Sample Case Study

Waste Bin Detection


Partner: A provider of waste management systems, including trash and recycling bins, to optimize waste reduction and removal.

One service provided is confirmation that waste was removed on schedule.

Previous approaches to this, if they existed, relied upon manual service confirmation from drivers. A manual approach, while accurate, contributes significant overhead to the driver’s existing workload.

Automation methods such as placing sensors within bins are unreliable and costly.


A computer vision-enabled, edge computing system with attached camera.

System is designed to ingest camera feeds, process images to identify that the correct bin was collected, and confirm service.

System is modularized to allow upgrades of each component independently.

Compute resources are optimized to allow desired functionality on the device without overheating the system.

dark circle

Increased Fidelity

Increased fidelity of properly-identified service confirmation.

dark circle

Improved Focus

Truck drivers are able to focus on more critical activities.

Service Highlights

Extract insights from image-based systems (e.g. images, videos, camera feeds, etc.) to produce faster and more accurate analysis. For instance, detecting specific objects and tracking those objects through time.

Leverage the power of deep learning algorithms on the edge or in the cloud to add value to existing data. Utilize a complete pipeline of data engineering, machine learning models, real-time insights, and advanced analytics. Integrate highly customizable models to fit each individual need and use case. With years of experience in computer vision, machine learning, deep learning, and cloud platforms, we’ve developed a suite of tightly integrated and optimized solutions to help businesses better leverage their vision-based data.