Help Content > Basic Module > Ship Detection

Purpose

This tool is used to extract a series of shape files containing the coastline detected in the input images, grouped by tide height.

To use this tool it is necessary to create a WorldTides account (https://www.worldtides.info/),. An authorization key can be obtained by creating a free account. The tide information is retrieved for free not using any credits, as specified in the website. To get the key, open the WorldTides website, click on Account and then Settings/API Key; copy the APY key code.

Technical Note

This tool is used to extract a series of shape files containing the coastline detected from the input images in correspondence of different tide heights. Here below the requirements regarding the input datastack:

•The input images must cover the same exact AOI, therefore it is important to import the images using the same Area of interest, so that the covered area is exactly the same. Alternatively it is possibile to apply the sample selection tool after the import

•The datastack must be coregistered since the analysis is performed pixel by pixel

•The datastack must be geocoded and the center of those images must be suitably close to open waters (less than 1 degree).

The workflow of the Coastline Extraction tool is the following:

•The tide information on the area of interest are retrieved from the website (https://www.worldtides.info/) at the acquisition time of each of the input images

•The min-max tide interval is divided into different height spans, the span distance is defined through the parameters named ''Tide height span'

•The input images are grouped into different groups(each corresponding to one tide height span), according to the tide height at the time of their acquisition. For each group at least "minimum group size" images are required (3 is the default value)

•The groups of images will be then used to extract the coastline for each single group by using multi/temporal indicators of the backscatter in the group

•Once the shape has been extracted some filters are applied to smooth the results and remove false positives in the acquisition.

To the resulting shape files three different filters are applied:

•the first filter smooths the results to remove gridded edges from the shape file

•the second filter removes every shape that is completely within an area of a defined size

•The third filter remove every shape with area size lower than a defined size.

All the values pertaining to these filters can be modified from the panel to achieve different levels of filtering.

The number of images must be carefully decided according to the tidal range, to the tide height span and to the minimum group size, please consider the following example:

Example:

•Tidal range= +/- 1m with respect to the zero, therefore our tidal range corresponds to 2m

•The tidal range (2 meters) is divided into different spans, the span distance is defined through the parameter named ''Tide height span'', in this case will be 25cm

•The input images are grouped into different groups (each corresponding to one tide height span), according to the tide height at the time of their acquisition, for this example leave the default value: 3

Which is the minimum number of images required?

•We have a coastline every 25cm between -1m and +1m which means: -1.0m, -0.75m, -0.5, -0.25m, 0m, 0.25m, 0.5m, 0.75, 1.0m, therefore we have 9 groups/shapes

•For every group we must have at least 3 images, therefore 9x3=27, that correspond to the minimum number of images required.

Note that those 27 images should correspond to the 9 different tide groups, and each group must have at least 3 images. This requirement can be achieved only if the user knows the tide high, for each of the 27 images, and the tide range a priori. For real case scenario these information are hardly obtainable, therefore the more images in input the better, to allow the tool to have enough input data to fill the statistical requirements (at least 3 images per group) for each group.

Input Files

Input file list

Please read carefully the technical notes to establish the minimum number of images required. This files are mandatory.

Parameters - Principal Parameters

Use tide to create groups

When this parameter is set to true, the input images are grouped into different groups (each corresponding to one tide height span), according to the tide height at the time of their acquisition. If set to false, only one group will be created. When this field is set to false the the User Auth Key does not need to be filled.

Tide Base URL

URL of the WorldTides website.

Tide User Auth Key

The authorization key to access the the WorldTides website. To get the key, open the WorldTides website, click on Account and then Settings/API Key; copy the APY key code.

Tide height span

The height span in meters that will be used to group the images.

Minimum group size

Number of images that need to be in a group to be declared valid. If a group contains less that this number of images it will be merged with the previous one.

Smoothing window size

Smoothing value for the first filter.

This value should usually be greater than 1 and less than 10 for optimal results.

Aggregation window size

Aggregation size used by the second filter.

This value should usually be greater than 1 and less than 20 for optimal results.

Filtering size

Aggregation size used by the third filter.

This value should usually be either 0 or a value between the Aggregation group size value and 150.

This is the most aggressive filter and will remove the most features if set, useful for very noisy shapes.

Minimum db threshold

This is the value that will determine the threshold between what will be considered water and land, everything under this value will be outside of the coastline shape.

Output Files

Output Root Name

Name of the output root. This file is mandatory.

The output files will contain a number between 0 and the number of groups ordered by tide height in ascending order.

General Functions

Exec

The processing step is executed.

Store Batch

The processing step is stored in the batch list. The Batch Browser button allows to load the batch processing list.

The window will be closed.

Help

Specific help document section.

Specific Function(s)

None.