Image services are a way to process and share collections of imagery to a wide constituency. They provide considerable flexibility and scalability for storing, processing, analyzing, and sharing imagery and raster products.
Image services are flexible and robust, and can be comprised of a variety of date types to match your needs and applications. If you have multiple input images, you can choose to create one imagery layer as an image mosaic or an image collection, or create one imagery layer for each image input. You can also create one imagery layer from a single image input.
Image services can be created from a single image that can have multiple bands or dimensions, and can be small in size or massive. They can be created from large collections of similar imagery. Image services can be comprised of a collection of orthomosaics covering a state or county, each attributed with different metadata and acquisition dates. You can also have a large collection of imagery directly captured by an aircraft or drone that you need to make quickly accessible as an orthorectified mosaic, yet provide access to the individual images and metadata. Image services can also include a large collection of digital terrain models or archives of satellite imagery covering a county or the world. Additionally, a data scientist with a multidimensional raster can set up an image service to be interactively analyzed and shared by many users interested in the results.
In all these cases, image services enable the imagery and rasters to be become quickly accessible to users connecting using desktop or web applications. As users access the image service, ArcGIS Image Server performs the required processing on the fly and dynamically mosaics the required images together. These applications can access the imagery as a seamless mosaic or query into the metadata of each image to display only a selection of the data or apply processing across the datasets to find differences or perform classification. These processes can include reprojection, orthorectification, clipping, applying complex image processing algorithms, and more.
These operations are performed in a distributed processing, storage, and sharing environment, where you configure a one-machine or multiple-machine deployment. Additionally, you can register and assign different roles to different image server machines, such as raster analysis and image hosting server roles, and a distributed raster data store in a raster analytics deployment.
There are many benefits of image services, such as the following:
- Provide quick access to hosted imagery and raster data in ArcGIS Pro, web map clients, and other desktop, mobile apps and OGC Web Map Service (WMS), Web Coverage Service (WCS), Web Map Tile Service (WMTS), and Keyhole Markup Language (KML) services. You can control which user groups and stakeholders have access to which image services, and what they can do with them.
- Manage large image collections, compile them in mosaic datasets, and provide updates in a systematic and straightforward manner.
- Process imagery quickly on the fly with dynamic image services. Dynamic image services allow you to create multiple image products on demand from a single source, without needing to store and maintain preprocessed data. You can change the functions and parameters for the images you want to process, and the results are displayed on the fly as you pan and zoom the imagery.
By making your image holdings accessible in this way, you enable users across your organization to apply your imagery and raster data in their image exploitation, analysis, and map production workflows.
Creating an image service
Image services are created when source imagery, processed imagery, and value-added raster products are published. When an image service is published, clients can connect to the image service using an ArcGIS Server connection or via REST. You can also publish an image service with WMS or WCS capabilities to allow clients to access your image service in an expanded variety of applications and devices.
When creating an image service, it is important to consider how it will be used and shared. Some fundamental questions to consider include the following:
- Is the image service comprised of one raster dataset, or multiple rasters? And if multiple rasters, is it an image mosaic, a collection of images, or multiple imagery layers?
- Is the raster data for viewing as an image backdrop for GIS layers, or as an input for analysis?
- Does the imagery have multiple bands, or need to be enhanced?
- Is any image processing required?
To learn more about important factors to consider when creating an image service, see Preparing image services.
Publishing image services
Image services allow you to serve and share imagery and raster data through a web service, making your data available for use in web clients, desktop applications, and field apps.
As the publisher of an image service, you determine what functionality is available to your stakeholders who use the image service, such as dynamic viewing, and defining templates for image processing.
Information about preparing, publishing, and processing image services is provided in a series of topics, starting with Key concepts for image services.
Caching image services
Image service caching improves the performance of image services in client applications. The cached image service can display an image quickly because ArcGIS Server does not have to generate the image dynamically. Instead, the image tiles are generated, indexed, and stored in your server cache directory. The server distributes these image tiles whenever there is a request, depending on the location of your panning and zoom levels.
When you cache an image service, you produce an image service that is accessed depending on its purpose. One purpose is to provide the fastest access to the image as a tiled service. The other purpose is to provide access to the data, for queries, downloading, access to individual items, and to use in processing and analysis.
Information about preparing and creating a cache image service is provided in a series of topics, starting with What is image service caching?