Neural Vision is an Artificial Intelligence and Deep Learning software at the forefront of machine vision.
Traditional machine vision by nature is highly dependent on strictly controlled environments, rigid positional tolerances, and ultimately the skill of the vision programmer. It's up to the engineers to come up with every algorithm required to inspect a certain part - whether that be measurements, color, correct location or anything in between. But what if they miss something? What if there are parameters they don't know that matter?
This has been a common struggle with tradition machine vision that results in large amounts of time being spent on testing and troubleshooting a wide array of possibilities. Simply put, in the traditional vision space, the system is only as good as the engineers who programmed it.
Cyth Systems created Neural Vision to allow a person with no machine vision experience to inspect and classify products. Traditionally, machine vision systems worked by having an experienced programmer decide on a unique combination of analysis algorithms to inspect an image. Through Neural Vision an operator simply clicks items of interest and these same operations are applied in millions of combinations behind the scenes, looking for any recurring pattern or key measurements that identify a good or bad product.
In this way, instead of having a customer rely entirely on a vision company's skills and expertise, the customer now has control of the inspection criteria. This gives them the unique ability to tune or design the vision inspection for their product.
The most significant difference between Neural Vision and other machine vision platforms is that the customer now controls the "programming". The customer decides what product images get included into the models for learning, but they don't actually have to know any programming. "Programming" is now simply a point-and-click operation.
The old model of companies being golden handcuffed to machine vision integrators is no longer valid, for this relationship is entirely based on machine vision functioning almost like a "black box" to the customer. Cyth's goal is to use the latest generation of technology to put the power in the hands of the customer so they determine what will pass or fail. With this new model, an integrator is needed for the initial design of the system but the customer is then free to design inspections as needed. The company will always be an expert in their own product so by giving them the ability to design inspections, they're able to leverage their existing knowledge.
Cyth is selling the tools rather than the expertise. Think of any off-the-shelf word processing software, those programs work as a tool which allows you to fill in your specific information and customize it again and again. Similarly, Cyth gives clients the tools to customize their inspection needs.
The poster above displays the kind of path the customer will follow early in the development cycle. This process wasn't possible five years ago. Neural Vision is a "try before you buy" model that traditional machine vision isn't capable of because the amount of programming involved.
This is a new and exciting frontier for machine vision. Neural Vision is powering breakthroughs across multiple industries due to the lack of vision knowledge required and its ability to handle what are often considered subjective or organic inspections of complex parts. This new software allows customers to gain a confident understanding of their solution before having to pay for full system development. Ultimately, Neural Vision gives companies the tools to exceed as experts in their own product inspections.