85%+ classification accuracy in live test scenarios
HANWHA - SECURITY SOLUTIONS
Pioneering AI innovation
Pioneering AI innovation
in video surveillance
We partnered with Hanwha Vision to develop an AI-powered object recognition engine that enhances video surveillance capabilities.
The result: more accurate detections, reduced manual workload, and scalable, intelligent security solutions.
85%+
OBJECT CLASSIFICATION ACCURACY
100%
CUSTOMIZABLE
DETECTION MODELS
DETECTION MODELS
INDUSTRY AWARD
WESCO INNOVATOR
OF THE YEAR
OF THE YEAR
About Hanwha Vision
Hanwha Vision is a global leader in next-generation video surveillance. As part of the Hanwha conglomerate, the company delivers high-performance cameras, processing infrastructure, and integrated security systems for a wide range of industries.
Their commitment to innovation and operational excellence makes them a key player in the global security landscape—and a strategic partner in pushing the boundaries of AI-driven surveillance.
The starting point
Hanwha’s Innovation Center set out to build—not fix—a new kind of solution: AI software that could match the sophistication of their hardware and open doors to scalable innovation. The goal was to move beyond reactive security and empower systems to recognize objects and patterns with precision and speed.
To succeed, Hanwha needed a human-centered, iterative approach to AI—one that would ensure accuracy, adaptability, and real-world value across diverse use cases.
THE GOAL
Laying the foundation for intelligent surveillance
Our goal was to create a flexible, AI-powered platform for object recognition—one that could evolve with each deployment and unlock new security capabilities across environments.
What we did
We collaborated with Hanwha’s Innovation Center to develop a robust AI engine through a human-in-the-loop process. Trained on real-world video data and refined through iterative feedback cycles, the system was built to continuously improve accuracy while handling real-world complexity.
Throughout the process, we focused on transparency, performance at scale, and empowering users with technology they can trust. Designed to adapt across environments—from retail to warehouse logistics and restricted areas—it integrates seamlessly with Hanwha’s broader security ecosystem.
Key initiatives included:
- Hypothesis-based modeling to shape object recognition logic
- Human-in-the-loop feedback to address edge cases and improve learning
- Statistical refinement to reduce false positives and increase accuracy
- Real-world data validation in complex operational environments
- Use case development for retail, logistics, and perimeter control
- Event showcasing at ISC West and GSX to gather stakeholder feedback
A STRONG SOLUTION
The tech stack
The project utilized a modern cloud-native architecture, combining Terraform for infrastructure as code and Kubernetes for container orchestration, ensuring scalable and reliable deployments. On the application side, React powered the frontend interfaces, while Node.js and Nest.js handled backend services with efficiency and modularity. Helm streamlined deployment workflows, enabling consistent and repeatable releases across environments.
By leveraging AWS as the cloud foundation, the team ensured high availability, security, and performance. This tech stack allowed Hanwha to accelerate development cycles, maintain operational resilience, and quickly adapt to evolving business requirements.
Results
AI that sees, understands, and adapts.
The solution we co-created with Hanwha set a new benchmark for intelligent video surveillance—one where automation and insight go hand in hand.
Smarter alerting and improved focus on critical incidents
Bandwidth and storage savings through intelligent filtering
Customizable detection models tailored to new environments
A scalable foundation for next-gen video analytics and global deployments
Award-Winning AI Innovation
Hanwha Vision’s FLEX AI was named Innovator of the Year by Wesco International for its breakthrough object recognition capabilities. With just 20 training images, FLEX AI enables cameras to detect custom objects—like shopping carts or forklifts—reducing losses and improving safety in real-world environments.
- We are product builders.
- We are result accountable.
- We are strategic partners.
Making Sense has provided me with a software-building team that looks to deliver beyond my expectations, offering scalable services and solutions to address the issues of my customers. I am only successful as a Product Manager for my current product due to their efforts.
Frederick Leung - Product Manager
at Hanwha Techwin
at Hanwha Techwin
Looking forward to making
sense with you
Tell us about your project challenges.