AI-Based Visual
Quality Control
With Maddox AI, you can reliably automate and digitise your quality control by simply annotating a few defect images - without any investment risk!
About our solution |
 |
Leading industrial companies rely on Maddox AI:
Benefits
Maddox AI: Your Benefits at a Glance
Maddox AI offers you numerous benefits over traditional quality control systems. Our cost-efficient solution can inspect even the most complex components, reducing recalibration efforts and pseudo-rejects. You can train, evaluate and optimise AI models for your quality control independently, without the need for deep AI knowledge. By processing the relevant production KPIs, our cloud software also provides valuable insights into production quality. And best of all, we take on the complete investment risk for each new inspection system!
0 x
recalibration
> 0 %
less pseudo scrap
0 %
risk free
0
AI-Expertise necessary
0 %
digitized analyses
Affiliated with
An official
Startup
Free White Paper: Why Aren't More Companies Using AI-Based Visual Inspection Systems?
Find out in our white paper why only a few companies are using AI systems in regular operation, although the majority is convinced of the great added value of the systems. Read now!
software-solution
Our Cloud Software at a Glance
Our cloud application allows you to create customised AI models without the need for AI expertise. The intuitive cloud software simplifies the model development process and ensures that you automate and digitise your inspection tasks quickly and efficiently.
Easy creation and maintenance of digital fault catalogues
Intuitive annotation of errors to be detected automatically
Transparent feedback on whether data quality is sufficient to train AI models
Train AI models with a few clicks
Reliable updating of models in your production lines
Easy creation and maintenance of digital fault catalogues
Intuitive annotation of errors to be detected automatically
Transparent feedback on whether data quality is sufficient to train AI models
Train AI models with a few clicks
Reliable updating of models in your production lines
Hardware-solution
Powerful Hardware Solutions for Individual Inspection Tasks
With the help of our flexible hardware components, a wide variety of applications can be reliably automated. Whether it is the inspection of components in the automotive industry, the quality control of injection-moulded parts in the plastics industry or the inspection of smallest elements in electronics production - we adapt our hardware precisely to your application to ensure precise and efficient quality control. If required, we can of course also reuse hardware that you have already installed. If you already have high-performance camera systems that basically produce sharp images but struggle with too high pseudo scrap, we can also optimise your existing camera systems with our solution and reduce your pseudo scrap.
Maddox PC Modul
The Maddox AI PC module with powerful NVIDIA GPU and 12" touchscreen monitor is an industrial AI system that can be seamlessly integrated into existing plants or with existing cameras thanks to standardized interfaces such as OPC-UA, REST API and DIO.
Maddox Eye
Maddox Eye is an image acquisition module with configurable camera, lens and lighting. Thanks to the robust water- and particle-proof housing and numerous standard accessories, our Maddox Eye can be flexibly integrated even in demanding environments.
Maddox Wire
Maddox Wire offers a compact solution for the inspection of endless products such as wires, hoses, tubes or rods. Using mirrors or multiple cameras, we enable reliable 360° inspection of products with diameters from 1 to 100 mm. Speeds of up to 15 meters/second can be handled without any problems.
Maddox Mini
Maddox Mini is the ideal solution for smaller production volumes while offering many customization options. With a wide range of accessories such as rotary modules, line cameras and integrated mandrel measurement, Maddox Mini enables comprehensive E2E inspection. If required, the recording logic can be transferred to a fully automated system at a later stage.
Maddox Maxi Uno
Maddox Maxi Uno is a compact stand-alone solution that enables precise inspection of a wide variety of products up to 700 mm x 400 mm. Sliding doors or drawers, combined with automatic image acquisition, allow ergonomic loading and unloading, as well as efficient cycle times.
Maddox Maxi Duo
Maddox Maxi Duo is a versatile inspection cell that offers an inspection capacity of up to 1000 mm x 700 mm. The system is flexibly expandable and can be equipped with a robot and a rotary module. This allows more demanding as well as more complex inspection tasks to be successfully and precisely handled.
Maddox Maxi Trio
Maddox Maxi Trio is our fully automatic stand-alone solution. Thanks to its modular design, it can be easily expanded to include automatic feeding and packaging and flexibly configured with a rotary indexing table, glass plate or conveyor belt system. Defective parts can thus be reliably separated from OK parts with short cycle times.
case studies
Many Inspection Tasks Can Be Automated With Maddox AI
Explore our case studies to find out how Maddox AI optimises quality control in industries such as automotive, plastics, electronics manufacturing and many more. Read about the individual challenges our customers have faced and how they benefit from our customised solutions. Use our filter options to search specifically by industry, application area or material properties and find relevant case studies.
Industry
Pharmaceutical and Medical
Pharmaceutical and Medical
- All
- Automotive
- Consumer Goods
- Electronical
- Energy Technology
- Food and Beverages
- Logistics
- Metal
- Pharmaceutical and Medical
- Plastic
Application
Component Detection
Component Detection
- All
- Code Recognition
- Component Detection
- Positioning Errors
- Surface Inspection
Material texture
Cardboard
Cardboard
- All
- Cardboard
- Copper
- Dough
- Metal
- Paper
- Plastic
- Rubber
- Steel
Fully Automated Wire Inspection
Surface defects such as dents, burnt areas or insulation defects can be reliably checked automatically with the help of Maddox AI.
Learn more
Inspection of Ball Bearings
Maddox AI successfully automates the inspection of impact points, incorrectly mounted components and scratches on ball bearings.
Learn more
Inspection of Printed Circuit Boards
Thanks to Maddox AI's device and software, incorrectly assembled components and other damages can be detected and tracked to a specific PCB.
Learn more
Cosmetic Container Inspection
Maddox AI's software, as well as the implementation of three individual camera systems, can reliably inspect surfaces, labels and imprints.
Learn more
Industry
Industry Overview: Diverse Applications in Quality Control
Maddox AI is already performing visual inspection tasks in a variety of industries. Our software is versatile and enables visual inspection of products in a wide range of application areas. Our team of experts will be happy to explain whether our solution can also automate your visual inspections.
Data Security
Your data is safe with us!
Our information management system is certified to meet the latest ISO 27001:2022 standards, ensuring the highest level of data security. Rest assured that your valuable data is well protected in our care. With continuous monitoring and improvement, we prioritize your information security at every step. Additionally, all data is processed and hosted exclusively in Germany.
FAQ
You Have Questions? We Have Answers:
Here you can find a collection of the most frequently asked questions on the topics of AI-based visual quality control as well as automation and digitalisation of inspection tasks. If you have any further questions, our team of experts will be happy to help!
What is the difference between classic and ML-based visual quality control?
Classic systems are given rules to recognise defects. A ML model, on the other hand, learns from examples whether a part is OK or NOK. In this learning process, the model derives implicit rules that do not require explicit programming (e.g. "NOK, if scratch on top left"). Machine Learning allows Maddox AI to automate even very complex use cases, such as surface recognition.
As a user, can I use Maddox AI without a technical or artificial intelligence background?
Yes, Maddox AI's software is designed in such a way that everybody can train AI models and put them into use after an initial thirty-minute training session.
How much does Maddox AI cost?
This depends on the specific use case. Before we start with a Maddox AI project, we show you transparently how much the use of Maddox AI would cost. Based on this, you can calculate your business case.
Can Maddox AI reliably inspect even at very high production speeds (e.g. several parts per second)?
In the vast majority of cases, high production speed is not a problem. Our models run on our industrial PC at your site, which means that the AI model analyses images within milliseconds. In our experience, it is the part handling rather than the AI evaluation that becomes the possible automation bottleneck.
How long does the implementation process of the Maddox AI system take?
This depends very much on how quickly we get enough data (pictures of OK/NOK parts). Since Maddox AI is an AI system, data is very important as a basis. We therefore always ask our customers to collect bad parts in advance. We can usually record these parts directly on the installation day and ideally already have enough data to train the model. In this case, the implementation only takes a few days. If we need to collect more data, the system can also run for a few weeks before the implementation is completed.
How much data (=annotated images) does Maddox AI need?
The important thing is not necessarily the amount of data, but its diversity. This means that different appearances of error types are relevant for a good and representative sample of defecets. We do not need many OK images, as the diversity in these images is usually smaller than for NOK images. Accordingly, our focus is on images with defects. We can give you the following rough rule of thumb: In 90% of the cases, 50 images per task (e.g. 50 NOK images of a defect class) are sufficient. In 10% of the cases, another 50 images (i.e. 100 in total) are needed to reliably adapt the Maddox AI system to your use case.
Can Maddox AI integrate with my software systems (MES, PLC, Data-Lake, etc.)?
Yes, this is not a problem. Maddox AI makes the results of the AI available via several protocols that are common in the industry, such as OPC-UA, REST API, MQTT, Modbus TCP or simple digital I/O.
Does Maddox AI make sure that there are no people visable in the images it takes?
Yes, Maddox AI places great emphasis on this. The Maddox AI system provides the ability to recognise people in the images to blur or black them out before uploading them to the cloud. No one has access to the people's identities - not even we can recover these images.
Will I receive new software updates regularly over time?
Yes, Maddox AI is a classic SaaS (software as service) solution. Accordingly, you will receive software updates regularly and free of charge.
Do I have to pay separately for software updates, maintenance and other after-sales services?
No, the software updates, maintenance and other after-sales services are included in the licence fee.