Case Studies | SaaS B2B

Spotting flood risk from the sky

Flow Connectivity Issue Identification PoC

About

A SaaS B2B business working with Cloud Combinator on AWS. The client is anonymised at their request.

Challenge

The challenge had three focus areas, each about turning a manual, expert task into a dependable automated service.

Teaching a model to read terrain

The solution had to interpret digital elevation maps and satellite imagery and accurately pick out flow issues within river basins, work that traditionally demands significant manual effort from specialists.

Making results precise and usable

A simple yes or no is not enough. The client needed structured output it could act on: results delivered as JSON with bounding boxes that pinpoint exactly where each flow issue sits in an image.

Secure, cost-aware operation

The client needed to upload sensitive imagery securely and run the service without paying for idle compute, which called for careful handling of data, model uptime and processing limits.

Solution

The client uploads imagery securely to AWS using pre-signed URLs, ideally a satellite image and a digital elevation map of the same region, though a single image type can be used when both are not available. The imagery is analysed by Amazon Rekognition models fine-tuned on a labelled dataset created with Amazon SageMaker Ground Truth, which returns each detection as a JSON object with bounding boxes marking the location of the flow issue.

2

Amazon Rekognition computer-vision models, for elevation maps and satellite imagery

5 min

Idle auto-shutdown of the model to avoid paying for unused compute

JSON

Structured output with bounding boxes pinpointing each flow issue

By the numbers:

  • 2 - Amazon Rekognition computer-vision models, for elevation maps and satellite imagery
  • 5 min - Idle auto-shutdown of the model to avoid paying for unused compute
  • JSON - Structured output with bounding boxes pinpointing each flow issue
Changes

The proof of concept was designed to give the client a streamlined, self-service system that analyses basin imagery with precision and returns actionable, geolocated results, replacing a slow manual process with an automated one hosted in their own AWS account.

  • Automated basin analysis — Computer vision identifies flow issues from DEM and satellite imagery, removing a task that traditionally required significant manual effort.
  • Precise, actionable output — Detections are returned as JSON with bounding boxes, so the client knows exactly where each issue is within an image.
  • Expertly labelled training data — Amazon SageMaker Ground Truth was used to build a labelled dataset to fine-tune the Rekognition models to the client's imagery.
  • Cost and reliability built in — On-demand model start-up, idle shutdown, request queuing and DynamoDB progress tracking keep the service both dependable and economical.
  • Handover — The engagement was designed to close with deployment and maintenance documentation, online knowledge-transfer sessions, and a documented code repository handed to the client.

By turning basin mapping into an automated AWS service, the work is designed to free the client's specialists from repetitive analysis and make expansion into new regions faster, supporting the company's ambition of accurate flood forecasting at global scale.

AWS Stack

Amazon Rekognition

For computer-vision detection of flow issues in DEM and satellite imagery.

Amazon SageMaker Ground Truth

For building the labelled dataset used to fine-tune the models.

Amazon S3

For secure image storage with intelligent tiering to control long-term cost.

Amazon DynamoDB

For tracking image progress through the pipeline and storing results.

Amazon VPC

For the secure network foundation, alongside data redundancy and lifecycle measures.

YOU MIGHT LIKE

Related success stories

View all case studies

Case Studies | Insights

Utilising Language Recognition, Speed, and Enhanced Security to Make Social Media a Force for Good

  • Here, we take a detailed look at how the Cloud Combinator team collaborated with another cutting-edge AI service provider that provides intelligent systems to “make social media more social” for brands and users alike.
  • Arwen AI is a UK-based startup specialising in AI solutions to manage and enhance brands’ social media interactions. Founded in 2020 by Matt McGrory, Dr. David Cole, and Joel Bailey, Arwen. AI focuses on using AI to automatically detect and remove spam, toxic comments, and other unwanted content from social media platforms.
  • The team at Arwen have three core products. ‘Moderate’ is focused on identifying and removing toxic content from social media channels. ‘Engage’ helps brands identify and engage with meaningful conversations on social media, and ‘Customize’ allows brands to apply bespoke algorithms to their channels - creating an even more effective moderation and engagement.
Read more
CONTACT US

Ready to turn AI into impact?

We'll help you spot the highest-value opportunities, reduce risk around your first AI initiative, and define a clear path to results from day one.

Why talk to us:

Outcome-driven recommendations

AWS-recognised delivery expertise

Risk-aware AI adoption

Clear next step, not a sales pitch

Start with a focused 20-minute conversation about your goals — no pressure, no commitment.

We use cookies

This website uses cookies to enhance user experience and to analyze performance and traffic on our website.

See our Privacy Policy for details.