Qgis Aerial Imagery Plugin

0 views
Skip to first unread message

Tiana Dubree

unread,
Aug 5, 2024, 8:56:29 AM8/5/24
to alacexin
TheSentinel Hub QGIS plugin is built on the OGC API in the Sentinel Hub system of APIs. The OGC API can provide a collection as a Web Mapping Service (WMS) that can be called in QGIS. It provides a graphical interface to the selection of image layers and dates, mosaicking, and cloud filtering that would be rather complicated to handle in a simple WMS request. You can use the default configurations available for each Sentinel satellite, or build your own configurations and visualizations, fully harnessing the power of custom scripts.

Now that you are logged in, you can access the Create panel. Here we create a WMS (or WMTS/WFS) data layer that can then be loaded and viewed in the QGIS map window. The first step is to select the desired Configuration. The dropdown menu will have each configuration that is available in your dashboard. The simplest way to create such a configuration is to use the readily available template options: there is a configuration for each Sentinel satellite sensor, some with several levels of pre-processing. Click New configuration, and you will see a list of templates. When you choose one, the list of Layers within the configuration is displayed. In the plugin, the layers you added to your configuration will be available for selection and you can switch between them or even add several of them in the same layer. For now, we will stick with the available layers, but we discuss creating user-defined layers later in this blog post.


The Create panel displays a calendar, where dates of available images are highlighted. The default setting creates a mosaic of the time interval shown, but if you select Use exact date, there is no mosaicking, and you see exactly the image of the date you selected. For optical sensor configurations, you can adjust the acceptable level of cloud coverage to filter out cloudy images.


In the final step, you have to decide whether to create a new layer in QGIS or update an existing one. For the first time, you have to select Create, and a new WMS layer will be added into your GIS, displaying the satellite imagery of your choice. Later, if you want to look at a different date or data product, you can Update this layer, or create a new one if necessary. Take care: if you click Update, the selected layer will be updated with the satellite imagery, regardless of what it was before.


Now that you have satellite imagery in your GIS, you can explore the world and find what interests you. QGIS has advanced tools for creating and exporting a print-ready map, but if you just want to download a view of your satellite image, you can use the Download panel. Choose the image format, set the area of interest and resolution, and you have your picture.


If you want a mosaic of the best images of your study area and time period, uncheck Use exact date, select a start and end date from the calendar and choose image priority. This is suitable for an overview of a large area. However, if you zoom out too far, you will encounter the zoom limit of the plugin and no image will be loaded.


One of the nice things in QGIS is that OpenStreetMap can be loaded using the QuickMapServices plugin. If you want some basic information to help your interpretation, you can experiment with adding OpenStreetMap and adjusting the layer transparency (Right-click on the layer name on the Layers panel, Properties, Transparency) so that you see both layers at the same time.


The default configurations already have a lot of options, but you can customize them further. If you select the Full WMS template for your configuration, you will have full freedom to select your data source and script. Click Add new layer, and choose the data source sensor from the menu. You can optionally also select a time range and a cloud coverage limit during the configuration. By clicking on the pencil icon by Data processing, you open a scripting window. Here you can select pre-defined data products and tweak their settings, but also access the custom script repository to copy and paste any scripts you want (except data fusion, for now).


Here we selected the Ulyssys Water Quality Viewer, a popular custom script for visualizing sediment and algae in water. After loading the newly created Water_Quality configuration and selecting the Sen_2_UWQV layer, the high algae concentration in the Orbetello Lagoon and the sediment-rich water of the Albegna River is shown.


Satellite imagery time series can be very useful crop type classification. However, identifying the crop type from a single image is only possible for some crops at some dates. If you want to map several crops, you will need to look at a time series.


By checking monthly images during the growing season, it is easy to separate summer and winter cereals, rapeseed based on the flowering period, and maize based on the timing of the harvest. Grasslands and fodder plants can be recognized based on their time of mowing, and the presence of catch crops can also be interpreted visually. Here we investigate a set of example parcels around the small town of Emertsham in Bavaria, Germany (you can take a closer look at the place in Copernicus Browser with this link). We use the True Colour layer of the Sentinel-2 L2A Configuration for interpretation, creating a time series of monthly mosaics from April to October. For this, we select the first and last date of each month as start and end dates, apply a 25% cloud cover limit, with the least cloudy image taking the highest priority.


QGIS offers many different tools for analysing and evaluating your data. Satellite imagery can serve as a nice background, offer an insight into processes as they happen, or even serve as a basis for visual interpretation of information from a time series. The Sentinel Hub QGIS plugin brings the Copernicus Data Space Ecosystem right into your desktop GIS software.


QGIS is a free tool used by people working in geographical information science, usually dealing with any kind of information related to geography, topology, cartography and such. But an insurance investigator, law enforcement agency or journalist can easily use QGIS in several ways for things like geolocation, or gain extra information on a specific area or situation. In this article I'll go over the installation of QGIS, some basics on the layout and how it can be fed with hi-res aerial imagery with a few simple clicks of the mouse.


But why would you choose to run QGIS, if you have the amazing free imagery of Google Earth, you say? Simply because there are towns or municipalities that offer extreme high resolution aerial imagery for free, sometimes up to 3.5cm, where Google Earth will 'only' provide up to 30cm resolution satellite imagery in Europe or the US, with some places up to 15cm. But the highest resolution is gained from aerial photography, and the resolution in those can be absolutely stunning. Not every city or town is covered, but I will give you the tools and tricks so you can find what is out there for your case.


Don't treat the sources that will becovered as a replacement for the latest Maxar imagery or from Google Earth that has multiple years within their timeline function. It is best to purely treat this as an extra resource. Sometimes it can be handy to have extremely hi-res imagery to identify specific details that are too blurry in satellite images. Or maybe you are looking for more context about an area that hasn't changed for years, or you want some high resolution media to use in reports or article. And maybe you have a case where you need to be able to actually read the graffiti in a certain skatepark where hi-res aerial imagery is available? In that case, keep on reading.


QGIS can be downloaded from their website, and offers downloads for Windows, macOS, Linux, BSD and there are even versions for Android and Windows tablets. In this article I have used the Windows installer, but for macOS it is a lot easier since QGIS is downloaded as a standalone program, where the Windows version comes with a OSGeo4W installer. This will give the user a plethora of options, software and libraries to install, but for our purpose, we are just going to be using QGIS and nothing else.


After that we need to approve the ever-present terms of service before we can start the actual installation. In total it downloads a few hundred MB's, so make sure you have enough time and a fast connection.


Start QGIS and the main work-area opens up that consists of a few general areas and loads of menu buttons. Don't get overwhelmed by this, since we won't be using a lot of the menus for now. On the top we have the main menu (1) with loads of buttons that open up different functionalities of QGIS. On the left we have the browser (2) where we can open local files, and under it the layers (3). The rest of the screen is filled with the general working area (4), where later on some high resolution imagery will appear.


What I personally like to do before I add any kind of external sources, is open a base layer, for instance from Open Street Map. This way I can already zoom towards my area of interest before I am going to look for specific datasets, and it can also be helpful to get multiple mapping services inside a single application. For this there are multiple plugins available that gives you such options. My choice would be either 'QuickMapServices'or the 'HCMGIS' plugin. With 'HCMGIS' one can choose between imagery from Google, Bing, ESRI For now I'll be concentrating on the first one. In order to install it you go to the 'Plugin' menu and open the 'Manage and Install Plugins'.


Search for the plugin 'QuickMapServices' and click 'Install Plugin'. After it has been installed and you have created a new project from the start page, you go to the 'web' menu or look for the little globe icon in the menu bar. From there you can quickly select the OSM layer to be displayed.


To illustrate why I like to add OSM as a base layer, can be seen in the next screenshot. On the left there is nothing besides the aerial imagery that was loaded, in this case the Canadian town of Regina in Saskatchewan. On the right I added a base layer, which gives me an indication of the surrounding areas, entry and exit roads, and when I turn off the imagery layer I can quickly go over the area to find a particular address, an area that looks interesting, or get some general bearing on the location before turning on the imagery layer again.

3a8082e126
Reply all
Reply to author
Forward
0 new messages