Indurion Suite
Industrial vision intelligence infrastructure
Indurion Suite is Robovision’s vision AI solution that enables governed inspection across industries.
Built for production
Classification and measurement happen locally on the agent running on each machine, be it on-premises or air-gapped. Inspection keeps going even when the network drops. Corrections are validated centrally or locally, then reconciled across every connected machine, so a fix made once reaches your whole fleet.
Go from prototype to production with Indurion Suite. Your own team can train and adjust models, while every model you build stays yours.
Two ways to govern vision AI
Choose your perspective and see how Indurion Suite fits into your operation.
Embed Indurion Suite across your machine portfolio under your own brand, without vendor lock-in.
-
01
White-labelling
Your logo, interface text, and documentation appear on every machine you ship. Set the branding once and apply it across your installed base.
-
02
Sovereignity
Run your own vision AI models within a standard workflow. Your datasets and configurations remain portable, and the resulting IP stays yours.
-
03
Fast time-to-resolution
If a model on a shipped machine returns a low-confidence result, it flags and logs the sample so your engineers can review and correct it.
-
04
Operational autonomy
Your engineers can define a new part variant themselves, using a handful of images. No Robovision support is required.
-
05
Fleet-wide optimization
Any centrally validated corrections can be applied across every connected machine in your fleet.
-
06
Environment-agnostic
Indurion Suite fits into the hardware, infrastructure, and workflows you already run.
Indurion Suite replaces manual inspection with flexible, automated grading that adapts to your entire product lineup, dramatically reducing your risk of costly recalls.
-
01
Data ownership
Your data remains yours and is never used to train models for other customers. Your datasets and configurations remain portable, even if you leave.
-
02
Fast time-to-resolution
Anything that falls outside your defined quality rules is captured and flagged for operator review.
-
03
Fleet-wide standards
Set one standardized quality benchmark and push validated recipe updates across production lines locally, regionally, or globally.
-
04
Operational Autonomy
Your operators can add new defect classes or product variants themselves through an intuitive, award-winning AI lifecycle, on their own schedule.
-
05
Offline resilience
Production lines continue running locally if connectivity is lost.
Industrial vision AI made simple
Award-winning design meets industrial simplicity. Built for data scientists and intuitive for line operators, insightful for quality managers and support teams.
1 . Data import
Capture visual data from your image sensors and upload it to Indurion Suite.
2. Data annotation
Label and annotate images in 2D or 3D using the latest tools, from bounding boxes and interactive segmentation to Self-Supervised Vision Transformers (SSL ViT).
3. Data curation
Resolve differences between annotators with clear analytics. Establish a single ground truth and lock it into an immutable training dataset. Every deployed recipe remains traceable to the exact data behind it.
4. Model training
Create a model from scratch or start with a pretrained model. Tune it against your process thresholds and compare runs side by side to build a production-ready model from the data you already have.
5. Model testing
Compare models against the same data or your own operators, then review results class by class. Record which model was validated against which dataset, making model promotion a documented decision.
6. Model optimization
Set the level of uncertainty you're willing to accept, and Indurion Suite derives a confidence threshold for each class. Low-confidence samples are flagged for review and feed into your next labeling round.
7. Model deployment
Deploy models into production while each agent sends back outliers and low-confidence samples. These insights feed the next data import, creating a continuous cycle of improvement.
The vision techniques
Most vision AI applications do not start from a blank slate. They run on pretrained core model types rather than custom-built architectures. Indurion Suite is built around that reality.
The real differentiator lies in what your team can do with these models: label images, train models, and deploy them directly into production, independently. Indurion Suite supports five core vision techniques:
Classification
Classifies what’s seen into defined categories. As a foundational building block of computer vision, it answers a simple question about an image or region of interest (ROI): “What am I looking at?”
Instance segmentation
Identifies individual objects and traces their exact outlines at pixel level. This technique makes it possible to distinguish between touching and overlapping, or densely packed items on a high-speed line.
Semantic segmentation
Maps every pixel in an image to a defined category, creating continuous regions based on surfaces and materials, as well as other characteristics. It makes it possible to analyze non-discrete, or variable-shaped items on a production line and answers the question: “Which surface, material, or region does each pixel belong to?”
Object detection
While basic image classification evaluates an entire image to give it a single label, object detection scans the image to identify and classify the coordinates of multiple items using bounding boxes. It answers two questions in real time: "What objects are in frame, and where are they located?"
Bring your own algorithm
Already have a model or proprietary architecture, or a tuned vision routine that works? Bring it into Indurion Suite through the open algorithm specification. It can run in its own framework, within the same governed workflow as your other recipes.
With a typical ROI within 12 months, these improvements can deliver transformational impact at industrial scale.
Support beyond deployment
Indurion Suite installations are backed by expert services throughout deployment and ongoing operation.
FAQs
Find answers to common questions about governing vision AI on the factory floor. Still need help? Please contact our team.
Indurion Suite classifies and measures visual defects captured by imaging devices while identifying patterns in the data. It does not diagnose why a defect occurred. A defect type it has never seen must be labelled with a handful of examples before it can be recognized.
Our solution works with industry-standard machine vision cameras in single or multi-camera setups. It integrates with existing industrial protocols and control systems for both inputs and outputs. Exact compatibility depends on your hardware and setup, which we confirm during the site assessment before you commit to a timeline.
Indurion Suite exposes its results through a standard integration point designed to connect with the systems already running your line, including your MES. In a typical setup, this does not require a custom build. Whether your specific MES needs additional configuration depends on your setup. We confirm this during the site assessment before you commit to a timeline.
Your process experts can manage the Indurion Suite directly from the interface. Tasks like labeling, training, and threshold adjustments don't require dedicated data scientists, freeing them to focus on high-impact work. The infrastructure runs on-premises or as a Robovision-managed service, supported by our ongoing training. We define the ideal setup for you during solution design.
Yes. Indurion Suite deploys on your own infrastructure in a single-tenant environment and is built to keep inspecting without a live connection to a central system. Some administrative functions, including licensing, may require a periodic check-in depending on your environment configuration. We confirm the requirements for a fully air-gapped setup during solution design.
Find out where Indurion Suite can create the most value
Tell us about your process, quality challenges, and targets. We’ll assess where vision AI can make the biggest impact and what implementation could look like for your operation.