Ship Detection using Remote Sensing
Updated: May 21
Satellite Imagery can be used effectively for several workflows - Climate monitoring & Land monitoring being the two applications which I had written about, before working on this piece. In this study, I have attempted Ship detection at Sea: a) in the Gulf of Kutch using Sentinel-1 satellite imagery (and subsequently in b) the Suez Canal in the aftermath of the massive Ever Given cargo vessel stranding) with the intention to see the extent of negative change in maritime activity due to the prevalent Coronavirus lockdown conditions. More importantly, my aim is to demonstrate using small and relatable studies, the scope and benefits of deploying Location Intelligence with the intention to monitor and improve operations.
The two cases covered in this article can be accessed using the section hyperlinks below:
Ship Detection in the Gulf of Kutch
Ship Detection in the Suez Canal
SHIP DETECTION IN THE GULF OF KUTCH, INDIA - EXTENT OF REDUCED ECONOMIC ACTIVITY DUE TO CORONAVIRUS LOCKDOWN
The result of the ship detection study is depicted in the image below - More than 33% reduction in shipping activity has been spotted in the Gulf of Kutch (based on the count of the ships detected).
I have used Sentinel-1A GRD data for this study. Much thanks to RUS Copernicus and Open Street Map for the invaluable training material
How is object detection possible from satellite imagery? How accurate is it?
The satellite in concern - Sentinel-1A - has a spatial resolution of 5m by 20m. The parameter for objects to be detected was kept at a minimum of 30m in size. (Reducing it further would increase the number of false detections (objects classified as ships but not ships in reality).
The principle behind Object Detection from Aerial Imagery is how electromagnetic waves interact with on-surface materials. In this study, a ship appears brighter to the C-SAR instrument on-board the Sentinel-1 satellite as compared to the water in the vicinity, which is darker - refer visual below. CFAR algorithm was one of the technique used during this processing workflow - the same methodology can be used to detect icebergs in colder regions.
That being said, the extent of accuracy depends on geological factors, satellite instrument properties, methodology of processing the data, and even local knowledge about the AoI (Area of Interest).
How to interpret these results?
A reduction in maritime activity was expected and ~37% seems plausible to me. Certain aspects to note, however, are-
a) Not all objects are commercial cargo / crude ships. Some can be fishing vessels, port vessels, navy vessels, tourism vessels, pirate vessels, other vessels & objects, or even false detections. However, one can safely assume that the vessels clustered around Mundra Port and Kandla Port - major maritime hubs in this region - would be commercial vessels - harbingers of trade i.e. With the object size of 30m that I have used in this study, a lot of shipping objects of the non-commercial variety would be filtered out as well. All else being constant, the difference in count between the two imagery products - 148 shipping objects - is also a very significant quantity.
b) Not all shipping objects detected would necessarily imply cargo movement or active commercial vessels-at-work. As you would notice, a fair number of ships are clustered around the coastline. Several of them could be docked or dysfunctional or undergoing maintenance as well.
c) The comparison of objects detected has been done during two points in time, one year apart. Adding more time periods i.e. performing a multitemporal analysis would contribute towards a more meaningful analysis.
Ship Detection in the Suez Canal, Egypt - pre, during & post the Ever Given Mega Container Ship Stranding
The grounding of Ever Given - Taiwanese mega container ship at the southern end of the Suez Canal in Egypt and the resulting halt in maritime activity in this narrow but vital shipping route which accounts for 12% of the world's trade, has sent economic shockwaves across the world, including in India. This unfortunate incident is an acute reminder of the interdependency involved in the world we live in today and how it can trigger supply chain discomfort worldwide.
Drawing inspiration from a wonderful initiative from Google Earth - Timelapse, here is my short Timelapse version of maritime activity in the Suez Canal over the last month, as detected using Sentinel-1 SAR imagery.
Video Case: Detecting Ships in the Suez Canal - Pre, During & Post the Ever Given Cargo Vessel stranding
What are some of the other uses of Object detection on Water using Satellite Imagery?
Satellite Imagery, when combined with other observation tools, are used by Security management professionals to compare the reported records of vessel whereabouts (mandatory for ships above a certain size) with the number actual vessels detected at sea using satellite imagery. This is done to spot and track 'dark ships' i.e. those that do not report in their location and direction information - these may be participating in smuggling / illegal activities or even pose a threat to national security. Icebergs and Oil Spills can also be detected and monitored using Object detection techniques on water.
Refer my Imagery Analytics work here - you'll find a few new use cases involving Object Detection as well
Intelloc Mapping Services | Mapmyops is engaged in providing mapping solutions to organizations which facilitate operations improvement, planning & monitoring workflows. These include but are not limited to Supply Chain Consulting, Drone Services, Location Analytics & Applications, Site Characterization, Satellite Imagery Analytics & Polluted Water Remediation. Projects can be conducted pan-India and overseas.
Several other mapping and operations workflows are documented on this Geo-blog. Reach out to us via email - firstname.lastname@example.org or book a paid consultation (video meeting) from the link placed at the footer of this site's landing page.