Visual Quality Inspection


The adoption of AI in multiple aspects of business operations has expanded quickly over a broad range of industries. In the domain of manufacturing, AI plays a significant role especially in the area of Visual Quality Inspection. Speed, accuracy and unambiguous judgement is what AI brings to this aspect of the business – this is a technology that manufacturers are closely watching & bringing into their operations.

A typical AI Quality Inspection setup involves samples produced from the machines, specialized lighting, industrial cameras and a workstation (typically equipped with a Graphics Processing Unit) where the AI engine is running. The AI Quality Inspection system will be able to detect Good or Defective samples while also categorising the types of defects detected, typically within a second or much lesser.


Being a leader in IoT imaging and monitoring technologies, we know imaging better than the competition This is demonstrated by our range of deep learning algorithms that can be deployed under different scenarios in order to get the best results.

Our intuitive user-interface enables users to quickly import images, label them & train the models from a single interface.

The system predicts the label for you and allows you to accept or re-label it, greatly simplifying the labelling process.

Once you are satisfied with the results of the model, simply select the destination workstation and deploy it instantaneously.

The results of the inspection can be configured to theformat that you need. Whether it is a simple XML/CSV file to be fed back to your AMES system, or an image indicating the location of the defect, the system provides the flexibility to meet your requirements.

Updates or patches are automatically installed if your workstation is accessible over the internet, saving the resources required for manual installation/patching.


Some of the applicable (but not limited to) scenarios.


Manual Inspection is usually based on human judgement & thus inconsistent due to various reasons such as vision, experience, fatigue etc

According to research, visual inspection errors typically range from 20% to 30% (Drury & Fox, 1975).

Misclassifying or overlooked defects (False Positive or Misses) can lead to a loss in quality, while false positives can cause unnecessary delays, increased production costs and overall wastage.

If an item is extremely small, zooming in or out to categorise the defect type will affect the speed of inspection.

The expense of hiring & re-training inspectors due to attrition is eliminated.

Use Cases

Breaking barriers of manufacturing in recent years, the digital revolution has turned the world into a playing field where only the most digitally advanced businesses survive.

Industry 4.0 technologies are reshaping manufacturing. Learn how the digitalisation of your manufacturing processes can optimise your business operations.