Making deep learning easy for quality assurance applications

27 January 2020

Yonatan Hyatt explains how manufacturers can benefit from deep learning in their machine vision applications.

As manufacturers look towards more intelligent machine vision systems, deep learning is becoming a more common technique. Indeed, a report by ABI Research predicted that deep learning-based machine vision techniques within smart manufacturing will experience a compound annual growth rate of 20% between 2017 and 2023.

One barrier to its adoption in quality assurance applications is that for many manufacturers it is cost prohibitive or too complex an engineering task. The traditional vendor mechanism for deep learning machine vision means that software is sold as a package separately from the other components, all of which must be put together as a hard-engineered solution. This solution will be applicable for inspecting a single product at a single location on one single line. Even the most advanced solution equipped with deep learning will not be truly flexible. 

Because the task is so complex, the manufacturer will arrange for a systems integrator to select the lighting, cameras, communication, housing and more. It will be the systems integrator that selects the deep learning software for use in the machine vision solution as often the manufacturer will not have the expertise in-house to set up, train and operate a traditional deep learning solution independently. 

The lack of flexibility in traditional machine vision solutions means that if a change on the line occurs, the systems integrator must be called in again to either adjust the solution, for example by developing new lighting conditions, or to replace the solution with something new.

Complex training
Once built, a machine vision solution equipped with deep learning requires a training process. The user must present hundreds to thousands, and sometimes even millions, of defective samples to the solution, to teach it what a defective product looks like. The integrator will have to set machine learning parameters, such as data-augmentation, network topologies and final classification thresholds. All of this can take months.

To change this scenario, an autonomous machine vision solution has been developed by Inspekto. The S70 gives manufacturers full control of their visual quality assurance. They can set up a fully operational quality assurance system out of the box in under an hour. Set up requires around 20 to 30 good samples, and no defective ones. 

The algorithm developed by Inspekto requires no prior knowledge of the object nor any expertise from the operator, the system can distinguish any nuisance changes in the field of view from any material changes which constitute a defect. Changes in the object’s orientation or in the lighting conditions will therefore not be flagged up as defects. The system can inspect any product, at any location on the line and under any environmental conditions

An autonomous machine vision solution will also be more flexible –  it should be easy for the system to be moved from one point on the line to another, offering the same simple set up at the new location. This flexibility also means that visual quality assurance can be performed on multiple products at the same location on the production line, with the system detecting and classifying each product as appropriate.

Yonatan Hyatt is CTO at Inspekto.


Contact Details and Archive...

Related Articles...

Print this page | E-mail this page