Remotely sensed imagery contains a myriad of valuable information, if you know how to find it. Satellite data, aerial photography, and drone imagery are increasingly used in commercial industries, scientific research, emergency management, resource management, reconnaissance and entertainment. You commonly see examples of image analysis in newspaper or scientific journal articles, and find a wide array of uses and applications that show why remotely sensed imagery is so important.
At first glance, imagery is critical because it allows us to inspect areas of interest that are hard to get to because they are too far away or difficult to access. For example, the spreading boundaries of wildfires are difficult and dangerous to assess from the ground, and aerial or high-resolution satellite imagery can help.
But visual inspection of imagery is just the tip of the iceberg. Prompt location-based insights from imagery are important for many industries. To make it possible to extract information from imagery, Esri has blended remote sensing and GIS technologies to bridge the gap between big data, machine learning, and getting answers.
Visual inspection of imagery is just the tip of the iceberg.
With image classification tools, we can identify features from imagery with software, removing the impacts of human error and increasing the chances of finding objects that are hard to see with the human eye. We can use the cloud to share information for even more effective data collection. With the ArcGIS Platform, we can go from remotely sensed imagery to an efficient task-oriented workflow managed in the cloud.
Esri offers an end-to-end imagery platform that incorporates image management, data preparation and training, deep learning model integration, and operational workflows that can be shared within your organization or to the public. You can even leverage distributed processing power with Raster Analytics in ArcGIS Enterprise to manage and analyze large amounts of imagery quickly and share it in the cloud.
We'll walk you through two "Platform Stories" to show you how we incorporated Esri's imagery workflows with the rest of the ArcGIS Platform for efficient, action-oriented problem solving.
Our first story takes us to an area northwest of Cairo, Egypt. Here, we are interested in structures such as houses or farms that are situated close to a pipeline.
Pipeline operators are responsible for the maintenance and safety of pipelines. It is especially important to educate residents and assess pipeline safety factors when new structures are built in close proximity to the pipeline. We use imagery and Deep Learning tools to identify structures close to a pipeline in Egypt, and compare it with existing data to identify only new structures to be added to a pipeline company database.
Next, we verify the structures using ArcGIS Online and employ an out-of-the-box application to assign specific properties to field technicians for assessment and community engagement.
Finally, we use management tools to monitor the completion of each assignment, all using the ArcGIS platform.
Click on the "Pipeline Encroachment" tab at the top of the story map to find out how we did it.
Our second story brings us to a coconut palm plantation in Tonga. Here, we used Deep Learning object identification tools to classify thousands of palm trees from aerial imagery. We then overlaid this data with a vegetation health index calculated from the same imagery.
Next, we published our data to the cloud and share it with our organization.
Field workers can then access the data on their mobile devices and easily locate, inspect, and treat unhealthy trees.
Click on the "Coconut Palm Health" tab at the top of the story map to see how we did it.
Story Map created with Story Map Cascade and Story Map Series
Check out this file on BOX for all photo credits and citations
Thanks to Jeff Liedtke for editing.
An error has occurred |