geoinformatics
Is it possible to only use SAR imagery for accurate flood mapping? It will be!! This research was partially supported by NOAA, United States (award NA21OAR4590358) and NASA, United States (award 80NSSC23M0051). The increasing severity, duration, and frequency of destructive #floods can be attributed to shifts in #climate, #infrastructure, #landuse, and population demographics. Obtaining precise and timely data about the extent of #floodwaters is crucial for effective #emergency preparedness and mitigation efforts. Deep convolutional neural networks (CNNs) have shown astonishing effectiveness in various remote sensing applications, including #flood mapping. One of the key limitations of #CNNs is that they can only predict whether a desired feature will appear in an image, not where it can be recognized. To address this limitation, the incorporation of self-attention mechanisms deployed in vision transformers (ViTs) can be particularly effective. However, the self-attention modules in the ViTs are complex and computationally expensive, and they require a wealth of ground data to attain their full capability in image classification/segmentation. Thus, in this paper, we develop the Residual Wave Vision U-Net (WVResU-Net), a deep learning segmentation architecture that utilizes advanced Vision Multi-Layer Perceptrons (#MLPs) and ResU-Net for accurate and reliable flood mapping using Sentinel-1 SAR’s dual polarization data. Results showed the significant superiority of the developed WVResU-Net algorithms over several well-known CNN and ViT deep learning models, including Swin U-Net, U-Net+++, Attention U-Net, R2U-Net, ResU-Net, TransU-Net and TransU-Net++. Paper: https://lnkd.in/gHK3FXzc Codes: https://lnkd.in/gHFhg4tk Authors: Ali Jamali, Swalpa Kumar Roy, Leila Hashemi-Beni, Biswajeet Pradhan, Jonathan Li, and Pedram Ghamisi #flood #sar #deeplearning #machinelearning #remotesensing #imageprocessing #cnn #disastermanagement
-
-
Postdoctoral Fellow (Remote Sensing) || Deep Learning || Python || Earth Observation Expert || GeoAI || GIS || Geoinformatics
Is it possible to only use SAR imagery for accurate flood mapping? It will be!! This research was partially supported by NOAA, United States (award NA21OAR4590358) and NASA, United States (award 80NSSC23M0051). The increasing severity, duration, and frequency of destructive #floods can be attributed to shifts in #climate, #infrastructure, #landuse, and population demographics. Obtaining precise and timely data about the extent of #floodwaters is crucial for effective #emergency preparedness and mitigation efforts. Deep convolutional neural networks (CNNs) have shown astonishing effectiveness in various remote sensing applications, including #flood mapping. One of the key limitations of #CNNs is that they can only predict whether a desired feature will appear in an image, not where it can be recognized. To address this limitation, the incorporation of self-attention mechanisms deployed in vision transformers (ViTs) can be particularly effective. However, the self-attention modules in the ViTs are complex and computationally expensive, and they require a wealth of ground data to attain their full capability in image classification/segmentation. Thus, in this paper, we develop the Residual Wave Vision U-Net (WVResU-Net), a deep learning segmentation architecture that utilizes advanced Vision Multi-Layer Perceptrons (#MLPs) and ResU-Net for accurate and reliable flood mapping using Sentinel-1 SAR’s dual polarization data. Results showed the significant superiority of the developed WVResU-Net algorithms over several well-known CNN and ViT deep learning models, including Swin U-Net, U-Net+++, Attention U-Net, R2U-Net, ResU-Net, TransU-Net and TransU-Net++. Paper: https://lnkd.in/gHK3FXzc Codes: https://lnkd.in/gHFhg4tk Authors: Ali Jamali, Swalpa Kumar Roy, Leila Hashemi-Beni, Biswajeet Pradhan, Jonathan Li, and Pedram Ghamisi #flood #sar #deeplearning #machinelearning #remotesensing #imageprocessing #cnn #disastermanagement
-
-
Semantic segmentation plays a pivotal role in geospatial mapping by enabling the automated identification and classification of objects within satellite or aerial imagery. I will be presenting an efficient and scalable Computer Vision method for extracting building footprints from satellite imagery at the upcoming Association of State Floodplain Managers, Inc. (ASFPM) Conference. Session D (Wednesday, June 26 - 11:30AM-noon) D7: Technological Innovations in Floodplain Management; Room 255A Title : Semantic Segmentation for Flood Mapping: Prioritizing areas of Risk Assessment with Computer Vision #CV #AI #GIS #ASFPM #SemanticSegmentation
Ingénieur d’études ( Coastal Engineering )
5moC’est la décomposition en mode empirique (EMD)