Eight people have lost their lives in flooding caused by heavy rain on the 9th and 10th of September in the port city of Livorno.
Authorities issued an ‘orange’ alert before the flood, but a month’s worth of rainfall in just four hours led to much worse flooding than expected.
On the 13th of September European Space Imaging collected imagery using WorldView-2, a very high resolution satellite in the DigitalGlobe constellation. The images show muddy water sitting in industrial areas and parts of the countryside.
“Being able to assess the scale of flood damage using satellite data is very useful for emergency services and local authorities,” says Adrian Zevenbergen, Managing Director at European Space Imaging. “It can help them accurately assess the worst hit areas, allowing them to make decisions about where best to direct their resources.”
The flood comes after a summer in which Italy has been plagued by drought.
New satellite imagery obtained by European Space Imaging shows flooding in Livorno by WorldView-2 @ 50cm resolution
Satellite image of Livorno on 13th September by WorldView-2 @ 50cm resolution
Flooded fields by WorldView-2 @ 50cm resolution
The muddy river flowing into a turbulent Mediterranean by WorldView-2 @ 50cm resolution
Three land-surveying authorities finished large-scale mapping projects using very high resolution satellite images in 2024. These are the challenges, solutions and results:
Redefining low latency data, European Space Imaging (EUSI) offers Near Real-Time (NRT) satellite image delivery in only 15 minutes after collection. The new EUSI DAF
Redefining low latency data, European Space Imaging (EUSI) offers Near Real-Time (NRT) satellite image delivery in only 15 minutes after collection. The new EUSI DAF
UNet architecture for semantic segmentation with ResNet34 as encoder or feature extraction part. ResNet34 is used as an encoder or feature extractor in the contracting path and the corresponding symmetric expanding path predicts the dense segmentation output.
UNet architecture for semantic segmentation with VGG16 as the encoder or feature extractor. VGG16 is used as an encoder or feature extractor in the contracting path and the corresponding symmetric expanding path predicts the dense segmentation output.
In this model, ResNet34 is used for feature extraction and the FCN operation remains as is. The feature of ResNet architecture is exploited where just like VGG, as the number of filters double, the feature map size gets halved. This gives a similarity to VGG and ResNet architecture while supporting deeper architecture and addressing the issue of vanishing gradients while also being faster. The fully connected layer at the output of ResNet34 is not used and instead converted to fully convolutional layer by means of 1×1 convolution.
In this model, VGG16 is used for feature extraction which also performs the function of an encoder. The fully connected layer of the VGG16 is not used and instead converted to fully convolutional layer by means of 1×1 convolution.