View original application note on Cyth System’s website here.
Machine Vision with Deep Learning and Artificial Intelligence are among the future trends for automated testing in many different industries, especially manufacturing. But what if you need a vision system to recognize or verify organic objects? What if you need a system that doesn’t just recognize parts, but is able to make a subjective decision like a human operator?
A customer specializing in the handling and testing of high throughput semiconductor chips was looking to partner with an American vision specialist to help bring AI and Deep Learning technologies into a next generation system for the semiconductor inspection marketplace. They were seeking out a disruptive technology with intuitive inspection, so they partnered with systems integration company Cyth Systems to find a solution.
One major inspection challenge was the size of the semiconductor chips; they were so small that the cameras needed to resolve to the single micron level, so the selection of optical components posed a significant hurdle. It would also be a challenge to find a camera-lens combination that had the ability to capture micron-level imagery, while mechanically fitting into the necessary footprint, as well as being capable of capturing images in the needed time frame. The customer also wanted the ability to selectively apply different solutions or criteria to their unique inspection needs at will.
The Inspection Solution
They worked with Cyth Systems to develop a solution that inspects the client’s product by utilizing High Intensity Line Lights from Advanced illumination along with high-resolution line scan cameras. The lights and cameras combined to build up a product image, singulate unique components, and then run those images through the Cyth’s Neural Vision software. The software then determined good or bad parts, classifying defects for over 50 inspections.
Cyth Systems integrated a vision subassembly into the loading mechanism along with a Festo motor with linear stage, provided by the client. To create the system, Cyth’s team used LabVIEW software with two 12k resolution Basler line scan cameras, lenses from Edmund Optics, and specialized custom optical spacers for precision light control. Two Advanced illumination High Intensity Red Line Lights were used, which provided the ideal image quality when inspecting a reflective part. Every product image required 1GB of data, so the team implemented a powerful PC to handle the needed processing power.
To utilize the final solution, the client will load a cassette of multiple lead frames, which will be indexed through the system for the inspection of individual parts. Each frame contains over 100 unique parts with over 50 inspections each and a target inspection time of 25 seconds. The output is a visual report detailing which components have been identified as rejects, with a detailed breakdown of defects based on client criteria, resulting in fast and accurate processing of frames.
The fully integrated solution gives the customer greater control over their inspections, accommodating existing spatial requirements while increasing speed and decreasing human error in the customer’s high-resolution inspection system.